NEWS
author bulwahn
Tue May 29 13:46:50 2012 +0200 (2012-05-29)
changeset 48013 44de84112a67
parent 47979 59ec72d3d0b9
child 48069 e9b2782c4f99
permissions -rw-r--r--
added optimisation for equational premises in Quickcheck; added some Quickcheck examples; NEWS
     1 Isabelle NEWS -- history user-relevant changes
     2 ==============================================
     3 
     4 New in this Isabelle version
     5 ----------------------------
     6 
     7 *** General ***
     8 
     9 * Discontinued obsolete method fastsimp / tactic fast_simp_tac, which
    10 is called fastforce / fast_force_tac already since Isabelle2011-1.
    11 
    12 
    13 *** System ***
    14 
    15 * Discontinued support for Poly/ML 5.2.1, which was the last version
    16 without exception positions and advanced ML compiler/toplevel
    17 configuration.
    18 
    19 
    20 *** HOL ***
    21 
    22 * Quickcheck:
    23 
    24   - added an optimisation for equality premises.
    25     It is switched on by default, and can be switched off by setting
    26     the configuration quickcheck_optimise_equality to false.    
    27 
    28 New in Isabelle2012 (May 2012)
    29 ------------------------------
    30 
    31 *** General ***
    32 
    33 * Prover IDE (PIDE) improvements:
    34 
    35   - more robust Sledgehammer integration (as before the sledgehammer
    36     command-line needs to be typed into the source buffer)
    37   - markup for bound variables
    38   - markup for types of term variables (displayed as tooltips)
    39   - support for user-defined Isar commands within the running session
    40   - improved support for Unicode outside original 16bit range
    41     e.g. glyph for \<A> (thanks to jEdit 4.5.1)
    42 
    43 * Forward declaration of outer syntax keywords within the theory
    44 header -- minor INCOMPATIBILITY for user-defined commands.  Allow new
    45 commands to be used in the same theory where defined.
    46 
    47 * Auxiliary contexts indicate block structure for specifications with
    48 additional parameters and assumptions.  Such unnamed contexts may be
    49 nested within other targets, like 'theory', 'locale', 'class',
    50 'instantiation' etc.  Results from the local context are generalized
    51 accordingly and applied to the enclosing target context.  Example:
    52 
    53   context
    54     fixes x y z :: 'a
    55     assumes xy: "x = y" and yz: "y = z"
    56   begin
    57 
    58   lemma my_trans: "x = z" using xy yz by simp
    59 
    60   end
    61 
    62   thm my_trans
    63 
    64 The most basic application is to factor-out context elements of
    65 several fixes/assumes/shows theorem statements, e.g. see
    66 ~~/src/HOL/Isar_Examples/Group_Context.thy
    67 
    68 Any other local theory specification element works within the "context
    69 ... begin ... end" block as well.
    70 
    71 * Bundled declarations associate attributed fact expressions with a
    72 given name in the context.  These may be later included in other
    73 contexts.  This allows to manage context extensions casually, without
    74 the logical dependencies of locales and locale interpretation.  See
    75 commands 'bundle', 'include', 'including' etc. in the isar-ref manual.
    76 
    77 * Commands 'lemmas' and 'theorems' allow local variables using 'for'
    78 declaration, and results are standardized before being stored.  Thus
    79 old-style "standard" after instantiation or composition of facts
    80 becomes obsolete.  Minor INCOMPATIBILITY, due to potential change of
    81 indices of schematic variables.
    82 
    83 * Rule attributes in local theory declarations (e.g. locale or class)
    84 are now statically evaluated: the resulting theorem is stored instead
    85 of the original expression.  INCOMPATIBILITY in rare situations, where
    86 the historic accident of dynamic re-evaluation in interpretations
    87 etc. was exploited.
    88 
    89 * New tutorial "Programming and Proving in Isabelle/HOL"
    90 ("prog-prove").  It completely supersedes "A Tutorial Introduction to
    91 Structured Isar Proofs" ("isar-overview"), which has been removed.  It
    92 also supersedes "Isabelle/HOL, A Proof Assistant for Higher-Order
    93 Logic" as the recommended beginners tutorial, but does not cover all
    94 of the material of that old tutorial.
    95 
    96 * Updated and extended reference manuals: "isar-ref",
    97 "implementation", "system"; reduced remaining material in old "ref"
    98 manual.
    99 
   100 
   101 *** Pure ***
   102 
   103 * Command 'definition' no longer exports the foundational "raw_def"
   104 into the user context.  Minor INCOMPATIBILITY, may use the regular
   105 "def" result with attribute "abs_def" to imitate the old version.
   106 
   107 * Attribute "abs_def" turns an equation of the form "f x y == t" into
   108 "f == %x y. t", which ensures that "simp" or "unfold" steps always
   109 expand it.  This also works for object-logic equality.  (Formerly
   110 undocumented feature.)
   111 
   112 * Sort constraints are now propagated in simultaneous statements, just
   113 like type constraints.  INCOMPATIBILITY in rare situations, where
   114 distinct sorts used to be assigned accidentally.  For example:
   115 
   116   lemma "P (x::'a::foo)" and "Q (y::'a::bar)"  -- "now illegal"
   117 
   118   lemma "P (x::'a)" and "Q (y::'a::bar)"
   119     -- "now uniform 'a::bar instead of default sort for first occurrence (!)"
   120 
   121 * Rule composition via attribute "OF" (or ML functions OF/MRS) is more
   122 tolerant against multiple unifiers, as long as the final result is
   123 unique.  (As before, rules are composed in canonical right-to-left
   124 order to accommodate newly introduced premises.)
   125 
   126 * Renamed some inner syntax categories:
   127 
   128     num ~> num_token
   129     xnum ~> xnum_token
   130     xstr ~> str_token
   131 
   132 Minor INCOMPATIBILITY.  Note that in practice "num_const" or
   133 "num_position" etc. are mainly used instead (which also include
   134 position information via constraints).
   135 
   136 * Simplified configuration options for syntax ambiguity: see
   137 "syntax_ambiguity_warning" and "syntax_ambiguity_limit" in isar-ref
   138 manual.  Minor INCOMPATIBILITY.
   139 
   140 * Discontinued configuration option "syntax_positions": atomic terms
   141 in parse trees are always annotated by position constraints.
   142 
   143 * Old code generator for SML and its commands 'code_module',
   144 'code_library', 'consts_code', 'types_code' have been discontinued.
   145 Use commands of the generic code generator instead.  INCOMPATIBILITY.
   146 
   147 * Redundant attribute "code_inline" has been discontinued. Use
   148 "code_unfold" instead.  INCOMPATIBILITY.
   149 
   150 * Dropped attribute "code_unfold_post" in favor of the its dual
   151 "code_abbrev", which yields a common pattern in definitions like
   152 
   153   definition [code_abbrev]: "f = t"
   154 
   155 INCOMPATIBILITY.
   156 
   157 * Obsolete 'types' command has been discontinued.  Use 'type_synonym'
   158 instead.  INCOMPATIBILITY.
   159 
   160 * Discontinued old "prems" fact, which used to refer to the accidental
   161 collection of foundational premises in the context (already marked as
   162 legacy since Isabelle2011).
   163 
   164 
   165 *** HOL ***
   166 
   167 * Type 'a set is now a proper type constructor (just as before
   168 Isabelle2008).  Definitions mem_def and Collect_def have disappeared.
   169 Non-trivial INCOMPATIBILITY.  For developments keeping predicates and
   170 sets separate, it is often sufficient to rephrase some set S that has
   171 been accidentally used as predicates by "%x. x : S", and some
   172 predicate P that has been accidentally used as set by "{x. P x}".
   173 Corresponding proofs in a first step should be pruned from any
   174 tinkering with former theorems mem_def and Collect_def as far as
   175 possible.
   176 
   177 For developments which deliberately mix predicates and sets, a
   178 planning step is necessary to determine what should become a predicate
   179 and what a set.  It can be helpful to carry out that step in
   180 Isabelle2011-1 before jumping right into the current release.
   181 
   182 * Code generation by default implements sets as container type rather
   183 than predicates.  INCOMPATIBILITY.
   184 
   185 * New type synonym 'a rel = ('a * 'a) set
   186 
   187 * The representation of numerals has changed.  Datatype "num"
   188 represents strictly positive binary numerals, along with functions
   189 "numeral :: num => 'a" and "neg_numeral :: num => 'a" to represent
   190 positive and negated numeric literals, respectively.  See also
   191 definitions in ~~/src/HOL/Num.thy.  Potential INCOMPATIBILITY, some
   192 user theories may require adaptations as follows:
   193 
   194   - Theorems with number_ring or number_semiring constraints: These
   195     classes are gone; use comm_ring_1 or comm_semiring_1 instead.
   196 
   197   - Theories defining numeric types: Remove number, number_semiring,
   198     and number_ring instances. Defer all theorems about numerals until
   199     after classes one and semigroup_add have been instantiated.
   200 
   201   - Numeral-only simp rules: Replace each rule having a "number_of v"
   202     pattern with two copies, one for numeral and one for neg_numeral.
   203 
   204   - Theorems about subclasses of semiring_1 or ring_1: These classes
   205     automatically support numerals now, so more simp rules and
   206     simprocs may now apply within the proof.
   207 
   208   - Definitions and theorems using old constructors Pls/Min/Bit0/Bit1:
   209     Redefine using other integer operations.
   210 
   211 * Transfer: New package intended to generalize the existing
   212 "descending" method and related theorem attributes from the Quotient
   213 package.  (Not all functionality is implemented yet, but future
   214 development will focus on Transfer as an eventual replacement for the
   215 corresponding parts of the Quotient package.)
   216 
   217   - transfer_rule attribute: Maintains a collection of transfer rules,
   218     which relate constants at two different types. Transfer rules may
   219     relate different type instances of the same polymorphic constant,
   220     or they may relate an operation on a raw type to a corresponding
   221     operation on an abstract type (quotient or subtype). For example:
   222 
   223     ((A ===> B) ===> list_all2 A ===> list_all2 B) map map
   224     (cr_int ===> cr_int ===> cr_int) (%(x,y) (u,v). (x+u, y+v)) plus_int
   225 
   226   - transfer method: Replaces a subgoal on abstract types with an
   227     equivalent subgoal on the corresponding raw types. Constants are
   228     replaced with corresponding ones according to the transfer rules.
   229     Goals are generalized over all free variables by default; this is
   230     necessary for variables whose types change, but can be overridden
   231     for specific variables with e.g. "transfer fixing: x y z".  The
   232     variant transfer' method allows replacing a subgoal with one that
   233     is logically stronger (rather than equivalent).
   234 
   235   - relator_eq attribute: Collects identity laws for relators of
   236     various type constructors, e.g. "list_all2 (op =) = (op =)".  The
   237     transfer method uses these lemmas to infer transfer rules for
   238     non-polymorphic constants on the fly.
   239 
   240   - transfer_prover method: Assists with proving a transfer rule for a
   241     new constant, provided the constant is defined in terms of other
   242     constants that already have transfer rules. It should be applied
   243     after unfolding the constant definitions.
   244 
   245   - HOL/ex/Transfer_Int_Nat.thy: Example theory demonstrating transfer
   246     from type nat to type int.
   247 
   248 * Lifting: New package intended to generalize the quotient_definition
   249 facility of the Quotient package; designed to work with Transfer.
   250 
   251   - lift_definition command: Defines operations on an abstract type in
   252     terms of a corresponding operation on a representation
   253     type.  Example syntax:
   254 
   255     lift_definition dlist_insert :: "'a => 'a dlist => 'a dlist"
   256       is List.insert
   257 
   258     Users must discharge a respectfulness proof obligation when each
   259     constant is defined. (For a type copy, i.e. a typedef with UNIV,
   260     the proof is discharged automatically.) The obligation is
   261     presented in a user-friendly, readable form; a respectfulness
   262     theorem in the standard format and a transfer rule are generated
   263     by the package.
   264 
   265   - Integration with code_abstype: For typedefs (e.g. subtypes
   266     corresponding to a datatype invariant, such as dlist),
   267     lift_definition generates a code certificate theorem and sets up
   268     code generation for each constant.
   269 
   270   - setup_lifting command: Sets up the Lifting package to work with a
   271     user-defined type. The user must provide either a quotient theorem
   272     or a type_definition theorem.  The package configures transfer
   273     rules for equality and quantifiers on the type, and sets up the
   274     lift_definition command to work with the type.
   275 
   276   - Usage examples: See Quotient_Examples/Lift_DList.thy,
   277     Quotient_Examples/Lift_RBT.thy, Quotient_Examples/Lift_FSet.thy,
   278     Word/Word.thy and Library/Float.thy.
   279 
   280 * Quotient package:
   281 
   282   - The 'quotient_type' command now supports a 'morphisms' option with
   283     rep and abs functions, similar to typedef.
   284 
   285   - 'quotient_type' sets up new types to work with the Lifting and
   286     Transfer packages, as with 'setup_lifting'.
   287 
   288   - The 'quotient_definition' command now requires the user to prove a
   289     respectfulness property at the point where the constant is
   290     defined, similar to lift_definition; INCOMPATIBILITY.
   291 
   292   - Renamed predicate 'Quotient' to 'Quotient3', and renamed theorems
   293     accordingly, INCOMPATIBILITY.
   294 
   295 * New diagnostic command 'find_unused_assms' to find potentially
   296 superfluous assumptions in theorems using Quickcheck.
   297 
   298 * Quickcheck:
   299 
   300   - Quickcheck returns variable assignments as counterexamples, which
   301     allows to reveal the underspecification of functions under test.
   302     For example, refuting "hd xs = x", it presents the variable
   303     assignment xs = [] and x = a1 as a counterexample, assuming that
   304     any property is false whenever "hd []" occurs in it.
   305 
   306     These counterexample are marked as potentially spurious, as
   307     Quickcheck also returns "xs = []" as a counterexample to the
   308     obvious theorem "hd xs = hd xs".
   309 
   310     After finding a potentially spurious counterexample, Quickcheck
   311     continues searching for genuine ones.
   312 
   313     By default, Quickcheck shows potentially spurious and genuine
   314     counterexamples. The option "genuine_only" sets quickcheck to only
   315     show genuine counterexamples.
   316 
   317   - The command 'quickcheck_generator' creates random and exhaustive
   318     value generators for a given type and operations.
   319 
   320     It generates values by using the operations as if they were
   321     constructors of that type.
   322 
   323   - Support for multisets.
   324 
   325   - Added "use_subtype" options.
   326 
   327   - Added "quickcheck_locale" configuration to specify how to process
   328     conjectures in a locale context.
   329 
   330 * Nitpick: Fixed infinite loop caused by the 'peephole_optim' option
   331 and affecting 'rat' and 'real'.
   332 
   333 * Sledgehammer:
   334   - Integrated more tightly with SPASS, as described in the ITP 2012
   335     paper "More SPASS with Isabelle".
   336   - Made it try "smt" as a fallback if "metis" fails or times out.
   337   - Added support for the following provers: Alt-Ergo (via Why3 and
   338     TFF1), iProver, iProver-Eq.
   339   - Replaced remote E-SInE with remote Satallax in the default setup.
   340   - Sped up the minimizer.
   341   - Added "lam_trans", "uncurry_aliases", and "minimize" options.
   342   - Renamed "slicing" ("no_slicing") option to "slice" ("dont_slice").
   343   - Renamed "sound" option to "strict".
   344 
   345 * Metis: Added possibility to specify lambda translations scheme as a
   346 parenthesized argument (e.g., "by (metis (lifting) ...)").
   347 
   348 * SMT: Renamed "smt_fixed" option to "smt_read_only_certificates".
   349 
   350 * Command 'try0': Renamed from 'try_methods'. INCOMPATIBILITY.
   351 
   352 * New "case_product" attribute to generate a case rule doing multiple
   353 case distinctions at the same time.  E.g.
   354 
   355   list.exhaust [case_product nat.exhaust]
   356 
   357 produces a rule which can be used to perform case distinction on both
   358 a list and a nat.
   359 
   360 * New "eventually_elim" method as a generalized variant of the
   361 eventually_elim* rules.  Supports structured proofs.
   362 
   363 * Typedef with implicit set definition is considered legacy.  Use
   364 "typedef (open)" form instead, which will eventually become the
   365 default.
   366 
   367 * Record: code generation can be switched off manually with
   368 
   369   declare [[record_coden = false]]  -- "default true"
   370 
   371 * Datatype: type parameters allow explicit sort constraints.
   372 
   373 * Concrete syntax for case expressions includes constraints for source
   374 positions, and thus produces Prover IDE markup for its bindings.
   375 INCOMPATIBILITY for old-style syntax translations that augment the
   376 pattern notation; e.g. see src/HOL/HOLCF/One.thy for translations of
   377 one_case.
   378 
   379 * Clarified attribute "mono_set": pure declaration without modifying
   380 the result of the fact expression.
   381 
   382 * More default pred/set conversions on a couple of relation operations
   383 and predicates.  Added powers of predicate relations.  Consolidation
   384 of some relation theorems:
   385 
   386   converse_def ~> converse_unfold
   387   rel_comp_def ~> relcomp_unfold
   388   symp_def ~> (modified, use symp_def and sym_def instead)
   389   transp_def ~> transp_trans
   390   Domain_def ~> Domain_unfold
   391   Range_def ~> Domain_converse [symmetric]
   392 
   393 Generalized theorems INF_INT_eq, INF_INT_eq2, SUP_UN_eq, SUP_UN_eq2.
   394 
   395 See theory "Relation" for examples for making use of pred/set
   396 conversions by means of attributes "to_set" and "to_pred".
   397 
   398 INCOMPATIBILITY.
   399 
   400 * Renamed facts about the power operation on relations, i.e., relpow
   401 to match the constant's name:
   402 
   403   rel_pow_1 ~> relpow_1
   404   rel_pow_0_I ~> relpow_0_I
   405   rel_pow_Suc_I ~> relpow_Suc_I
   406   rel_pow_Suc_I2 ~> relpow_Suc_I2
   407   rel_pow_0_E ~> relpow_0_E
   408   rel_pow_Suc_E ~> relpow_Suc_E
   409   rel_pow_E ~> relpow_E
   410   rel_pow_Suc_D2 ~> relpow_Suc_D2
   411   rel_pow_Suc_E2 ~> relpow_Suc_E2
   412   rel_pow_Suc_D2' ~> relpow_Suc_D2'
   413   rel_pow_E2 ~> relpow_E2
   414   rel_pow_add ~> relpow_add
   415   rel_pow_commute ~> relpow
   416   rel_pow_empty ~> relpow_empty:
   417   rtrancl_imp_UN_rel_pow ~> rtrancl_imp_UN_relpow
   418   rel_pow_imp_rtrancl ~> relpow_imp_rtrancl
   419   rtrancl_is_UN_rel_pow ~> rtrancl_is_UN_relpow
   420   rtrancl_imp_rel_pow ~> rtrancl_imp_relpow
   421   rel_pow_fun_conv ~> relpow_fun_conv
   422   rel_pow_finite_bounded1 ~> relpow_finite_bounded1
   423   rel_pow_finite_bounded ~> relpow_finite_bounded
   424   rtrancl_finite_eq_rel_pow ~> rtrancl_finite_eq_relpow
   425   trancl_finite_eq_rel_pow ~> trancl_finite_eq_relpow
   426   single_valued_rel_pow ~> single_valued_relpow
   427 
   428 INCOMPATIBILITY.
   429 
   430 * Theory Relation: Consolidated constant name for relation composition
   431 and corresponding theorem names:
   432 
   433   - Renamed constant rel_comp to relcomp.
   434 
   435   - Dropped abbreviation pred_comp. Use relcompp instead.
   436 
   437   - Renamed theorems:
   438 
   439     rel_compI ~> relcompI
   440     rel_compEpair ~> relcompEpair
   441     rel_compE ~> relcompE
   442     pred_comp_rel_comp_eq ~> relcompp_relcomp_eq
   443     rel_comp_empty1 ~> relcomp_empty1
   444     rel_comp_mono ~> relcomp_mono
   445     rel_comp_subset_Sigma ~> relcomp_subset_Sigma
   446     rel_comp_distrib ~> relcomp_distrib
   447     rel_comp_distrib2 ~> relcomp_distrib2
   448     rel_comp_UNION_distrib ~> relcomp_UNION_distrib
   449     rel_comp_UNION_distrib2 ~> relcomp_UNION_distrib2
   450     single_valued_rel_comp ~> single_valued_relcomp
   451     rel_comp_def ~> relcomp_unfold
   452     converse_rel_comp ~> converse_relcomp
   453     pred_compI ~> relcomppI
   454     pred_compE ~> relcomppE
   455     pred_comp_bot1 ~> relcompp_bot1
   456     pred_comp_bot2 ~> relcompp_bot2
   457     transp_pred_comp_less_eq ~> transp_relcompp_less_eq
   458     pred_comp_mono ~> relcompp_mono
   459     pred_comp_distrib ~> relcompp_distrib
   460     pred_comp_distrib2 ~> relcompp_distrib2
   461     converse_pred_comp ~> converse_relcompp
   462 
   463     finite_rel_comp ~> finite_relcomp
   464 
   465     set_rel_comp ~> set_relcomp
   466 
   467 INCOMPATIBILITY.
   468 
   469 * Theory Divides: Discontinued redundant theorems about div and mod.
   470 INCOMPATIBILITY, use the corresponding generic theorems instead.
   471 
   472   DIVISION_BY_ZERO ~> div_by_0, mod_by_0
   473   zdiv_self ~> div_self
   474   zmod_self ~> mod_self
   475   zdiv_zero ~> div_0
   476   zmod_zero ~> mod_0
   477   zdiv_zmod_equality ~> div_mod_equality2
   478   zdiv_zmod_equality2 ~> div_mod_equality
   479   zmod_zdiv_trivial ~> mod_div_trivial
   480   zdiv_zminus_zminus ~> div_minus_minus
   481   zmod_zminus_zminus ~> mod_minus_minus
   482   zdiv_zminus2 ~> div_minus_right
   483   zmod_zminus2 ~> mod_minus_right
   484   zdiv_minus1_right ~> div_minus1_right
   485   zmod_minus1_right ~> mod_minus1_right
   486   zdvd_mult_div_cancel ~> dvd_mult_div_cancel
   487   zmod_zmult1_eq ~> mod_mult_right_eq
   488   zpower_zmod ~> power_mod
   489   zdvd_zmod ~> dvd_mod
   490   zdvd_zmod_imp_zdvd ~> dvd_mod_imp_dvd
   491   mod_mult_distrib ~> mult_mod_left
   492   mod_mult_distrib2 ~> mult_mod_right
   493 
   494 * Removed redundant theorems nat_mult_2 and nat_mult_2_right; use
   495 generic mult_2 and mult_2_right instead. INCOMPATIBILITY.
   496 
   497 * Finite_Set.fold now qualified.  INCOMPATIBILITY.
   498 
   499 * Consolidated theorem names concerning fold combinators:
   500 
   501   inf_INFI_fold_inf ~> inf_INF_fold_inf
   502   sup_SUPR_fold_sup ~> sup_SUP_fold_sup
   503   INFI_fold_inf ~> INF_fold_inf
   504   SUPR_fold_sup ~> SUP_fold_sup
   505   union_set ~> union_set_fold
   506   minus_set ~> minus_set_fold
   507   INFI_set_fold ~> INF_set_fold
   508   SUPR_set_fold ~> SUP_set_fold
   509   INF_code ~> INF_set_foldr
   510   SUP_code ~> SUP_set_foldr
   511   foldr.simps ~> foldr.simps (in point-free formulation)
   512   foldr_fold_rev ~> foldr_conv_fold
   513   foldl_fold ~> foldl_conv_fold
   514   foldr_foldr ~> foldr_conv_foldl
   515   foldl_foldr ~> foldl_conv_foldr
   516   fold_set_remdups ~> fold_set_fold_remdups
   517   fold_set ~> fold_set_fold
   518   fold1_set ~> fold1_set_fold
   519 
   520 INCOMPATIBILITY.
   521 
   522 * Dropped rarely useful theorems concerning fold combinators:
   523 foldl_apply, foldl_fun_comm, foldl_rev, fold_weak_invariant,
   524 rev_foldl_cons, fold_set_remdups, fold_set, fold_set1,
   525 concat_conv_foldl, foldl_weak_invariant, foldl_invariant,
   526 foldr_invariant, foldl_absorb0, foldl_foldr1_lemma, foldl_foldr1,
   527 listsum_conv_fold, listsum_foldl, sort_foldl_insort, foldl_assoc,
   528 foldr_conv_foldl, start_le_sum, elem_le_sum, sum_eq_0_conv.
   529 INCOMPATIBILITY.  For the common phrases "%xs. List.foldr plus xs 0"
   530 and "List.foldl plus 0", prefer "List.listsum".  Otherwise it can be
   531 useful to boil down "List.foldr" and "List.foldl" to "List.fold" by
   532 unfolding "foldr_conv_fold" and "foldl_conv_fold".
   533 
   534 * Dropped lemmas minus_set_foldr, union_set_foldr, union_coset_foldr,
   535 inter_coset_foldr, Inf_fin_set_foldr, Sup_fin_set_foldr,
   536 Min_fin_set_foldr, Max_fin_set_foldr, Inf_set_foldr, Sup_set_foldr,
   537 INF_set_foldr, SUP_set_foldr.  INCOMPATIBILITY.  Prefer corresponding
   538 lemmas over fold rather than foldr, or make use of lemmas
   539 fold_conv_foldr and fold_rev.
   540 
   541 * Congruence rules Option.map_cong and Option.bind_cong for recursion
   542 through option types.
   543 
   544 * "Transitive_Closure.ntrancl": bounded transitive closure on
   545 relations.
   546 
   547 * Constant "Set.not_member" now qualified.  INCOMPATIBILITY.
   548 
   549 * Theory Int: Discontinued many legacy theorems specific to type int.
   550 INCOMPATIBILITY, use the corresponding generic theorems instead.
   551 
   552   zminus_zminus ~> minus_minus
   553   zminus_0 ~> minus_zero
   554   zminus_zadd_distrib ~> minus_add_distrib
   555   zadd_commute ~> add_commute
   556   zadd_assoc ~> add_assoc
   557   zadd_left_commute ~> add_left_commute
   558   zadd_ac ~> add_ac
   559   zmult_ac ~> mult_ac
   560   zadd_0 ~> add_0_left
   561   zadd_0_right ~> add_0_right
   562   zadd_zminus_inverse2 ~> left_minus
   563   zmult_zminus ~> mult_minus_left
   564   zmult_commute ~> mult_commute
   565   zmult_assoc ~> mult_assoc
   566   zadd_zmult_distrib ~> left_distrib
   567   zadd_zmult_distrib2 ~> right_distrib
   568   zdiff_zmult_distrib ~> left_diff_distrib
   569   zdiff_zmult_distrib2 ~> right_diff_distrib
   570   zmult_1 ~> mult_1_left
   571   zmult_1_right ~> mult_1_right
   572   zle_refl ~> order_refl
   573   zle_trans ~> order_trans
   574   zle_antisym ~> order_antisym
   575   zle_linear ~> linorder_linear
   576   zless_linear ~> linorder_less_linear
   577   zadd_left_mono ~> add_left_mono
   578   zadd_strict_right_mono ~> add_strict_right_mono
   579   zadd_zless_mono ~> add_less_le_mono
   580   int_0_less_1 ~> zero_less_one
   581   int_0_neq_1 ~> zero_neq_one
   582   zless_le ~> less_le
   583   zpower_zadd_distrib ~> power_add
   584   zero_less_zpower_abs_iff ~> zero_less_power_abs_iff
   585   zero_le_zpower_abs ~> zero_le_power_abs
   586 
   587 * Theory Deriv: Renamed
   588 
   589   DERIV_nonneg_imp_nonincreasing ~> DERIV_nonneg_imp_nondecreasing
   590 
   591 * Theory Library/Multiset: Improved code generation of multisets.
   592 
   593 * Theory HOL/Library/Set_Algebras: Addition and multiplication on sets
   594 are expressed via type classes again. The special syntax
   595 \<oplus>/\<otimes> has been replaced by plain +/*. Removed constant
   596 setsum_set, which is now subsumed by Big_Operators.setsum.
   597 INCOMPATIBILITY.
   598 
   599 * Theory HOL/Library/Diagonalize has been removed. INCOMPATIBILITY,
   600 use theory HOL/Library/Nat_Bijection instead.
   601 
   602 * Theory HOL/Library/RBT_Impl: Backing implementation of red-black
   603 trees is now inside a type class context.  Names of affected
   604 operations and lemmas have been prefixed by rbt_.  INCOMPATIBILITY for
   605 theories working directly with raw red-black trees, adapt the names as
   606 follows:
   607 
   608   Operations:
   609   bulkload -> rbt_bulkload
   610   del_from_left -> rbt_del_from_left
   611   del_from_right -> rbt_del_from_right
   612   del -> rbt_del
   613   delete -> rbt_delete
   614   ins -> rbt_ins
   615   insert -> rbt_insert
   616   insertw -> rbt_insert_with
   617   insert_with_key -> rbt_insert_with_key
   618   map_entry -> rbt_map_entry
   619   lookup -> rbt_lookup
   620   sorted -> rbt_sorted
   621   tree_greater -> rbt_greater
   622   tree_less -> rbt_less
   623   tree_less_symbol -> rbt_less_symbol
   624   union -> rbt_union
   625   union_with -> rbt_union_with
   626   union_with_key -> rbt_union_with_key
   627 
   628   Lemmas:
   629   balance_left_sorted -> balance_left_rbt_sorted
   630   balance_left_tree_greater -> balance_left_rbt_greater
   631   balance_left_tree_less -> balance_left_rbt_less
   632   balance_right_sorted -> balance_right_rbt_sorted
   633   balance_right_tree_greater -> balance_right_rbt_greater
   634   balance_right_tree_less -> balance_right_rbt_less
   635   balance_sorted -> balance_rbt_sorted
   636   balance_tree_greater -> balance_rbt_greater
   637   balance_tree_less -> balance_rbt_less
   638   bulkload_is_rbt -> rbt_bulkload_is_rbt
   639   combine_sorted -> combine_rbt_sorted
   640   combine_tree_greater -> combine_rbt_greater
   641   combine_tree_less -> combine_rbt_less
   642   delete_in_tree -> rbt_delete_in_tree
   643   delete_is_rbt -> rbt_delete_is_rbt
   644   del_from_left_tree_greater -> rbt_del_from_left_rbt_greater
   645   del_from_left_tree_less -> rbt_del_from_left_rbt_less
   646   del_from_right_tree_greater -> rbt_del_from_right_rbt_greater
   647   del_from_right_tree_less -> rbt_del_from_right_rbt_less
   648   del_in_tree -> rbt_del_in_tree
   649   del_inv1_inv2 -> rbt_del_inv1_inv2
   650   del_sorted -> rbt_del_rbt_sorted
   651   del_tree_greater -> rbt_del_rbt_greater
   652   del_tree_less -> rbt_del_rbt_less
   653   dom_lookup_Branch -> dom_rbt_lookup_Branch
   654   entries_lookup -> entries_rbt_lookup
   655   finite_dom_lookup -> finite_dom_rbt_lookup
   656   insert_sorted -> rbt_insert_rbt_sorted
   657   insertw_is_rbt -> rbt_insertw_is_rbt
   658   insertwk_is_rbt -> rbt_insertwk_is_rbt
   659   insertwk_sorted -> rbt_insertwk_rbt_sorted
   660   insertw_sorted -> rbt_insertw_rbt_sorted
   661   ins_sorted -> ins_rbt_sorted
   662   ins_tree_greater -> ins_rbt_greater
   663   ins_tree_less -> ins_rbt_less
   664   is_rbt_sorted -> is_rbt_rbt_sorted
   665   lookup_balance -> rbt_lookup_balance
   666   lookup_bulkload -> rbt_lookup_rbt_bulkload
   667   lookup_delete -> rbt_lookup_rbt_delete
   668   lookup_Empty -> rbt_lookup_Empty
   669   lookup_from_in_tree -> rbt_lookup_from_in_tree
   670   lookup_in_tree -> rbt_lookup_in_tree
   671   lookup_ins -> rbt_lookup_ins
   672   lookup_insert -> rbt_lookup_rbt_insert
   673   lookup_insertw -> rbt_lookup_rbt_insertw
   674   lookup_insertwk -> rbt_lookup_rbt_insertwk
   675   lookup_keys -> rbt_lookup_keys
   676   lookup_map -> rbt_lookup_map
   677   lookup_map_entry -> rbt_lookup_rbt_map_entry
   678   lookup_tree_greater -> rbt_lookup_rbt_greater
   679   lookup_tree_less -> rbt_lookup_rbt_less
   680   lookup_union -> rbt_lookup_rbt_union
   681   map_entry_color_of -> rbt_map_entry_color_of
   682   map_entry_inv1 -> rbt_map_entry_inv1
   683   map_entry_inv2 -> rbt_map_entry_inv2
   684   map_entry_is_rbt -> rbt_map_entry_is_rbt
   685   map_entry_sorted -> rbt_map_entry_rbt_sorted
   686   map_entry_tree_greater -> rbt_map_entry_rbt_greater
   687   map_entry_tree_less -> rbt_map_entry_rbt_less
   688   map_tree_greater -> map_rbt_greater
   689   map_tree_less -> map_rbt_less
   690   map_sorted -> map_rbt_sorted
   691   paint_sorted -> paint_rbt_sorted
   692   paint_lookup -> paint_rbt_lookup
   693   paint_tree_greater -> paint_rbt_greater
   694   paint_tree_less -> paint_rbt_less
   695   sorted_entries -> rbt_sorted_entries
   696   tree_greater_eq_trans -> rbt_greater_eq_trans
   697   tree_greater_nit -> rbt_greater_nit
   698   tree_greater_prop -> rbt_greater_prop
   699   tree_greater_simps -> rbt_greater_simps
   700   tree_greater_trans -> rbt_greater_trans
   701   tree_less_eq_trans -> rbt_less_eq_trans
   702   tree_less_nit -> rbt_less_nit
   703   tree_less_prop -> rbt_less_prop
   704   tree_less_simps -> rbt_less_simps
   705   tree_less_trans -> rbt_less_trans
   706   tree_ord_props -> rbt_ord_props
   707   union_Branch -> rbt_union_Branch
   708   union_is_rbt -> rbt_union_is_rbt
   709   unionw_is_rbt -> rbt_unionw_is_rbt
   710   unionwk_is_rbt -> rbt_unionwk_is_rbt
   711   unionwk_sorted -> rbt_unionwk_rbt_sorted
   712 
   713 * Theory HOL/Library/Float: Floating point numbers are now defined as
   714 a subset of the real numbers.  All operations are defined using the
   715 lifing-framework and proofs use the transfer method.  INCOMPATIBILITY.
   716 
   717   Changed Operations:
   718   float_abs -> abs
   719   float_nprt -> nprt
   720   float_pprt -> pprt
   721   pow2 -> use powr
   722   round_down -> float_round_down
   723   round_up -> float_round_up
   724   scale -> exponent
   725 
   726   Removed Operations:
   727   ceiling_fl, lb_mult, lb_mod, ub_mult, ub_mod
   728 
   729   Renamed Lemmas:
   730   abs_float_def -> Float.compute_float_abs
   731   bitlen_ge0 -> bitlen_nonneg
   732   bitlen.simps -> Float.compute_bitlen
   733   float_components -> Float_mantissa_exponent
   734   float_divl.simps -> Float.compute_float_divl
   735   float_divr.simps -> Float.compute_float_divr
   736   float_eq_odd -> mult_powr_eq_mult_powr_iff
   737   float_power -> real_of_float_power
   738   lapprox_posrat_def -> Float.compute_lapprox_posrat
   739   lapprox_rat.simps -> Float.compute_lapprox_rat
   740   le_float_def' -> Float.compute_float_le
   741   le_float_def -> less_eq_float.rep_eq
   742   less_float_def' -> Float.compute_float_less
   743   less_float_def -> less_float.rep_eq
   744   normfloat_def -> Float.compute_normfloat
   745   normfloat_imp_odd_or_zero -> mantissa_not_dvd and mantissa_noteq_0
   746   normfloat -> normfloat_def
   747   normfloat_unique -> use normfloat_def
   748   number_of_float_Float -> Float.compute_float_numeral, Float.compute_float_neg_numeral
   749   one_float_def -> Float.compute_float_one
   750   plus_float_def -> Float.compute_float_plus
   751   rapprox_posrat_def -> Float.compute_rapprox_posrat
   752   rapprox_rat.simps -> Float.compute_rapprox_rat
   753   real_of_float_0 -> zero_float.rep_eq
   754   real_of_float_1 -> one_float.rep_eq
   755   real_of_float_abs -> abs_float.rep_eq
   756   real_of_float_add -> plus_float.rep_eq
   757   real_of_float_minus -> uminus_float.rep_eq
   758   real_of_float_mult -> times_float.rep_eq
   759   real_of_float_simp -> Float.rep_eq
   760   real_of_float_sub -> minus_float.rep_eq
   761   round_down.simps -> Float.compute_float_round_down
   762   round_up.simps -> Float.compute_float_round_up
   763   times_float_def -> Float.compute_float_times
   764   uminus_float_def -> Float.compute_float_uminus
   765   zero_float_def -> Float.compute_float_zero
   766 
   767   Lemmas not necessary anymore, use the transfer method:
   768   bitlen_B0, bitlen_B1, bitlen_ge1, bitlen_Min, bitlen_Pls, float_divl,
   769   float_divr, float_le_simp, float_less1_mantissa_bound,
   770   float_less_simp, float_less_zero, float_le_zero,
   771   float_pos_less1_e_neg, float_pos_m_pos, float_split, float_split2,
   772   floor_pos_exp, lapprox_posrat, lapprox_posrat_bottom, lapprox_rat,
   773   lapprox_rat_bottom, normalized_float, rapprox_posrat,
   774   rapprox_posrat_le1, rapprox_rat, real_of_float_ge0_exp,
   775   real_of_float_neg_exp, real_of_float_nge0_exp, round_down floor_fl,
   776   round_up, zero_le_float, zero_less_float
   777 
   778 * New theory HOL/Library/DAList provides an abstract type for
   779 association lists with distinct keys.
   780 
   781 * Session HOL/IMP: Added new theory of abstract interpretation of
   782 annotated commands.
   783 
   784 * Session HOL-Import: Re-implementation from scratch is faster,
   785 simpler, and more scalable.  Requires a proof bundle, which is
   786 available as an external component.  Discontinued old (and mostly
   787 dead) Importer for HOL4 and HOL Light.  INCOMPATIBILITY.
   788 
   789 * Session HOL-Word: Discontinued many redundant theorems specific to
   790 type 'a word. INCOMPATIBILITY, use the corresponding generic theorems
   791 instead.
   792 
   793   word_sub_alt ~> word_sub_wi
   794   word_add_alt ~> word_add_def
   795   word_mult_alt ~> word_mult_def
   796   word_minus_alt ~> word_minus_def
   797   word_0_alt ~> word_0_wi
   798   word_1_alt ~> word_1_wi
   799   word_add_0 ~> add_0_left
   800   word_add_0_right ~> add_0_right
   801   word_mult_1 ~> mult_1_left
   802   word_mult_1_right ~> mult_1_right
   803   word_add_commute ~> add_commute
   804   word_add_assoc ~> add_assoc
   805   word_add_left_commute ~> add_left_commute
   806   word_mult_commute ~> mult_commute
   807   word_mult_assoc ~> mult_assoc
   808   word_mult_left_commute ~> mult_left_commute
   809   word_left_distrib ~> left_distrib
   810   word_right_distrib ~> right_distrib
   811   word_left_minus ~> left_minus
   812   word_diff_0_right ~> diff_0_right
   813   word_diff_self ~> diff_self
   814   word_sub_def ~> diff_minus
   815   word_diff_minus ~> diff_minus
   816   word_add_ac ~> add_ac
   817   word_mult_ac ~> mult_ac
   818   word_plus_ac0 ~> add_0_left add_0_right add_ac
   819   word_times_ac1 ~> mult_1_left mult_1_right mult_ac
   820   word_order_trans ~> order_trans
   821   word_order_refl ~> order_refl
   822   word_order_antisym ~> order_antisym
   823   word_order_linear ~> linorder_linear
   824   lenw1_zero_neq_one ~> zero_neq_one
   825   word_number_of_eq ~> number_of_eq
   826   word_of_int_add_hom ~> wi_hom_add
   827   word_of_int_sub_hom ~> wi_hom_sub
   828   word_of_int_mult_hom ~> wi_hom_mult
   829   word_of_int_minus_hom ~> wi_hom_neg
   830   word_of_int_succ_hom ~> wi_hom_succ
   831   word_of_int_pred_hom ~> wi_hom_pred
   832   word_of_int_0_hom ~> word_0_wi
   833   word_of_int_1_hom ~> word_1_wi
   834 
   835 * Session HOL-Word: New proof method "word_bitwise" for splitting
   836 machine word equalities and inequalities into logical circuits,
   837 defined in HOL/Word/WordBitwise.thy.  Supports addition, subtraction,
   838 multiplication, shifting by constants, bitwise operators and numeric
   839 constants.  Requires fixed-length word types, not 'a word.  Solves
   840 many standard word identities outright and converts more into first
   841 order problems amenable to blast or similar.  See also examples in
   842 HOL/Word/Examples/WordExamples.thy.
   843 
   844 * Session HOL-Probability: Introduced the type "'a measure" to
   845 represent measures, this replaces the records 'a algebra and 'a
   846 measure_space.  The locales based on subset_class now have two
   847 locale-parameters the space \<Omega> and the set of measurable sets M.
   848 The product of probability spaces uses now the same constant as the
   849 finite product of sigma-finite measure spaces "PiM :: ('i => 'a)
   850 measure".  Most constants are defined now outside of locales and gain
   851 an additional parameter, like null_sets, almost_eventually or \<mu>'.
   852 Measure space constructions for distributions and densities now got
   853 their own constants distr and density.  Instead of using locales to
   854 describe measure spaces with a finite space, the measure count_space
   855 and point_measure is introduced.  INCOMPATIBILITY.
   856 
   857   Renamed constants:
   858   measure -> emeasure
   859   finite_measure.\<mu>' -> measure
   860   product_algebra_generator -> prod_algebra
   861   product_prob_space.emb -> prod_emb
   862   product_prob_space.infprod_algebra -> PiM
   863 
   864   Removed locales:
   865   completeable_measure_space
   866   finite_measure_space
   867   finite_prob_space
   868   finite_product_finite_prob_space
   869   finite_product_sigma_algebra
   870   finite_sigma_algebra
   871   measure_space
   872   pair_finite_prob_space
   873   pair_finite_sigma_algebra
   874   pair_finite_space
   875   pair_sigma_algebra
   876   product_sigma_algebra
   877 
   878   Removed constants:
   879   conditional_space
   880   distribution -> use distr measure, or distributed predicate
   881   image_space
   882   joint_distribution -> use distr measure, or distributed predicate
   883   pair_measure_generator
   884   product_prob_space.infprod_algebra -> use PiM
   885   subvimage
   886 
   887   Replacement theorems:
   888   finite_additivity_sufficient -> ring_of_sets.countably_additiveI_finite
   889   finite_measure.empty_measure -> measure_empty
   890   finite_measure.finite_continuity_from_above -> finite_measure.finite_Lim_measure_decseq
   891   finite_measure.finite_continuity_from_below -> finite_measure.finite_Lim_measure_incseq
   892   finite_measure.finite_measure_countably_subadditive -> finite_measure.finite_measure_subadditive_countably
   893   finite_measure.finite_measure_eq -> finite_measure.emeasure_eq_measure
   894   finite_measure.finite_measure -> finite_measure.emeasure_finite
   895   finite_measure.finite_measure_finite_singleton -> finite_measure.finite_measure_eq_setsum_singleton
   896   finite_measure.positive_measure' -> measure_nonneg
   897   finite_measure.real_measure -> finite_measure.emeasure_real
   898   finite_product_prob_space.finite_measure_times -> finite_product_prob_space.finite_measure_PiM_emb
   899   finite_product_sigma_algebra.in_P -> sets_PiM_I_finite
   900   finite_product_sigma_algebra.P_empty -> space_PiM_empty, sets_PiM_empty
   901   information_space.conditional_entropy_eq -> information_space.conditional_entropy_simple_distributed
   902   information_space.conditional_entropy_positive -> information_space.conditional_entropy_nonneg_simple
   903   information_space.conditional_mutual_information_eq_mutual_information -> information_space.conditional_mutual_information_eq_mutual_information_simple
   904   information_space.conditional_mutual_information_generic_positive -> information_space.conditional_mutual_information_nonneg_simple
   905   information_space.conditional_mutual_information_positive -> information_space.conditional_mutual_information_nonneg_simple
   906   information_space.entropy_commute -> information_space.entropy_commute_simple
   907   information_space.entropy_eq -> information_space.entropy_simple_distributed
   908   information_space.entropy_generic_eq -> information_space.entropy_simple_distributed
   909   information_space.entropy_positive -> information_space.entropy_nonneg_simple
   910   information_space.entropy_uniform_max -> information_space.entropy_uniform
   911   information_space.KL_eq_0_imp -> information_space.KL_eq_0_iff_eq
   912   information_space.KL_eq_0 -> information_space.KL_same_eq_0
   913   information_space.KL_ge_0 -> information_space.KL_nonneg
   914   information_space.mutual_information_eq -> information_space.mutual_information_simple_distributed
   915   information_space.mutual_information_positive -> information_space.mutual_information_nonneg_simple
   916   Int_stable_cuboids -> Int_stable_atLeastAtMost
   917   Int_stable_product_algebra_generator -> positive_integral
   918   measure_preserving -> equality "distr M N f = N" "f : measurable M N"
   919   measure_space.additive -> emeasure_additive
   920   measure_space.AE_iff_null_set -> AE_iff_null
   921   measure_space.almost_everywhere_def -> eventually_ae_filter
   922   measure_space.almost_everywhere_vimage -> AE_distrD
   923   measure_space.continuity_from_above -> INF_emeasure_decseq
   924   measure_space.continuity_from_above_Lim -> Lim_emeasure_decseq
   925   measure_space.continuity_from_below_Lim -> Lim_emeasure_incseq
   926   measure_space.continuity_from_below -> SUP_emeasure_incseq
   927   measure_space_density -> emeasure_density
   928   measure_space.density_is_absolutely_continuous -> absolutely_continuousI_density
   929   measure_space.integrable_vimage -> integrable_distr
   930   measure_space.integral_translated_density -> integral_density
   931   measure_space.integral_vimage -> integral_distr
   932   measure_space.measure_additive -> plus_emeasure
   933   measure_space.measure_compl -> emeasure_compl
   934   measure_space.measure_countable_increasing -> emeasure_countable_increasing
   935   measure_space.measure_countably_subadditive -> emeasure_subadditive_countably
   936   measure_space.measure_decseq -> decseq_emeasure
   937   measure_space.measure_Diff -> emeasure_Diff
   938   measure_space.measure_Diff_null_set -> emeasure_Diff_null_set
   939   measure_space.measure_eq_0 -> emeasure_eq_0
   940   measure_space.measure_finitely_subadditive -> emeasure_subadditive_finite
   941   measure_space.measure_finite_singleton -> emeasure_eq_setsum_singleton
   942   measure_space.measure_incseq -> incseq_emeasure
   943   measure_space.measure_insert -> emeasure_insert
   944   measure_space.measure_mono -> emeasure_mono
   945   measure_space.measure_not_negative -> emeasure_not_MInf
   946   measure_space.measure_preserving_Int_stable -> measure_eqI_generator_eq
   947   measure_space.measure_setsum -> setsum_emeasure
   948   measure_space.measure_setsum_split -> setsum_emeasure_cover
   949   measure_space.measure_space_vimage -> emeasure_distr
   950   measure_space.measure_subadditive_finite -> emeasure_subadditive_finite
   951   measure_space.measure_subadditive -> subadditive
   952   measure_space.measure_top -> emeasure_space
   953   measure_space.measure_UN_eq_0 -> emeasure_UN_eq_0
   954   measure_space.measure_Un_null_set -> emeasure_Un_null_set
   955   measure_space.positive_integral_translated_density -> positive_integral_density
   956   measure_space.positive_integral_vimage -> positive_integral_distr
   957   measure_space.real_continuity_from_above -> Lim_measure_decseq
   958   measure_space.real_continuity_from_below -> Lim_measure_incseq
   959   measure_space.real_measure_countably_subadditive -> measure_subadditive_countably
   960   measure_space.real_measure_Diff -> measure_Diff
   961   measure_space.real_measure_finite_Union -> measure_finite_Union
   962   measure_space.real_measure_setsum_singleton -> measure_eq_setsum_singleton
   963   measure_space.real_measure_subadditive -> measure_subadditive
   964   measure_space.real_measure_Union -> measure_Union
   965   measure_space.real_measure_UNION -> measure_UNION
   966   measure_space.simple_function_vimage -> simple_function_comp
   967   measure_space.simple_integral_vimage -> simple_integral_distr
   968   measure_space.simple_integral_vimage -> simple_integral_distr
   969   measure_unique_Int_stable -> measure_eqI_generator_eq
   970   measure_unique_Int_stable_vimage -> measure_eqI_generator_eq
   971   pair_sigma_algebra.measurable_cut_fst -> sets_Pair1
   972   pair_sigma_algebra.measurable_cut_snd -> sets_Pair2
   973   pair_sigma_algebra.measurable_pair_image_fst -> measurable_Pair1
   974   pair_sigma_algebra.measurable_pair_image_snd -> measurable_Pair2
   975   pair_sigma_algebra.measurable_product_swap -> measurable_pair_swap_iff
   976   pair_sigma_algebra.pair_sigma_algebra_measurable -> measurable_pair_swap
   977   pair_sigma_algebra.pair_sigma_algebra_swap_measurable -> measurable_pair_swap'
   978   pair_sigma_algebra.sets_swap -> sets_pair_swap
   979   pair_sigma_finite.measure_cut_measurable_fst -> pair_sigma_finite.measurable_emeasure_Pair1
   980   pair_sigma_finite.measure_cut_measurable_snd -> pair_sigma_finite.measurable_emeasure_Pair2
   981   pair_sigma_finite.measure_preserving_swap -> pair_sigma_finite.distr_pair_swap
   982   pair_sigma_finite.pair_measure_alt2 -> pair_sigma_finite.emeasure_pair_measure_alt2
   983   pair_sigma_finite.pair_measure_alt -> pair_sigma_finite.emeasure_pair_measure_alt
   984   pair_sigma_finite.pair_measure_times -> pair_sigma_finite.emeasure_pair_measure_Times
   985   prob_space.indep_distribution_eq_measure -> prob_space.indep_vars_iff_distr_eq_PiM
   986   prob_space.indep_var_distributionD -> prob_space.indep_var_distribution_eq
   987   prob_space.measure_space_1 -> prob_space.emeasure_space_1
   988   prob_space.prob_space_vimage -> prob_space_distr
   989   prob_space.random_variable_restrict -> measurable_restrict
   990   prob_space_unique_Int_stable -> measure_eqI_prob_space
   991   product_algebraE -> prod_algebraE_all
   992   product_algebra_generator_der -> prod_algebra_eq_finite
   993   product_algebra_generator_into_space -> prod_algebra_sets_into_space
   994   product_algebraI -> sets_PiM_I_finite
   995   product_measure_exists -> product_sigma_finite.sigma_finite
   996   product_prob_space.finite_index_eq_finite_product -> product_prob_space.sets_PiM_generator
   997   product_prob_space.finite_measure_infprod_emb_Pi -> product_prob_space.measure_PiM_emb
   998   product_prob_space.infprod_spec -> product_prob_space.emeasure_PiM_emb_not_empty
   999   product_prob_space.measurable_component -> measurable_component_singleton
  1000   product_prob_space.measurable_emb -> measurable_prod_emb
  1001   product_prob_space.measurable_into_infprod_algebra -> measurable_PiM_single
  1002   product_prob_space.measurable_singleton_infprod -> measurable_component_singleton
  1003   product_prob_space.measure_emb -> emeasure_prod_emb
  1004   product_prob_space.measure_preserving_restrict -> product_prob_space.distr_restrict
  1005   product_sigma_algebra.product_algebra_into_space -> space_closed
  1006   product_sigma_finite.measure_fold -> product_sigma_finite.distr_merge
  1007   product_sigma_finite.measure_preserving_component_singelton -> product_sigma_finite.distr_singleton
  1008   product_sigma_finite.measure_preserving_merge -> product_sigma_finite.distr_merge
  1009   sequence_space.measure_infprod -> sequence_space.measure_PiM_countable
  1010   sets_product_algebra -> sets_PiM
  1011   sigma_algebra.measurable_sigma -> measurable_measure_of
  1012   sigma_finite_measure.disjoint_sigma_finite -> sigma_finite_disjoint
  1013   sigma_finite_measure.RN_deriv_vimage -> sigma_finite_measure.RN_deriv_distr
  1014   sigma_product_algebra_sigma_eq -> sigma_prod_algebra_sigma_eq
  1015   space_product_algebra -> space_PiM
  1016 
  1017 * Session HOL-TPTP: support to parse and import TPTP problems (all
  1018 languages) into Isabelle/HOL.
  1019 
  1020 
  1021 *** FOL ***
  1022 
  1023 * New "case_product" attribute (see HOL).
  1024 
  1025 
  1026 *** ZF ***
  1027 
  1028 * Greater support for structured proofs involving induction or case
  1029 analysis.
  1030 
  1031 * Much greater use of mathematical symbols.
  1032 
  1033 * Removal of many ML theorem bindings.  INCOMPATIBILITY.
  1034 
  1035 
  1036 *** ML ***
  1037 
  1038 * Antiquotation @{keyword "name"} produces a parser for outer syntax
  1039 from a minor keyword introduced via theory header declaration.
  1040 
  1041 * Antiquotation @{command_spec "name"} produces the
  1042 Outer_Syntax.command_spec from a major keyword introduced via theory
  1043 header declaration; it can be passed to Outer_Syntax.command etc.
  1044 
  1045 * Local_Theory.define no longer hard-wires default theorem name
  1046 "foo_def", but retains the binding as given.  If that is Binding.empty
  1047 / Attrib.empty_binding, the result is not registered as user-level
  1048 fact.  The Local_Theory.define_internal variant allows to specify a
  1049 non-empty name (used for the foundation in the background theory),
  1050 while omitting the fact binding in the user-context.  Potential
  1051 INCOMPATIBILITY for derived definitional packages: need to specify
  1052 naming policy for primitive definitions more explicitly.
  1053 
  1054 * Renamed Thm.capply to Thm.apply, and Thm.cabs to Thm.lambda in
  1055 conformance with similar operations in structure Term and Logic.
  1056 
  1057 * Antiquotation @{attributes [...]} embeds attribute source
  1058 representation into the ML text, which is particularly useful with
  1059 declarations like Local_Theory.note.
  1060 
  1061 * Structure Proof_Context follows standard naming scheme.  Old
  1062 ProofContext has been discontinued.  INCOMPATIBILITY.
  1063 
  1064 * Refined Local_Theory.declaration {syntax, pervasive}, with subtle
  1065 change of semantics: update is applied to auxiliary local theory
  1066 context as well.
  1067 
  1068 * Modernized some old-style infix operations:
  1069 
  1070   addeqcongs    ~> Simplifier.add_eqcong
  1071   deleqcongs    ~> Simplifier.del_eqcong
  1072   addcongs      ~> Simplifier.add_cong
  1073   delcongs      ~> Simplifier.del_cong
  1074   setmksimps    ~> Simplifier.set_mksimps
  1075   setmkcong     ~> Simplifier.set_mkcong
  1076   setmksym      ~> Simplifier.set_mksym
  1077   setmkeqTrue   ~> Simplifier.set_mkeqTrue
  1078   settermless   ~> Simplifier.set_termless
  1079   setsubgoaler  ~> Simplifier.set_subgoaler
  1080   addsplits     ~> Splitter.add_split
  1081   delsplits     ~> Splitter.del_split
  1082 
  1083 
  1084 *** System ***
  1085 
  1086 * USER_HOME settings variable points to cross-platform user home
  1087 directory, which coincides with HOME on POSIX systems only.  Likewise,
  1088 the Isabelle path specification "~" now expands to $USER_HOME, instead
  1089 of former $HOME.  A different default for USER_HOME may be set
  1090 explicitly in shell environment, before Isabelle settings are
  1091 evaluated.  Minor INCOMPATIBILITY: need to adapt Isabelle path where
  1092 the generic user home was intended.
  1093 
  1094 * ISABELLE_HOME_WINDOWS refers to ISABELLE_HOME in windows file name
  1095 notation, which is useful for the jEdit file browser, for example.
  1096 
  1097 * ISABELLE_JDK_HOME settings variable points to JDK with javac and jar
  1098 (not just JRE).
  1099 
  1100 
  1101 
  1102 New in Isabelle2011-1 (October 2011)
  1103 ------------------------------------
  1104 
  1105 *** General ***
  1106 
  1107 * Improved Isabelle/jEdit Prover IDE (PIDE), which can be invoked as
  1108 "isabelle jedit" or "ISABELLE_HOME/Isabelle" on the command line.
  1109 
  1110   - Management of multiple theory files directly from the editor
  1111     buffer store -- bypassing the file-system (no requirement to save
  1112     files for checking).
  1113 
  1114   - Markup of formal entities within the text buffer, with semantic
  1115     highlighting, tooltips and hyperlinks to jump to defining source
  1116     positions.
  1117 
  1118   - Improved text rendering, with sub/superscripts in the source
  1119     buffer (including support for copy/paste wrt. output panel, HTML
  1120     theory output and other non-Isabelle text boxes).
  1121 
  1122   - Refined scheduling of proof checking and printing of results,
  1123     based on interactive editor view.  (Note: jEdit folding and
  1124     narrowing allows to restrict buffer perspectives explicitly.)
  1125 
  1126   - Reduced CPU performance requirements, usable on machines with few
  1127     cores.
  1128 
  1129   - Reduced memory requirements due to pruning of unused document
  1130     versions (garbage collection).
  1131 
  1132 See also ~~/src/Tools/jEdit/README.html for further information,
  1133 including some remaining limitations.
  1134 
  1135 * Theory loader: source files are exclusively located via the master
  1136 directory of each theory node (where the .thy file itself resides).
  1137 The global load path (such as src/HOL/Library) has been discontinued.
  1138 Note that the path element ~~ may be used to reference theories in the
  1139 Isabelle home folder -- for instance, "~~/src/HOL/Library/FuncSet".
  1140 INCOMPATIBILITY.
  1141 
  1142 * Theory loader: source files are identified by content via SHA1
  1143 digests.  Discontinued former path/modtime identification and optional
  1144 ISABELLE_FILE_IDENT plugin scripts.
  1145 
  1146 * Parallelization of nested Isar proofs is subject to
  1147 Goal.parallel_proofs_threshold (default 100).  See also isabelle
  1148 usedir option -Q.
  1149 
  1150 * Name space: former unsynchronized references are now proper
  1151 configuration options, with more conventional names:
  1152 
  1153   long_names   ~> names_long
  1154   short_names  ~> names_short
  1155   unique_names ~> names_unique
  1156 
  1157 Minor INCOMPATIBILITY, need to declare options in context like this:
  1158 
  1159   declare [[names_unique = false]]
  1160 
  1161 * Literal facts `prop` may contain dummy patterns, e.g. `_ = _`.  Note
  1162 that the result needs to be unique, which means fact specifications
  1163 may have to be refined after enriching a proof context.
  1164 
  1165 * Attribute "case_names" has been refined: the assumptions in each case
  1166 can be named now by following the case name with [name1 name2 ...].
  1167 
  1168 * Isabelle/Isar reference manual has been updated and extended:
  1169   - "Synopsis" provides a catalog of main Isar language concepts.
  1170   - Formal references in syntax diagrams, via @{rail} antiquotation.
  1171   - Updated material from classic "ref" manual, notably about
  1172     "Classical Reasoner".
  1173 
  1174 
  1175 *** HOL ***
  1176 
  1177 * Class bot and top require underlying partial order rather than
  1178 preorder: uniqueness of bot and top is guaranteed.  INCOMPATIBILITY.
  1179 
  1180 * Class complete_lattice: generalized a couple of lemmas from sets;
  1181 generalized theorems INF_cong and SUP_cong.  New type classes for
  1182 complete boolean algebras and complete linear orders.  Lemmas
  1183 Inf_less_iff, less_Sup_iff, INF_less_iff, less_SUP_iff now reside in
  1184 class complete_linorder.
  1185 
  1186 Changed proposition of lemmas Inf_bool_def, Sup_bool_def, Inf_fun_def,
  1187 Sup_fun_def, Inf_apply, Sup_apply.
  1188 
  1189 Removed redundant lemmas (the right hand side gives hints how to
  1190 replace them for (metis ...), or (simp only: ...) proofs):
  1191 
  1192   Inf_singleton ~> Inf_insert [where A="{}", unfolded Inf_empty inf_top_right]
  1193   Sup_singleton ~> Sup_insert [where A="{}", unfolded Sup_empty sup_bot_right]
  1194   Inf_binary ~> Inf_insert, Inf_empty, and inf_top_right
  1195   Sup_binary ~> Sup_insert, Sup_empty, and sup_bot_right
  1196   Int_eq_Inter ~> Inf_insert, Inf_empty, and inf_top_right
  1197   Un_eq_Union ~> Sup_insert, Sup_empty, and sup_bot_right
  1198   Inter_def ~> INF_def, image_def
  1199   Union_def ~> SUP_def, image_def
  1200   INT_eq ~> INF_def, and image_def
  1201   UN_eq ~> SUP_def, and image_def
  1202   INF_subset ~> INF_superset_mono [OF _ order_refl]
  1203 
  1204 More consistent and comprehensive names:
  1205 
  1206   INTER_eq_Inter_image ~> INF_def
  1207   UNION_eq_Union_image ~> SUP_def
  1208   INFI_def ~> INF_def
  1209   SUPR_def ~> SUP_def
  1210   INF_leI ~> INF_lower
  1211   INF_leI2 ~> INF_lower2
  1212   le_INFI ~> INF_greatest
  1213   le_SUPI ~> SUP_upper
  1214   le_SUPI2 ~> SUP_upper2
  1215   SUP_leI ~> SUP_least
  1216   INFI_bool_eq ~> INF_bool_eq
  1217   SUPR_bool_eq ~> SUP_bool_eq
  1218   INFI_apply ~> INF_apply
  1219   SUPR_apply ~> SUP_apply
  1220   INTER_def ~> INTER_eq
  1221   UNION_def ~> UNION_eq
  1222 
  1223 INCOMPATIBILITY.
  1224 
  1225 * Renamed theory Complete_Lattice to Complete_Lattices.
  1226 INCOMPATIBILITY.
  1227 
  1228 * Theory Complete_Lattices: lemmas Inf_eq_top_iff, INF_eq_top_iff,
  1229 INF_image, Inf_insert, INF_top, Inf_top_conv, INF_top_conv, SUP_bot,
  1230 Sup_bot_conv, SUP_bot_conv, Sup_eq_top_iff, SUP_eq_top_iff, SUP_image,
  1231 Sup_insert are now declared as [simp].  INCOMPATIBILITY.
  1232 
  1233 * Theory Lattice: lemmas compl_inf_bot, compl_le_comp_iff,
  1234 compl_sup_top, inf_idem, inf_left_idem, inf_sup_absorb, sup_idem,
  1235 sup_inf_absob, sup_left_idem are now declared as [simp].  Minor
  1236 INCOMPATIBILITY.
  1237 
  1238 * Added syntactic classes "inf" and "sup" for the respective
  1239 constants.  INCOMPATIBILITY: Changes in the argument order of the
  1240 (mostly internal) locale predicates for some derived classes.
  1241 
  1242 * Theorem collections ball_simps and bex_simps do not contain theorems
  1243 referring to UNION any longer; these have been moved to collection
  1244 UN_ball_bex_simps.  INCOMPATIBILITY.
  1245 
  1246 * Theory Archimedean_Field: floor now is defined as parameter of a
  1247 separate type class floor_ceiling.
  1248 
  1249 * Theory Finite_Set: more coherent development of fold_set locales:
  1250 
  1251     locale fun_left_comm ~> locale comp_fun_commute
  1252     locale fun_left_comm_idem ~> locale comp_fun_idem
  1253 
  1254 Both use point-free characterization; interpretation proofs may need
  1255 adjustment.  INCOMPATIBILITY.
  1256 
  1257 * Theory Limits: Type "'a net" has been renamed to "'a filter", in
  1258 accordance with standard mathematical terminology. INCOMPATIBILITY.
  1259 
  1260 * Theory Complex_Main: The locale interpretations for the
  1261 bounded_linear and bounded_bilinear locales have been removed, in
  1262 order to reduce the number of duplicate lemmas. Users must use the
  1263 original names for distributivity theorems, potential INCOMPATIBILITY.
  1264 
  1265   divide.add ~> add_divide_distrib
  1266   divide.diff ~> diff_divide_distrib
  1267   divide.setsum ~> setsum_divide_distrib
  1268   mult.add_right ~> right_distrib
  1269   mult.diff_right ~> right_diff_distrib
  1270   mult_right.setsum ~> setsum_right_distrib
  1271   mult_left.diff ~> left_diff_distrib
  1272 
  1273 * Theory Complex_Main: Several redundant theorems have been removed or
  1274 replaced by more general versions. INCOMPATIBILITY.
  1275 
  1276   real_diff_def ~> minus_real_def
  1277   real_divide_def ~> divide_real_def
  1278   real_less_def ~> less_le
  1279   real_abs_def ~> abs_real_def
  1280   real_sgn_def ~> sgn_real_def
  1281   real_mult_commute ~> mult_commute
  1282   real_mult_assoc ~> mult_assoc
  1283   real_mult_1 ~> mult_1_left
  1284   real_add_mult_distrib ~> left_distrib
  1285   real_zero_not_eq_one ~> zero_neq_one
  1286   real_mult_inverse_left ~> left_inverse
  1287   INVERSE_ZERO ~> inverse_zero
  1288   real_le_refl ~> order_refl
  1289   real_le_antisym ~> order_antisym
  1290   real_le_trans ~> order_trans
  1291   real_le_linear ~> linear
  1292   real_le_eq_diff ~> le_iff_diff_le_0
  1293   real_add_left_mono ~> add_left_mono
  1294   real_mult_order ~> mult_pos_pos
  1295   real_mult_less_mono2 ~> mult_strict_left_mono
  1296   real_of_int_real_of_nat ~> real_of_int_of_nat_eq
  1297   real_0_le_divide_iff ~> zero_le_divide_iff
  1298   realpow_two_disj ~> power2_eq_iff
  1299   real_squared_diff_one_factored ~> square_diff_one_factored
  1300   realpow_two_diff ~> square_diff_square_factored
  1301   reals_complete2 ~> complete_real
  1302   real_sum_squared_expand ~> power2_sum
  1303   exp_ln_eq ~> ln_unique
  1304   expi_add ~> exp_add
  1305   expi_zero ~> exp_zero
  1306   lemma_DERIV_subst ~> DERIV_cong
  1307   LIMSEQ_Zfun_iff ~> tendsto_Zfun_iff
  1308   LIMSEQ_const ~> tendsto_const
  1309   LIMSEQ_norm ~> tendsto_norm
  1310   LIMSEQ_add ~> tendsto_add
  1311   LIMSEQ_minus ~> tendsto_minus
  1312   LIMSEQ_minus_cancel ~> tendsto_minus_cancel
  1313   LIMSEQ_diff ~> tendsto_diff
  1314   bounded_linear.LIMSEQ ~> bounded_linear.tendsto
  1315   bounded_bilinear.LIMSEQ ~> bounded_bilinear.tendsto
  1316   LIMSEQ_mult ~> tendsto_mult
  1317   LIMSEQ_inverse ~> tendsto_inverse
  1318   LIMSEQ_divide ~> tendsto_divide
  1319   LIMSEQ_pow ~> tendsto_power
  1320   LIMSEQ_setsum ~> tendsto_setsum
  1321   LIMSEQ_setprod ~> tendsto_setprod
  1322   LIMSEQ_norm_zero ~> tendsto_norm_zero_iff
  1323   LIMSEQ_rabs_zero ~> tendsto_rabs_zero_iff
  1324   LIMSEQ_imp_rabs ~> tendsto_rabs
  1325   LIMSEQ_add_minus ~> tendsto_add [OF _ tendsto_minus]
  1326   LIMSEQ_add_const ~> tendsto_add [OF _ tendsto_const]
  1327   LIMSEQ_diff_const ~> tendsto_diff [OF _ tendsto_const]
  1328   LIMSEQ_Complex ~> tendsto_Complex
  1329   LIM_ident ~> tendsto_ident_at
  1330   LIM_const ~> tendsto_const
  1331   LIM_add ~> tendsto_add
  1332   LIM_add_zero ~> tendsto_add_zero
  1333   LIM_minus ~> tendsto_minus
  1334   LIM_diff ~> tendsto_diff
  1335   LIM_norm ~> tendsto_norm
  1336   LIM_norm_zero ~> tendsto_norm_zero
  1337   LIM_norm_zero_cancel ~> tendsto_norm_zero_cancel
  1338   LIM_norm_zero_iff ~> tendsto_norm_zero_iff
  1339   LIM_rabs ~> tendsto_rabs
  1340   LIM_rabs_zero ~> tendsto_rabs_zero
  1341   LIM_rabs_zero_cancel ~> tendsto_rabs_zero_cancel
  1342   LIM_rabs_zero_iff ~> tendsto_rabs_zero_iff
  1343   LIM_compose ~> tendsto_compose
  1344   LIM_mult ~> tendsto_mult
  1345   LIM_scaleR ~> tendsto_scaleR
  1346   LIM_of_real ~> tendsto_of_real
  1347   LIM_power ~> tendsto_power
  1348   LIM_inverse ~> tendsto_inverse
  1349   LIM_sgn ~> tendsto_sgn
  1350   isCont_LIM_compose ~> isCont_tendsto_compose
  1351   bounded_linear.LIM ~> bounded_linear.tendsto
  1352   bounded_linear.LIM_zero ~> bounded_linear.tendsto_zero
  1353   bounded_bilinear.LIM ~> bounded_bilinear.tendsto
  1354   bounded_bilinear.LIM_prod_zero ~> bounded_bilinear.tendsto_zero
  1355   bounded_bilinear.LIM_left_zero ~> bounded_bilinear.tendsto_left_zero
  1356   bounded_bilinear.LIM_right_zero ~> bounded_bilinear.tendsto_right_zero
  1357   LIM_inverse_fun ~> tendsto_inverse [OF tendsto_ident_at]
  1358 
  1359 * Theory Complex_Main: The definition of infinite series was
  1360 generalized.  Now it is defined on the type class {topological_space,
  1361 comm_monoid_add}.  Hence it is useable also for extended real numbers.
  1362 
  1363 * Theory Complex_Main: The complex exponential function "expi" is now
  1364 a type-constrained abbreviation for "exp :: complex => complex"; thus
  1365 several polymorphic lemmas about "exp" are now applicable to "expi".
  1366 
  1367 * Code generation:
  1368 
  1369   - Theory Library/Code_Char_ord provides native ordering of
  1370     characters in the target language.
  1371 
  1372   - Commands code_module and code_library are legacy, use export_code
  1373     instead.
  1374 
  1375   - Method "evaluation" is legacy, use method "eval" instead.
  1376 
  1377   - Legacy evaluator "SML" is deactivated by default.  May be
  1378     reactivated by the following theory command:
  1379 
  1380       setup {* Value.add_evaluator ("SML", Codegen.eval_term) *}
  1381 
  1382 * Declare ext [intro] by default.  Rare INCOMPATIBILITY.
  1383 
  1384 * New proof method "induction" that gives induction hypotheses the
  1385 name "IH", thus distinguishing them from further hypotheses that come
  1386 from rule induction.  The latter are still called "hyps".  Method
  1387 "induction" is a thin wrapper around "induct" and follows the same
  1388 syntax.
  1389 
  1390 * Method "fastsimp" has been renamed to "fastforce", but "fastsimp" is
  1391 still available as a legacy feature for some time.
  1392 
  1393 * Nitpick:
  1394   - Added "need" and "total_consts" options.
  1395   - Reintroduced "show_skolems" option by popular demand.
  1396   - Renamed attribute: nitpick_def ~> nitpick_unfold.
  1397     INCOMPATIBILITY.
  1398 
  1399 * Sledgehammer:
  1400   - Use quasi-sound (and efficient) translations by default.
  1401   - Added support for the following provers: E-ToFoF, LEO-II,
  1402     Satallax, SNARK, Waldmeister, and Z3 with TPTP syntax.
  1403   - Automatically preplay and minimize proofs before showing them if
  1404     this can be done within reasonable time.
  1405   - sledgehammer available_provers ~> sledgehammer supported_provers.
  1406     INCOMPATIBILITY.
  1407   - Added "preplay_timeout", "slicing", "type_enc", "sound",
  1408     "max_mono_iters", and "max_new_mono_instances" options.
  1409   - Removed "explicit_apply" and "full_types" options as well as "Full
  1410     Types" Proof General menu item. INCOMPATIBILITY.
  1411 
  1412 * Metis:
  1413   - Removed "metisF" -- use "metis" instead. INCOMPATIBILITY.
  1414   - Obsoleted "metisFT" -- use "metis (full_types)" instead.
  1415     INCOMPATIBILITY.
  1416 
  1417 * Command 'try':
  1418   - Renamed 'try_methods' and added "simp:", "intro:", "dest:", and
  1419     "elim:" options. INCOMPATIBILITY.
  1420   - Introduced 'try' that not only runs 'try_methods' but also
  1421     'solve_direct', 'sledgehammer', 'quickcheck', and 'nitpick'.
  1422 
  1423 * Quickcheck:
  1424   - Added "eval" option to evaluate terms for the found counterexample
  1425     (currently only supported by the default (exhaustive) tester).
  1426   - Added post-processing of terms to obtain readable counterexamples
  1427     (currently only supported by the default (exhaustive) tester).
  1428   - New counterexample generator quickcheck[narrowing] enables
  1429     narrowing-based testing.  Requires the Glasgow Haskell compiler
  1430     with its installation location defined in the Isabelle settings
  1431     environment as ISABELLE_GHC.
  1432   - Removed quickcheck tester "SML" based on the SML code generator
  1433     (formly in HOL/Library).
  1434 
  1435 * Function package: discontinued option "tailrec".  INCOMPATIBILITY,
  1436 use 'partial_function' instead.
  1437 
  1438 * Theory Library/Extended_Reals replaces now the positive extended
  1439 reals found in probability theory. This file is extended by
  1440 Multivariate_Analysis/Extended_Real_Limits.
  1441 
  1442 * Theory Library/Old_Recdef: old 'recdef' package has been moved here,
  1443 from where it must be imported explicitly if it is really required.
  1444 INCOMPATIBILITY.
  1445 
  1446 * Theory Library/Wfrec: well-founded recursion combinator "wfrec" has
  1447 been moved here.  INCOMPATIBILITY.
  1448 
  1449 * Theory Library/Saturated provides type of numbers with saturated
  1450 arithmetic.
  1451 
  1452 * Theory Library/Product_Lattice defines a pointwise ordering for the
  1453 product type 'a * 'b, and provides instance proofs for various order
  1454 and lattice type classes.
  1455 
  1456 * Theory Library/Countable now provides the "countable_datatype" proof
  1457 method for proving "countable" class instances for datatypes.
  1458 
  1459 * Theory Library/Cset_Monad allows do notation for computable sets
  1460 (cset) via the generic monad ad-hoc overloading facility.
  1461 
  1462 * Library: Theories of common data structures are split into theories
  1463 for implementation, an invariant-ensuring type, and connection to an
  1464 abstract type. INCOMPATIBILITY.
  1465 
  1466   - RBT is split into RBT and RBT_Mapping.
  1467   - AssocList is split and renamed into AList and AList_Mapping.
  1468   - DList is split into DList_Impl, DList, and DList_Cset.
  1469   - Cset is split into Cset and List_Cset.
  1470 
  1471 * Theory Library/Nat_Infinity has been renamed to
  1472 Library/Extended_Nat, with name changes of the following types and
  1473 constants:
  1474 
  1475   type inat   ~> type enat
  1476   Fin         ~> enat
  1477   Infty       ~> infinity (overloaded)
  1478   iSuc        ~> eSuc
  1479   the_Fin     ~> the_enat
  1480 
  1481 Every theorem name containing "inat", "Fin", "Infty", or "iSuc" has
  1482 been renamed accordingly. INCOMPATIBILITY.
  1483 
  1484 * Session Multivariate_Analysis: The euclidean_space type class now
  1485 fixes a constant "Basis :: 'a set" consisting of the standard
  1486 orthonormal basis for the type. Users now have the option of
  1487 quantifying over this set instead of using the "basis" function, e.g.
  1488 "ALL x:Basis. P x" vs "ALL i<DIM('a). P (basis i)".
  1489 
  1490 * Session Multivariate_Analysis: Type "('a, 'b) cart" has been renamed
  1491 to "('a, 'b) vec" (the syntax "'a ^ 'b" remains unaffected). Constants
  1492 "Cart_nth" and "Cart_lambda" have been respectively renamed to
  1493 "vec_nth" and "vec_lambda"; theorems mentioning those names have
  1494 changed to match. Definition theorems for overloaded constants now use
  1495 the standard "foo_vec_def" naming scheme. A few other theorems have
  1496 been renamed as follows (INCOMPATIBILITY):
  1497 
  1498   Cart_eq          ~> vec_eq_iff
  1499   dist_nth_le_cart ~> dist_vec_nth_le
  1500   tendsto_vector   ~> vec_tendstoI
  1501   Cauchy_vector    ~> vec_CauchyI
  1502 
  1503 * Session Multivariate_Analysis: Several duplicate theorems have been
  1504 removed, and other theorems have been renamed or replaced with more
  1505 general versions. INCOMPATIBILITY.
  1506 
  1507   finite_choice ~> finite_set_choice
  1508   eventually_conjI ~> eventually_conj
  1509   eventually_and ~> eventually_conj_iff
  1510   eventually_false ~> eventually_False
  1511   setsum_norm ~> norm_setsum
  1512   Lim_sequentially ~> LIMSEQ_def
  1513   Lim_ident_at ~> LIM_ident
  1514   Lim_const ~> tendsto_const
  1515   Lim_cmul ~> tendsto_scaleR [OF tendsto_const]
  1516   Lim_neg ~> tendsto_minus
  1517   Lim_add ~> tendsto_add
  1518   Lim_sub ~> tendsto_diff
  1519   Lim_mul ~> tendsto_scaleR
  1520   Lim_vmul ~> tendsto_scaleR [OF _ tendsto_const]
  1521   Lim_null_norm ~> tendsto_norm_zero_iff [symmetric]
  1522   Lim_linear ~> bounded_linear.tendsto
  1523   Lim_component ~> tendsto_euclidean_component
  1524   Lim_component_cart ~> tendsto_vec_nth
  1525   Lim_inner ~> tendsto_inner [OF tendsto_const]
  1526   dot_lsum ~> inner_setsum_left
  1527   dot_rsum ~> inner_setsum_right
  1528   continuous_cmul ~> continuous_scaleR [OF continuous_const]
  1529   continuous_neg ~> continuous_minus
  1530   continuous_sub ~> continuous_diff
  1531   continuous_vmul ~> continuous_scaleR [OF _ continuous_const]
  1532   continuous_mul ~> continuous_scaleR
  1533   continuous_inv ~> continuous_inverse
  1534   continuous_at_within_inv ~> continuous_at_within_inverse
  1535   continuous_at_inv ~> continuous_at_inverse
  1536   continuous_at_norm ~> continuous_norm [OF continuous_at_id]
  1537   continuous_at_infnorm ~> continuous_infnorm [OF continuous_at_id]
  1538   continuous_at_component ~> continuous_component [OF continuous_at_id]
  1539   continuous_on_neg ~> continuous_on_minus
  1540   continuous_on_sub ~> continuous_on_diff
  1541   continuous_on_cmul ~> continuous_on_scaleR [OF continuous_on_const]
  1542   continuous_on_vmul ~> continuous_on_scaleR [OF _ continuous_on_const]
  1543   continuous_on_mul ~> continuous_on_scaleR
  1544   continuous_on_mul_real ~> continuous_on_mult
  1545   continuous_on_inner ~> continuous_on_inner [OF continuous_on_const]
  1546   continuous_on_norm ~> continuous_on_norm [OF continuous_on_id]
  1547   continuous_on_inverse ~> continuous_on_inv
  1548   uniformly_continuous_on_neg ~> uniformly_continuous_on_minus
  1549   uniformly_continuous_on_sub ~> uniformly_continuous_on_diff
  1550   subset_interior ~> interior_mono
  1551   subset_closure ~> closure_mono
  1552   closure_univ ~> closure_UNIV
  1553   real_arch_lt ~> reals_Archimedean2
  1554   real_arch ~> reals_Archimedean3
  1555   real_abs_norm ~> abs_norm_cancel
  1556   real_abs_sub_norm ~> norm_triangle_ineq3
  1557   norm_cauchy_schwarz_abs ~> Cauchy_Schwarz_ineq2
  1558 
  1559 * Session HOL-Probability:
  1560   - Caratheodory's extension lemma is now proved for ring_of_sets.
  1561   - Infinite products of probability measures are now available.
  1562   - Sigma closure is independent, if the generator is independent
  1563   - Use extended reals instead of positive extended
  1564     reals. INCOMPATIBILITY.
  1565 
  1566 * Session HOLCF: Discontinued legacy theorem names, INCOMPATIBILITY.
  1567 
  1568   expand_fun_below ~> fun_below_iff
  1569   below_fun_ext ~> fun_belowI
  1570   expand_cfun_eq ~> cfun_eq_iff
  1571   ext_cfun ~> cfun_eqI
  1572   expand_cfun_below ~> cfun_below_iff
  1573   below_cfun_ext ~> cfun_belowI
  1574   monofun_fun_fun ~> fun_belowD
  1575   monofun_fun_arg ~> monofunE
  1576   monofun_lub_fun ~> adm_monofun [THEN admD]
  1577   cont_lub_fun ~> adm_cont [THEN admD]
  1578   cont2cont_Rep_CFun ~> cont2cont_APP
  1579   cont_Rep_CFun_app ~> cont_APP_app
  1580   cont_Rep_CFun_app_app ~> cont_APP_app_app
  1581   cont_cfun_fun ~> cont_Rep_cfun1 [THEN contE]
  1582   cont_cfun_arg ~> cont_Rep_cfun2 [THEN contE]
  1583   contlub_cfun ~> lub_APP [symmetric]
  1584   contlub_LAM ~> lub_LAM [symmetric]
  1585   thelubI ~> lub_eqI
  1586   UU_I ~> bottomI
  1587   lift_distinct1 ~> lift.distinct(1)
  1588   lift_distinct2 ~> lift.distinct(2)
  1589   Def_not_UU ~> lift.distinct(2)
  1590   Def_inject ~> lift.inject
  1591   below_UU_iff ~> below_bottom_iff
  1592   eq_UU_iff ~> eq_bottom_iff
  1593 
  1594 
  1595 *** Document preparation ***
  1596 
  1597 * Antiquotation @{rail} layouts railroad syntax diagrams, see also
  1598 isar-ref manual, both for description and actual application of the
  1599 same.
  1600 
  1601 * Antiquotation @{value} evaluates the given term and presents its
  1602 result.
  1603 
  1604 * Antiquotations: term style "isub" provides ad-hoc conversion of
  1605 variables x1, y23 into subscripted form x\<^isub>1,
  1606 y\<^isub>2\<^isub>3.
  1607 
  1608 * Predefined LaTeX macros for Isabelle symbols \<bind> and \<then>
  1609 (e.g. see ~~/src/HOL/Library/Monad_Syntax.thy).
  1610 
  1611 * Localized \isabellestyle switch can be used within blocks or groups
  1612 like this:
  1613 
  1614   \isabellestyle{it}  %preferred default
  1615   {\isabellestylett @{text "typewriter stuff"}}
  1616 
  1617 * Discontinued special treatment of hard tabulators.  Implicit
  1618 tab-width is now defined as 1.  Potential INCOMPATIBILITY for visual
  1619 layouts.
  1620 
  1621 
  1622 *** ML ***
  1623 
  1624 * The inner syntax of sort/type/term/prop supports inlined YXML
  1625 representations within quoted string tokens.  By encoding logical
  1626 entities via Term_XML (in ML or Scala) concrete syntax can be
  1627 bypassed, which is particularly useful for producing bits of text
  1628 under external program control.
  1629 
  1630 * Antiquotations for ML and document preparation are managed as theory
  1631 data, which requires explicit setup.
  1632 
  1633 * Isabelle_Process.is_active allows tools to check if the official
  1634 process wrapper is running (Isabelle/Scala/jEdit) or the old TTY loop
  1635 (better known as Proof General).
  1636 
  1637 * Structure Proof_Context follows standard naming scheme.  Old
  1638 ProofContext is still available for some time as legacy alias.
  1639 
  1640 * Structure Timing provides various operations for timing; supersedes
  1641 former start_timing/end_timing etc.
  1642 
  1643 * Path.print is the official way to show file-system paths to users
  1644 (including quotes etc.).
  1645 
  1646 * Inner syntax: identifiers in parse trees of generic categories
  1647 "logic", "aprop", "idt" etc. carry position information (disguised as
  1648 type constraints).  Occasional INCOMPATIBILITY with non-compliant
  1649 translations that choke on unexpected type constraints.  Positions can
  1650 be stripped in ML translations via Syntax.strip_positions /
  1651 Syntax.strip_positions_ast, or via the syntax constant
  1652 "_strip_positions" within parse trees.  As last resort, positions can
  1653 be disabled via the configuration option Syntax.positions, which is
  1654 called "syntax_positions" in Isar attribute syntax.
  1655 
  1656 * Discontinued special status of various ML structures that contribute
  1657 to structure Syntax (Ast, Lexicon, Mixfix, Parser, Printer etc.): less
  1658 pervasive content, no inclusion in structure Syntax.  INCOMPATIBILITY,
  1659 refer directly to Ast.Constant, Lexicon.is_identifier,
  1660 Syntax_Trans.mk_binder_tr etc.
  1661 
  1662 * Typed print translation: discontinued show_sorts argument, which is
  1663 already available via context of "advanced" translation.
  1664 
  1665 * Refined PARALLEL_GOALS tactical: degrades gracefully for schematic
  1666 goal states; body tactic needs to address all subgoals uniformly.
  1667 
  1668 * Slightly more special eq_list/eq_set, with shortcut involving
  1669 pointer equality (assumes that eq relation is reflexive).
  1670 
  1671 * Classical tactics use proper Proof.context instead of historic types
  1672 claset/clasimpset.  Old-style declarations like addIs, addEs, addDs
  1673 operate directly on Proof.context.  Raw type claset retains its use as
  1674 snapshot of the classical context, which can be recovered via
  1675 (put_claset HOL_cs) etc.  Type clasimpset has been discontinued.
  1676 INCOMPATIBILITY, classical tactics and derived proof methods require
  1677 proper Proof.context.
  1678 
  1679 
  1680 *** System ***
  1681 
  1682 * Discontinued support for Poly/ML 5.2, which was the last version
  1683 without proper multithreading and TimeLimit implementation.
  1684 
  1685 * Discontinued old lib/scripts/polyml-platform, which has been
  1686 obsolete since Isabelle2009-2.
  1687 
  1688 * Various optional external tools are referenced more robustly and
  1689 uniformly by explicit Isabelle settings as follows:
  1690 
  1691   ISABELLE_CSDP   (formerly CSDP_EXE)
  1692   ISABELLE_GHC    (formerly EXEC_GHC or GHC_PATH)
  1693   ISABELLE_OCAML  (formerly EXEC_OCAML)
  1694   ISABELLE_SWIPL  (formerly EXEC_SWIPL)
  1695   ISABELLE_YAP    (formerly EXEC_YAP)
  1696 
  1697 Note that automated detection from the file-system or search path has
  1698 been discontinued.  INCOMPATIBILITY.
  1699 
  1700 * Scala layer provides JVM method invocation service for static
  1701 methods of type (String)String, see Invoke_Scala.method in ML.  For
  1702 example:
  1703 
  1704   Invoke_Scala.method "java.lang.System.getProperty" "java.home"
  1705 
  1706 Together with YXML.string_of_body/parse_body and XML.Encode/Decode
  1707 this allows to pass structured values between ML and Scala.
  1708 
  1709 * The IsabelleText fonts includes some further glyphs to support the
  1710 Prover IDE.  Potential INCOMPATIBILITY: users who happen to have
  1711 installed a local copy (which is normally *not* required) need to
  1712 delete or update it from ~~/lib/fonts/.
  1713 
  1714 
  1715 
  1716 New in Isabelle2011 (January 2011)
  1717 ----------------------------------
  1718 
  1719 *** General ***
  1720 
  1721 * Experimental Prover IDE based on Isabelle/Scala and jEdit (see
  1722 src/Tools/jEdit).  This also serves as IDE for Isabelle/ML, with
  1723 useful tooltips and hyperlinks produced from its static analysis.  The
  1724 bundled component provides an executable Isabelle tool that can be run
  1725 like this:
  1726 
  1727   Isabelle2011/bin/isabelle jedit
  1728 
  1729 * Significantly improved Isabelle/Isar implementation manual.
  1730 
  1731 * System settings: ISABELLE_HOME_USER now includes ISABELLE_IDENTIFIER
  1732 (and thus refers to something like $HOME/.isabelle/Isabelle2011),
  1733 while the default heap location within that directory lacks that extra
  1734 suffix.  This isolates multiple Isabelle installations from each
  1735 other, avoiding problems with old settings in new versions.
  1736 INCOMPATIBILITY, need to copy/upgrade old user settings manually.
  1737 
  1738 * Source files are always encoded as UTF-8, instead of old-fashioned
  1739 ISO-Latin-1.  INCOMPATIBILITY.  Isabelle LaTeX documents might require
  1740 the following package declarations:
  1741 
  1742   \usepackage[utf8]{inputenc}
  1743   \usepackage{textcomp}
  1744 
  1745 * Explicit treatment of UTF-8 sequences as Isabelle symbols, such that
  1746 a Unicode character is treated as a single symbol, not a sequence of
  1747 non-ASCII bytes as before.  Since Isabelle/ML string literals may
  1748 contain symbols without further backslash escapes, Unicode can now be
  1749 used here as well.  Recall that Symbol.explode in ML provides a
  1750 consistent view on symbols, while raw explode (or String.explode)
  1751 merely give a byte-oriented representation.
  1752 
  1753 * Theory loader: source files are primarily located via the master
  1754 directory of each theory node (where the .thy file itself resides).
  1755 The global load path is still partially available as legacy feature.
  1756 Minor INCOMPATIBILITY due to subtle change in file lookup: use
  1757 explicit paths, relatively to the theory.
  1758 
  1759 * Special treatment of ML file names has been discontinued.
  1760 Historically, optional extensions .ML or .sml were added on demand --
  1761 at the cost of clarity of file dependencies.  Recall that Isabelle/ML
  1762 files exclusively use the .ML extension.  Minor INCOMPATIBILTY.
  1763 
  1764 * Various options that affect pretty printing etc. are now properly
  1765 handled within the context via configuration options, instead of
  1766 unsynchronized references or print modes.  There are both ML Config.T
  1767 entities and Isar declaration attributes to access these.
  1768 
  1769   ML (Config.T)                 Isar (attribute)
  1770 
  1771   eta_contract                  eta_contract
  1772   show_brackets                 show_brackets
  1773   show_sorts                    show_sorts
  1774   show_types                    show_types
  1775   show_question_marks           show_question_marks
  1776   show_consts                   show_consts
  1777   show_abbrevs                  show_abbrevs
  1778 
  1779   Syntax.ast_trace              syntax_ast_trace
  1780   Syntax.ast_stat               syntax_ast_stat
  1781   Syntax.ambiguity_level        syntax_ambiguity_level
  1782 
  1783   Goal_Display.goals_limit      goals_limit
  1784   Goal_Display.show_main_goal   show_main_goal
  1785 
  1786   Method.rule_trace             rule_trace
  1787 
  1788   Thy_Output.display            thy_output_display
  1789   Thy_Output.quotes             thy_output_quotes
  1790   Thy_Output.indent             thy_output_indent
  1791   Thy_Output.source             thy_output_source
  1792   Thy_Output.break              thy_output_break
  1793 
  1794 Note that corresponding "..._default" references in ML may only be
  1795 changed globally at the ROOT session setup, but *not* within a theory.
  1796 The option "show_abbrevs" supersedes the former print mode
  1797 "no_abbrevs" with inverted meaning.
  1798 
  1799 * More systematic naming of some configuration options.
  1800 INCOMPATIBILITY.
  1801 
  1802   trace_simp  ~>  simp_trace
  1803   debug_simp  ~>  simp_debug
  1804 
  1805 * Support for real valued configuration options, using simplistic
  1806 floating-point notation that coincides with the inner syntax for
  1807 float_token.
  1808 
  1809 * Support for real valued preferences (with approximative PGIP type):
  1810 front-ends need to accept "pgint" values in float notation.
  1811 INCOMPATIBILITY.
  1812 
  1813 * The IsabelleText font now includes Cyrillic, Hebrew, Arabic from
  1814 DejaVu Sans.
  1815 
  1816 * Discontinued support for Poly/ML 5.0 and 5.1 versions.
  1817 
  1818 
  1819 *** Pure ***
  1820 
  1821 * Command 'type_synonym' (with single argument) replaces somewhat
  1822 outdated 'types', which is still available as legacy feature for some
  1823 time.
  1824 
  1825 * Command 'nonterminal' (with 'and' separated list of arguments)
  1826 replaces somewhat outdated 'nonterminals'.  INCOMPATIBILITY.
  1827 
  1828 * Command 'notepad' replaces former 'example_proof' for
  1829 experimentation in Isar without any result.  INCOMPATIBILITY.
  1830 
  1831 * Locale interpretation commands 'interpret' and 'sublocale' accept
  1832 lists of equations to map definitions in a locale to appropriate
  1833 entities in the context of the interpretation.  The 'interpretation'
  1834 command already provided this functionality.
  1835 
  1836 * Diagnostic command 'print_dependencies' prints the locale instances
  1837 that would be activated if the specified expression was interpreted in
  1838 the current context.  Variant "print_dependencies!" assumes a context
  1839 without interpretations.
  1840 
  1841 * Diagnostic command 'print_interps' prints interpretations in proofs
  1842 in addition to interpretations in theories.
  1843 
  1844 * Discontinued obsolete 'global' and 'local' commands to manipulate
  1845 the theory name space.  Rare INCOMPATIBILITY.  The ML functions
  1846 Sign.root_path and Sign.local_path may be applied directly where this
  1847 feature is still required for historical reasons.
  1848 
  1849 * Discontinued obsolete 'constdefs' command.  INCOMPATIBILITY, use
  1850 'definition' instead.
  1851 
  1852 * The "prems" fact, which refers to the accidental collection of
  1853 foundational premises in the context, is now explicitly marked as
  1854 legacy feature and will be discontinued soon.  Consider using "assms"
  1855 of the head statement or reference facts by explicit names.
  1856 
  1857 * Document antiquotations @{class} and @{type} print classes and type
  1858 constructors.
  1859 
  1860 * Document antiquotation @{file} checks file/directory entries within
  1861 the local file system.
  1862 
  1863 
  1864 *** HOL ***
  1865 
  1866 * Coercive subtyping: functions can be declared as coercions and type
  1867 inference will add them as necessary upon input of a term.  Theory
  1868 Complex_Main declares real :: nat => real and real :: int => real as
  1869 coercions. A coercion function f is declared like this:
  1870 
  1871   declare [[coercion f]]
  1872 
  1873 To lift coercions through type constructors (e.g. from nat => real to
  1874 nat list => real list), map functions can be declared, e.g.
  1875 
  1876   declare [[coercion_map map]]
  1877 
  1878 Currently coercion inference is activated only in theories including
  1879 real numbers, i.e. descendants of Complex_Main.  This is controlled by
  1880 the configuration option "coercion_enabled", e.g. it can be enabled in
  1881 other theories like this:
  1882 
  1883   declare [[coercion_enabled]]
  1884 
  1885 * Command 'partial_function' provides basic support for recursive
  1886 function definitions over complete partial orders.  Concrete instances
  1887 are provided for i) the option type, ii) tail recursion on arbitrary
  1888 types, and iii) the heap monad of Imperative_HOL.  See
  1889 src/HOL/ex/Fundefs.thy and src/HOL/Imperative_HOL/ex/Linked_Lists.thy
  1890 for examples.
  1891 
  1892 * Function package: f.psimps rules are no longer implicitly declared
  1893 as [simp].  INCOMPATIBILITY.
  1894 
  1895 * Datatype package: theorems generated for executable equality (class
  1896 "eq") carry proper names and are treated as default code equations.
  1897 
  1898 * Inductive package: now offers command 'inductive_simps' to
  1899 automatically derive instantiated and simplified equations for
  1900 inductive predicates, similar to 'inductive_cases'.
  1901 
  1902 * Command 'enriched_type' allows to register properties of the
  1903 functorial structure of types.
  1904 
  1905 * Improved infrastructure for term evaluation using code generator
  1906 techniques, in particular static evaluation conversions.
  1907 
  1908 * Code generator: Scala (2.8 or higher) has been added to the target
  1909 languages.
  1910 
  1911 * Code generator: globbing constant expressions "*" and "Theory.*"
  1912 have been replaced by the more idiomatic "_" and "Theory._".
  1913 INCOMPATIBILITY.
  1914 
  1915 * Code generator: export_code without explicit file declaration prints
  1916 to standard output.  INCOMPATIBILITY.
  1917 
  1918 * Code generator: do not print function definitions for case
  1919 combinators any longer.
  1920 
  1921 * Code generator: simplification with rules determined with
  1922 src/Tools/Code/code_simp.ML and method "code_simp".
  1923 
  1924 * Code generator for records: more idiomatic representation of record
  1925 types.  Warning: records are not covered by ancient SML code
  1926 generation any longer.  INCOMPATIBILITY.  In cases of need, a suitable
  1927 rep_datatype declaration helps to succeed then:
  1928 
  1929   record 'a foo = ...
  1930   ...
  1931   rep_datatype foo_ext ...
  1932 
  1933 * Records: logical foundation type for records does not carry a
  1934 '_type' suffix any longer (obsolete due to authentic syntax).
  1935 INCOMPATIBILITY.
  1936 
  1937 * Quickcheck now by default uses exhaustive testing instead of random
  1938 testing.  Random testing can be invoked by "quickcheck [random]",
  1939 exhaustive testing by "quickcheck [exhaustive]".
  1940 
  1941 * Quickcheck instantiates polymorphic types with small finite
  1942 datatypes by default. This enables a simple execution mechanism to
  1943 handle quantifiers and function equality over the finite datatypes.
  1944 
  1945 * Quickcheck random generator has been renamed from "code" to
  1946 "random".  INCOMPATIBILITY.
  1947 
  1948 * Quickcheck now has a configurable time limit which is set to 30
  1949 seconds by default. This can be changed by adding [timeout = n] to the
  1950 quickcheck command. The time limit for Auto Quickcheck is still set
  1951 independently.
  1952 
  1953 * Quickcheck in locales considers interpretations of that locale for
  1954 counter example search.
  1955 
  1956 * Sledgehammer:
  1957   - Added "smt" and "remote_smt" provers based on the "smt" proof
  1958     method. See the Sledgehammer manual for details ("isabelle doc
  1959     sledgehammer").
  1960   - Renamed commands:
  1961     sledgehammer atp_info ~> sledgehammer running_provers
  1962     sledgehammer atp_kill ~> sledgehammer kill_provers
  1963     sledgehammer available_atps ~> sledgehammer available_provers
  1964     INCOMPATIBILITY.
  1965   - Renamed options:
  1966     sledgehammer [atps = ...] ~> sledgehammer [provers = ...]
  1967     sledgehammer [atp = ...] ~> sledgehammer [prover = ...]
  1968     sledgehammer [timeout = 77 s] ~> sledgehammer [timeout = 77]
  1969     (and "ms" and "min" are no longer supported)
  1970     INCOMPATIBILITY.
  1971 
  1972 * Nitpick:
  1973   - Renamed options:
  1974     nitpick [timeout = 77 s] ~> nitpick [timeout = 77]
  1975     nitpick [tac_timeout = 777 ms] ~> nitpick [tac_timeout = 0.777]
  1976     INCOMPATIBILITY.
  1977   - Added support for partial quotient types.
  1978   - Added local versions of the "Nitpick.register_xxx" functions.
  1979   - Added "whack" option.
  1980   - Allow registration of quotient types as codatatypes.
  1981   - Improved "merge_type_vars" option to merge more types.
  1982   - Removed unsound "fast_descrs" option.
  1983   - Added custom symmetry breaking for datatypes, making it possible to reach
  1984     higher cardinalities.
  1985   - Prevent the expansion of too large definitions.
  1986 
  1987 * Proof methods "metis" and "meson" now have configuration options
  1988 "meson_trace", "metis_trace", and "metis_verbose" that can be enabled
  1989 to diagnose these tools. E.g.
  1990 
  1991     using [[metis_trace = true]]
  1992 
  1993 * Auto Solve: Renamed "Auto Solve Direct".  The tool is now available
  1994 manually as command 'solve_direct'.
  1995 
  1996 * The default SMT solver Z3 must be enabled explicitly (due to
  1997 licensing issues) by setting the environment variable
  1998 Z3_NON_COMMERCIAL in etc/settings of the component, for example.  For
  1999 commercial applications, the SMT solver CVC3 is provided as fall-back;
  2000 changing the SMT solver is done via the configuration option
  2001 "smt_solver".
  2002 
  2003 * Remote SMT solvers need to be referred to by the "remote_" prefix,
  2004 i.e. "remote_cvc3" and "remote_z3".
  2005 
  2006 * Added basic SMT support for datatypes, records, and typedefs using
  2007 the oracle mode (no proofs).  Direct support of pairs has been dropped
  2008 in exchange (pass theorems fst_conv snd_conv pair_collapse to the SMT
  2009 support for a similar behavior).  Minor INCOMPATIBILITY.
  2010 
  2011 * Changed SMT configuration options:
  2012   - Renamed:
  2013     z3_proofs ~> smt_oracle (with inverted meaning)
  2014     z3_trace_assms ~> smt_trace_used_facts
  2015     INCOMPATIBILITY.
  2016   - Added:
  2017     smt_verbose
  2018     smt_random_seed
  2019     smt_datatypes
  2020     smt_infer_triggers
  2021     smt_monomorph_limit
  2022     cvc3_options
  2023     remote_cvc3_options
  2024     remote_z3_options
  2025     yices_options
  2026 
  2027 * Boogie output files (.b2i files) need to be declared in the theory
  2028 header.
  2029 
  2030 * Simplification procedure "list_to_set_comprehension" rewrites list
  2031 comprehensions applied to List.set to set comprehensions.  Occasional
  2032 INCOMPATIBILITY, may be deactivated like this:
  2033 
  2034   declare [[simproc del: list_to_set_comprehension]]
  2035 
  2036 * Removed old version of primrec package.  INCOMPATIBILITY.
  2037 
  2038 * Removed simplifier congruence rule of "prod_case", as has for long
  2039 been the case with "split".  INCOMPATIBILITY.
  2040 
  2041 * String.literal is a type, but not a datatype.  INCOMPATIBILITY.
  2042 
  2043 * Removed [split_format ... and ... and ...] version of
  2044 [split_format].  Potential INCOMPATIBILITY.
  2045 
  2046 * Predicate "sorted" now defined inductively, with nice induction
  2047 rules.  INCOMPATIBILITY: former sorted.simps now named sorted_simps.
  2048 
  2049 * Constant "contents" renamed to "the_elem", to free the generic name
  2050 contents for other uses.  INCOMPATIBILITY.
  2051 
  2052 * Renamed class eq and constant eq (for code generation) to class
  2053 equal and constant equal, plus renaming of related facts and various
  2054 tuning.  INCOMPATIBILITY.
  2055 
  2056 * Dropped type classes mult_mono and mult_mono1.  INCOMPATIBILITY.
  2057 
  2058 * Removed output syntax "'a ~=> 'b" for "'a => 'b option".
  2059 INCOMPATIBILITY.
  2060 
  2061 * Renamed theory Fset to Cset, type Fset.fset to Cset.set, in order to
  2062 avoid confusion with finite sets.  INCOMPATIBILITY.
  2063 
  2064 * Abandoned locales equiv, congruent and congruent2 for equivalence
  2065 relations.  INCOMPATIBILITY: use equivI rather than equiv_intro (same
  2066 for congruent(2)).
  2067 
  2068 * Some previously unqualified names have been qualified:
  2069 
  2070   types
  2071     bool ~> HOL.bool
  2072     nat ~> Nat.nat
  2073 
  2074   constants
  2075     Trueprop ~> HOL.Trueprop
  2076     True ~> HOL.True
  2077     False ~> HOL.False
  2078     op & ~> HOL.conj
  2079     op | ~> HOL.disj
  2080     op --> ~> HOL.implies
  2081     op = ~> HOL.eq
  2082     Not ~> HOL.Not
  2083     The ~> HOL.The
  2084     All ~> HOL.All
  2085     Ex ~> HOL.Ex
  2086     Ex1 ~> HOL.Ex1
  2087     Let ~> HOL.Let
  2088     If ~> HOL.If
  2089     Ball ~> Set.Ball
  2090     Bex ~> Set.Bex
  2091     Suc ~> Nat.Suc
  2092     Pair ~> Product_Type.Pair
  2093     fst ~> Product_Type.fst
  2094     snd ~> Product_Type.snd
  2095     curry ~> Product_Type.curry
  2096     op : ~> Set.member
  2097     Collect ~> Set.Collect
  2098 
  2099 INCOMPATIBILITY.
  2100 
  2101 * More canonical naming convention for some fundamental definitions:
  2102 
  2103     bot_bool_eq ~> bot_bool_def
  2104     top_bool_eq ~> top_bool_def
  2105     inf_bool_eq ~> inf_bool_def
  2106     sup_bool_eq ~> sup_bool_def
  2107     bot_fun_eq  ~> bot_fun_def
  2108     top_fun_eq  ~> top_fun_def
  2109     inf_fun_eq  ~> inf_fun_def
  2110     sup_fun_eq  ~> sup_fun_def
  2111 
  2112 INCOMPATIBILITY.
  2113 
  2114 * More stylized fact names:
  2115 
  2116   expand_fun_eq ~> fun_eq_iff
  2117   expand_set_eq ~> set_eq_iff
  2118   set_ext       ~> set_eqI
  2119   nat_number    ~> eval_nat_numeral
  2120 
  2121 INCOMPATIBILITY.
  2122 
  2123 * Refactoring of code-generation specific operations in theory List:
  2124 
  2125   constants
  2126     null ~> List.null
  2127 
  2128   facts
  2129     mem_iff ~> member_def
  2130     null_empty ~> null_def
  2131 
  2132 INCOMPATIBILITY.  Note that these were not supposed to be used
  2133 regularly unless for striking reasons; their main purpose was code
  2134 generation.
  2135 
  2136 Various operations from the Haskell prelude are used for generating
  2137 Haskell code.
  2138 
  2139 * Term "bij f" is now an abbreviation of "bij_betw f UNIV UNIV".  Term
  2140 "surj f" is now an abbreviation of "range f = UNIV".  The theorems
  2141 bij_def and surj_def are unchanged.  INCOMPATIBILITY.
  2142 
  2143 * Abolished some non-alphabetic type names: "prod" and "sum" replace
  2144 "*" and "+" respectively.  INCOMPATIBILITY.
  2145 
  2146 * Name "Plus" of disjoint sum operator "<+>" is now hidden.  Write
  2147 "Sum_Type.Plus" instead.
  2148 
  2149 * Constant "split" has been merged with constant "prod_case"; names of
  2150 ML functions, facts etc. involving split have been retained so far,
  2151 though.  INCOMPATIBILITY.
  2152 
  2153 * Dropped old infix syntax "_ mem _" for List.member; use "_ : set _"
  2154 instead.  INCOMPATIBILITY.
  2155 
  2156 * Removed lemma "Option.is_none_none" which duplicates "is_none_def".
  2157 INCOMPATIBILITY.
  2158 
  2159 * Former theory Library/Enum is now part of the HOL-Main image.
  2160 INCOMPATIBILITY: all constants of the Enum theory now have to be
  2161 referred to by its qualified name.
  2162 
  2163   enum    ~>  Enum.enum
  2164   nlists  ~>  Enum.nlists
  2165   product ~>  Enum.product
  2166 
  2167 * Theory Library/Monad_Syntax provides do-syntax for monad types.
  2168 Syntax in Library/State_Monad has been changed to avoid ambiguities.
  2169 INCOMPATIBILITY.
  2170 
  2171 * Theory Library/SetsAndFunctions has been split into
  2172 Library/Function_Algebras and Library/Set_Algebras; canonical names
  2173 for instance definitions for functions; various improvements.
  2174 INCOMPATIBILITY.
  2175 
  2176 * Theory Library/Multiset provides stable quicksort implementation of
  2177 sort_key.
  2178 
  2179 * Theory Library/Multiset: renamed empty_idemp ~> empty_neutral.
  2180 INCOMPATIBILITY.
  2181 
  2182 * Session Multivariate_Analysis: introduced a type class for euclidean
  2183 space.  Most theorems are now stated in terms of euclidean spaces
  2184 instead of finite cartesian products.
  2185 
  2186   types
  2187     real ^ 'n ~>  'a::real_vector
  2188               ~>  'a::euclidean_space
  2189               ~>  'a::ordered_euclidean_space
  2190         (depends on your needs)
  2191 
  2192   constants
  2193      _ $ _        ~> _ $$ _
  2194      \<chi> x. _  ~> \<chi>\<chi> x. _
  2195      CARD('n)     ~> DIM('a)
  2196 
  2197 Also note that the indices are now natural numbers and not from some
  2198 finite type. Finite cartesian products of euclidean spaces, products
  2199 of euclidean spaces the real and complex numbers are instantiated to
  2200 be euclidean_spaces.  INCOMPATIBILITY.
  2201 
  2202 * Session Probability: introduced pextreal as positive extended real
  2203 numbers.  Use pextreal as value for measures.  Introduce the
  2204 Radon-Nikodym derivative, product spaces and Fubini's theorem for
  2205 arbitrary sigma finite measures.  Introduces Lebesgue measure based on
  2206 the integral in Multivariate Analysis.  INCOMPATIBILITY.
  2207 
  2208 * Session Imperative_HOL: revamped, corrected dozens of inadequacies.
  2209 INCOMPATIBILITY.
  2210 
  2211 * Session SPARK (with image HOL-SPARK) provides commands to load and
  2212 prove verification conditions generated by the SPARK Ada program
  2213 verifier.  See also src/HOL/SPARK and src/HOL/SPARK/Examples.
  2214 
  2215 
  2216 *** HOL-Algebra ***
  2217 
  2218 * Theorems for additive ring operations (locale abelian_monoid and
  2219 descendants) are generated by interpretation from their multiplicative
  2220 counterparts.  Names (in particular theorem names) have the mandatory
  2221 qualifier 'add'.  Previous theorem names are redeclared for
  2222 compatibility.
  2223 
  2224 * Structure "int_ring" is now an abbreviation (previously a
  2225 definition).  This fits more natural with advanced interpretations.
  2226 
  2227 
  2228 *** HOLCF ***
  2229 
  2230 * The domain package now runs in definitional mode by default: The
  2231 former command 'new_domain' is now called 'domain'.  To use the domain
  2232 package in its original axiomatic mode, use 'domain (unsafe)'.
  2233 INCOMPATIBILITY.
  2234 
  2235 * The new class "domain" is now the default sort.  Class "predomain"
  2236 is an unpointed version of "domain". Theories can be updated by
  2237 replacing sort annotations as shown below.  INCOMPATIBILITY.
  2238 
  2239   'a::type ~> 'a::countable
  2240   'a::cpo  ~> 'a::predomain
  2241   'a::pcpo ~> 'a::domain
  2242 
  2243 * The old type class "rep" has been superseded by class "domain".
  2244 Accordingly, users of the definitional package must remove any
  2245 "default_sort rep" declarations.  INCOMPATIBILITY.
  2246 
  2247 * The domain package (definitional mode) now supports unpointed
  2248 predomain argument types, as long as they are marked 'lazy'. (Strict
  2249 arguments must be in class "domain".) For example, the following
  2250 domain definition now works:
  2251 
  2252   domain natlist = nil | cons (lazy "nat discr") (lazy "natlist")
  2253 
  2254 * Theory HOLCF/Library/HOL_Cpo provides cpo and predomain class
  2255 instances for types from main HOL: bool, nat, int, char, 'a + 'b,
  2256 'a option, and 'a list.  Additionally, it configures fixrec and the
  2257 domain package to work with these types.  For example:
  2258 
  2259   fixrec isInl :: "('a + 'b) u -> tr"
  2260     where "isInl$(up$(Inl x)) = TT" | "isInl$(up$(Inr y)) = FF"
  2261 
  2262   domain V = VFun (lazy "V -> V") | VCon (lazy "nat") (lazy "V list")
  2263 
  2264 * The "(permissive)" option of fixrec has been replaced with a
  2265 per-equation "(unchecked)" option. See
  2266 src/HOL/HOLCF/Tutorial/Fixrec_ex.thy for examples. INCOMPATIBILITY.
  2267 
  2268 * The "bifinite" class no longer fixes a constant "approx"; the class
  2269 now just asserts that such a function exists.  INCOMPATIBILITY.
  2270 
  2271 * Former type "alg_defl" has been renamed to "defl".  HOLCF no longer
  2272 defines an embedding of type 'a defl into udom by default; instances
  2273 of "bifinite" and "domain" classes are available in
  2274 src/HOL/HOLCF/Library/Defl_Bifinite.thy.
  2275 
  2276 * The syntax "REP('a)" has been replaced with "DEFL('a)".
  2277 
  2278 * The predicate "directed" has been removed.  INCOMPATIBILITY.
  2279 
  2280 * The type class "finite_po" has been removed.  INCOMPATIBILITY.
  2281 
  2282 * The function "cprod_map" has been renamed to "prod_map".
  2283 INCOMPATIBILITY.
  2284 
  2285 * The monadic bind operator on each powerdomain has new binder syntax
  2286 similar to sets, e.g. "\<Union>\<sharp>x\<in>xs. t" represents
  2287 "upper_bind\<cdot>xs\<cdot>(\<Lambda> x. t)".
  2288 
  2289 * The infix syntax for binary union on each powerdomain has changed
  2290 from e.g. "+\<sharp>" to "\<union>\<sharp>", for consistency with set
  2291 syntax.  INCOMPATIBILITY.
  2292 
  2293 * The constant "UU" has been renamed to "bottom".  The syntax "UU" is
  2294 still supported as an input translation.
  2295 
  2296 * Renamed some theorems (the original names are also still available).
  2297 
  2298   expand_fun_below   ~> fun_below_iff
  2299   below_fun_ext      ~> fun_belowI
  2300   expand_cfun_eq     ~> cfun_eq_iff
  2301   ext_cfun           ~> cfun_eqI
  2302   expand_cfun_below  ~> cfun_below_iff
  2303   below_cfun_ext     ~> cfun_belowI
  2304   cont2cont_Rep_CFun ~> cont2cont_APP
  2305 
  2306 * The Abs and Rep functions for various types have changed names.
  2307 Related theorem names have also changed to match. INCOMPATIBILITY.
  2308 
  2309   Rep_CFun  ~> Rep_cfun
  2310   Abs_CFun  ~> Abs_cfun
  2311   Rep_Sprod ~> Rep_sprod
  2312   Abs_Sprod ~> Abs_sprod
  2313   Rep_Ssum  ~> Rep_ssum
  2314   Abs_Ssum  ~> Abs_ssum
  2315 
  2316 * Lemmas with names of the form *_defined_iff or *_strict_iff have
  2317 been renamed to *_bottom_iff.  INCOMPATIBILITY.
  2318 
  2319 * Various changes to bisimulation/coinduction with domain package:
  2320 
  2321   - Definitions of "bisim" constants no longer mention definedness.
  2322   - With mutual recursion, "bisim" predicate is now curried.
  2323   - With mutual recursion, each type gets a separate coind theorem.
  2324   - Variable names in bisim_def and coinduct rules have changed.
  2325 
  2326 INCOMPATIBILITY.
  2327 
  2328 * Case combinators generated by the domain package for type "foo" are
  2329 now named "foo_case" instead of "foo_when".  INCOMPATIBILITY.
  2330 
  2331 * Several theorems have been renamed to more accurately reflect the
  2332 names of constants and types involved.  INCOMPATIBILITY.
  2333 
  2334   thelub_const    ~> lub_const
  2335   lub_const       ~> is_lub_const
  2336   thelubI         ~> lub_eqI
  2337   is_lub_lub      ~> is_lubD2
  2338   lubI            ~> is_lub_lub
  2339   unique_lub      ~> is_lub_unique
  2340   is_ub_lub       ~> is_lub_rangeD1
  2341   lub_bin_chain   ~> is_lub_bin_chain
  2342   lub_fun         ~> is_lub_fun
  2343   thelub_fun      ~> lub_fun
  2344   thelub_cfun     ~> lub_cfun
  2345   thelub_Pair     ~> lub_Pair
  2346   lub_cprod       ~> is_lub_prod
  2347   thelub_cprod    ~> lub_prod
  2348   minimal_cprod   ~> minimal_prod
  2349   inst_cprod_pcpo ~> inst_prod_pcpo
  2350   UU_I            ~> bottomI
  2351   compact_UU      ~> compact_bottom
  2352   deflation_UU    ~> deflation_bottom
  2353   finite_deflation_UU ~> finite_deflation_bottom
  2354 
  2355 * Many legacy theorem names have been discontinued.  INCOMPATIBILITY.
  2356 
  2357   sq_ord_less_eq_trans ~> below_eq_trans
  2358   sq_ord_eq_less_trans ~> eq_below_trans
  2359   refl_less            ~> below_refl
  2360   trans_less           ~> below_trans
  2361   antisym_less         ~> below_antisym
  2362   antisym_less_inverse ~> po_eq_conv [THEN iffD1]
  2363   box_less             ~> box_below
  2364   rev_trans_less       ~> rev_below_trans
  2365   not_less2not_eq      ~> not_below2not_eq
  2366   less_UU_iff          ~> below_UU_iff
  2367   flat_less_iff        ~> flat_below_iff
  2368   adm_less             ~> adm_below
  2369   adm_not_less         ~> adm_not_below
  2370   adm_compact_not_less ~> adm_compact_not_below
  2371   less_fun_def         ~> below_fun_def
  2372   expand_fun_less      ~> fun_below_iff
  2373   less_fun_ext         ~> fun_belowI
  2374   less_discr_def       ~> below_discr_def
  2375   discr_less_eq        ~> discr_below_eq
  2376   less_unit_def        ~> below_unit_def
  2377   less_cprod_def       ~> below_prod_def
  2378   prod_lessI           ~> prod_belowI
  2379   Pair_less_iff        ~> Pair_below_iff
  2380   fst_less_iff         ~> fst_below_iff
  2381   snd_less_iff         ~> snd_below_iff
  2382   expand_cfun_less     ~> cfun_below_iff
  2383   less_cfun_ext        ~> cfun_belowI
  2384   injection_less       ~> injection_below
  2385   less_up_def          ~> below_up_def
  2386   not_Iup_less         ~> not_Iup_below
  2387   Iup_less             ~> Iup_below
  2388   up_less              ~> up_below
  2389   Def_inject_less_eq   ~> Def_below_Def
  2390   Def_less_is_eq       ~> Def_below_iff
  2391   spair_less_iff       ~> spair_below_iff
  2392   less_sprod           ~> below_sprod
  2393   spair_less           ~> spair_below
  2394   sfst_less_iff        ~> sfst_below_iff
  2395   ssnd_less_iff        ~> ssnd_below_iff
  2396   fix_least_less       ~> fix_least_below
  2397   dist_less_one        ~> dist_below_one
  2398   less_ONE             ~> below_ONE
  2399   ONE_less_iff         ~> ONE_below_iff
  2400   less_sinlD           ~> below_sinlD
  2401   less_sinrD           ~> below_sinrD
  2402 
  2403 
  2404 *** FOL and ZF ***
  2405 
  2406 * All constant names are now qualified internally and use proper
  2407 identifiers, e.g. "IFOL.eq" instead of "op =".  INCOMPATIBILITY.
  2408 
  2409 
  2410 *** ML ***
  2411 
  2412 * Antiquotation @{assert} inlines a function bool -> unit that raises
  2413 Fail if the argument is false.  Due to inlining the source position of
  2414 failed assertions is included in the error output.
  2415 
  2416 * Discontinued antiquotation @{theory_ref}, which is obsolete since ML
  2417 text is in practice always evaluated with a stable theory checkpoint.
  2418 Minor INCOMPATIBILITY, use (Theory.check_thy @{theory}) instead.
  2419 
  2420 * Antiquotation @{theory A} refers to theory A from the ancestry of
  2421 the current context, not any accidental theory loader state as before.
  2422 Potential INCOMPATIBILITY, subtle change in semantics.
  2423 
  2424 * Syntax.pretty_priority (default 0) configures the required priority
  2425 of pretty-printed output and thus affects insertion of parentheses.
  2426 
  2427 * Syntax.default_root (default "any") configures the inner syntax
  2428 category (nonterminal symbol) for parsing of terms.
  2429 
  2430 * Former exception Library.UnequalLengths now coincides with
  2431 ListPair.UnequalLengths.
  2432 
  2433 * Renamed structure MetaSimplifier to Raw_Simplifier.  Note that the
  2434 main functionality is provided by structure Simplifier.
  2435 
  2436 * Renamed raw "explode" function to "raw_explode" to emphasize its
  2437 meaning.  Note that internally to Isabelle, Symbol.explode is used in
  2438 almost all situations.
  2439 
  2440 * Discontinued obsolete function sys_error and exception SYS_ERROR.
  2441 See implementation manual for further details on exceptions in
  2442 Isabelle/ML.
  2443 
  2444 * Renamed setmp_noncritical to Unsynchronized.setmp to emphasize its
  2445 meaning.
  2446 
  2447 * Renamed structure PureThy to Pure_Thy and moved most of its
  2448 operations to structure Global_Theory, to emphasize that this is
  2449 rarely-used global-only stuff.
  2450 
  2451 * Discontinued Output.debug.  Minor INCOMPATIBILITY, use plain writeln
  2452 instead (or tracing for high-volume output).
  2453 
  2454 * Configuration option show_question_marks only affects regular pretty
  2455 printing of types and terms, not raw Term.string_of_vname.
  2456 
  2457 * ML_Context.thm and ML_Context.thms are no longer pervasive.  Rare
  2458 INCOMPATIBILITY, superseded by static antiquotations @{thm} and
  2459 @{thms} for most purposes.
  2460 
  2461 * ML structure Unsynchronized is never opened, not even in Isar
  2462 interaction mode as before.  Old Unsynchronized.set etc. have been
  2463 discontinued -- use plain := instead.  This should be *rare* anyway,
  2464 since modern tools always work via official context data, notably
  2465 configuration options.
  2466 
  2467 * Parallel and asynchronous execution requires special care concerning
  2468 interrupts.  Structure Exn provides some convenience functions that
  2469 avoid working directly with raw Interrupt.  User code must not absorb
  2470 interrupts -- intermediate handling (for cleanup etc.) needs to be
  2471 followed by re-raising of the original exception.  Another common
  2472 source of mistakes are "handle _" patterns, which make the meaning of
  2473 the program subject to physical effects of the environment.
  2474 
  2475 
  2476 
  2477 New in Isabelle2009-2 (June 2010)
  2478 ---------------------------------
  2479 
  2480 *** General ***
  2481 
  2482 * Authentic syntax for *all* logical entities (type classes, type
  2483 constructors, term constants): provides simple and robust
  2484 correspondence between formal entities and concrete syntax.  Within
  2485 the parse tree / AST representations, "constants" are decorated by
  2486 their category (class, type, const) and spelled out explicitly with
  2487 their full internal name.
  2488 
  2489 Substantial INCOMPATIBILITY concerning low-level syntax declarations
  2490 and translations (translation rules and translation functions in ML).
  2491 Some hints on upgrading:
  2492 
  2493   - Many existing uses of 'syntax' and 'translations' can be replaced
  2494     by more modern 'type_notation', 'notation' and 'abbreviation',
  2495     which are independent of this issue.
  2496 
  2497   - 'translations' require markup within the AST; the term syntax
  2498     provides the following special forms:
  2499 
  2500       CONST c   -- produces syntax version of constant c from context
  2501       XCONST c  -- literally c, checked as constant from context
  2502       c         -- literally c, if declared by 'syntax'
  2503 
  2504     Plain identifiers are treated as AST variables -- occasionally the
  2505     system indicates accidental variables via the error "rhs contains
  2506     extra variables".
  2507 
  2508     Type classes and type constructors are marked according to their
  2509     concrete syntax.  Some old translations rules need to be written
  2510     for the "type" category, using type constructor application
  2511     instead of pseudo-term application of the default category
  2512     "logic".
  2513 
  2514   - 'parse_translation' etc. in ML may use the following
  2515     antiquotations:
  2516 
  2517       @{class_syntax c}   -- type class c within parse tree / AST
  2518       @{term_syntax c}    -- type constructor c within parse tree / AST
  2519       @{const_syntax c}   -- ML version of "CONST c" above
  2520       @{syntax_const c}   -- literally c (checked wrt. 'syntax' declarations)
  2521 
  2522   - Literal types within 'typed_print_translations', i.e. those *not*
  2523     represented as pseudo-terms are represented verbatim.  Use @{class
  2524     c} or @{type_name c} here instead of the above syntax
  2525     antiquotations.
  2526 
  2527 Note that old non-authentic syntax was based on unqualified base
  2528 names, so all of the above "constant" names would coincide.  Recall
  2529 that 'print_syntax' and ML_command "set Syntax.trace_ast" help to
  2530 diagnose syntax problems.
  2531 
  2532 * Type constructors admit general mixfix syntax, not just infix.
  2533 
  2534 * Concrete syntax may be attached to local entities without a proof
  2535 body, too.  This works via regular mixfix annotations for 'fix',
  2536 'def', 'obtain' etc. or via the explicit 'write' command, which is
  2537 similar to the 'notation' command in theory specifications.
  2538 
  2539 * Discontinued unnamed infix syntax (legacy feature for many years) --
  2540 need to specify constant name and syntax separately.  Internal ML
  2541 datatype constructors have been renamed from InfixName to Infix etc.
  2542 Minor INCOMPATIBILITY.
  2543 
  2544 * Schematic theorem statements need to be explicitly markup as such,
  2545 via commands 'schematic_lemma', 'schematic_theorem',
  2546 'schematic_corollary'.  Thus the relevance of the proof is made
  2547 syntactically clear, which impacts performance in a parallel or
  2548 asynchronous interactive environment.  Minor INCOMPATIBILITY.
  2549 
  2550 * Use of cumulative prems via "!" in some proof methods has been
  2551 discontinued (old legacy feature).
  2552 
  2553 * References 'trace_simp' and 'debug_simp' have been replaced by
  2554 configuration options stored in the context. Enabling tracing (the
  2555 case of debugging is similar) in proofs works via
  2556 
  2557   using [[trace_simp = true]]
  2558 
  2559 Tracing is then active for all invocations of the simplifier in
  2560 subsequent goal refinement steps. Tracing may also still be enabled or
  2561 disabled via the ProofGeneral settings menu.
  2562 
  2563 * Separate commands 'hide_class', 'hide_type', 'hide_const',
  2564 'hide_fact' replace the former 'hide' KIND command.  Minor
  2565 INCOMPATIBILITY.
  2566 
  2567 * Improved parallelism of proof term normalization: usedir -p2 -q0 is
  2568 more efficient than combinations with -q1 or -q2.
  2569 
  2570 
  2571 *** Pure ***
  2572 
  2573 * Proofterms record type-class reasoning explicitly, using the
  2574 "unconstrain" operation internally.  This eliminates all sort
  2575 constraints from a theorem and proof, introducing explicit
  2576 OFCLASS-premises.  On the proof term level, this operation is
  2577 automatically applied at theorem boundaries, such that closed proofs
  2578 are always free of sort constraints.  INCOMPATIBILITY for tools that
  2579 inspect proof terms.
  2580 
  2581 * Local theory specifications may depend on extra type variables that
  2582 are not present in the result type -- arguments TYPE('a) :: 'a itself
  2583 are added internally.  For example:
  2584 
  2585   definition unitary :: bool where "unitary = (ALL (x::'a) y. x = y)"
  2586 
  2587 * Predicates of locales introduced by classes carry a mandatory
  2588 "class" prefix.  INCOMPATIBILITY.
  2589 
  2590 * Vacuous class specifications observe default sort.  INCOMPATIBILITY.
  2591 
  2592 * Old 'axclass' command has been discontinued.  INCOMPATIBILITY, use
  2593 'class' instead.
  2594 
  2595 * Command 'code_reflect' allows to incorporate generated ML code into
  2596 runtime environment; replaces immature code_datatype antiquotation.
  2597 INCOMPATIBILITY.
  2598 
  2599 * Code generator: simple concept for abstract datatypes obeying
  2600 invariants.
  2601 
  2602 * Code generator: details of internal data cache have no impact on the
  2603 user space functionality any longer.
  2604 
  2605 * Methods "unfold_locales" and "intro_locales" ignore non-locale
  2606 subgoals.  This is more appropriate for interpretations with 'where'.
  2607 INCOMPATIBILITY.
  2608 
  2609 * Command 'example_proof' opens an empty proof body.  This allows to
  2610 experiment with Isar, without producing any persistent result.
  2611 
  2612 * Commands 'type_notation' and 'no_type_notation' declare type syntax
  2613 within a local theory context, with explicit checking of the
  2614 constructors involved (in contrast to the raw 'syntax' versions).
  2615 
  2616 * Commands 'types' and 'typedecl' now work within a local theory
  2617 context -- without introducing dependencies on parameters or
  2618 assumptions, which is not possible in Isabelle/Pure.
  2619 
  2620 * Command 'defaultsort' has been renamed to 'default_sort', it works
  2621 within a local theory context.  Minor INCOMPATIBILITY.
  2622 
  2623 
  2624 *** HOL ***
  2625 
  2626 * Command 'typedef' now works within a local theory context -- without
  2627 introducing dependencies on parameters or assumptions, which is not
  2628 possible in Isabelle/Pure/HOL.  Note that the logical environment may
  2629 contain multiple interpretations of local typedefs (with different
  2630 non-emptiness proofs), even in a global theory context.
  2631 
  2632 * New package for quotient types.  Commands 'quotient_type' and
  2633 'quotient_definition' may be used for defining types and constants by
  2634 quotient constructions.  An example is the type of integers created by
  2635 quotienting pairs of natural numbers:
  2636 
  2637   fun
  2638     intrel :: "(nat * nat) => (nat * nat) => bool"
  2639   where
  2640     "intrel (x, y) (u, v) = (x + v = u + y)"
  2641 
  2642   quotient_type int = "nat * nat" / intrel
  2643     by (auto simp add: equivp_def expand_fun_eq)
  2644 
  2645   quotient_definition
  2646     "0::int" is "(0::nat, 0::nat)"
  2647 
  2648 The method "lifting" can be used to lift of theorems from the
  2649 underlying "raw" type to the quotient type.  The example
  2650 src/HOL/Quotient_Examples/FSet.thy includes such a quotient
  2651 construction and provides a reasoning infrastructure for finite sets.
  2652 
  2653 * Renamed Library/Quotient.thy to Library/Quotient_Type.thy to avoid
  2654 clash with new theory Quotient in Main HOL.
  2655 
  2656 * Moved the SMT binding into the main HOL session, eliminating
  2657 separate HOL-SMT session.
  2658 
  2659 * List membership infix mem operation is only an input abbreviation.
  2660 INCOMPATIBILITY.
  2661 
  2662 * Theory Library/Word.thy has been removed.  Use library Word/Word.thy
  2663 for future developements; former Library/Word.thy is still present in
  2664 the AFP entry RSAPPS.
  2665 
  2666 * Theorem Int.int_induct renamed to Int.int_of_nat_induct and is no
  2667 longer shadowed.  INCOMPATIBILITY.
  2668 
  2669 * Dropped theorem duplicate comp_arith; use semiring_norm instead.
  2670 INCOMPATIBILITY.
  2671 
  2672 * Dropped theorem RealPow.real_sq_order; use power2_le_imp_le instead.
  2673 INCOMPATIBILITY.
  2674 
  2675 * Dropped normalizing_semiring etc; use the facts in semiring classes
  2676 instead.  INCOMPATIBILITY.
  2677 
  2678 * Dropped several real-specific versions of lemmas about floor and
  2679 ceiling; use the generic lemmas from theory "Archimedean_Field"
  2680 instead.  INCOMPATIBILITY.
  2681 
  2682   floor_number_of_eq         ~> floor_number_of
  2683   le_floor_eq_number_of      ~> number_of_le_floor
  2684   le_floor_eq_zero           ~> zero_le_floor
  2685   le_floor_eq_one            ~> one_le_floor
  2686   floor_less_eq_number_of    ~> floor_less_number_of
  2687   floor_less_eq_zero         ~> floor_less_zero
  2688   floor_less_eq_one          ~> floor_less_one
  2689   less_floor_eq_number_of    ~> number_of_less_floor
  2690   less_floor_eq_zero         ~> zero_less_floor
  2691   less_floor_eq_one          ~> one_less_floor
  2692   floor_le_eq_number_of      ~> floor_le_number_of
  2693   floor_le_eq_zero           ~> floor_le_zero
  2694   floor_le_eq_one            ~> floor_le_one
  2695   floor_subtract_number_of   ~> floor_diff_number_of
  2696   floor_subtract_one         ~> floor_diff_one
  2697   ceiling_number_of_eq       ~> ceiling_number_of
  2698   ceiling_le_eq_number_of    ~> ceiling_le_number_of
  2699   ceiling_le_zero_eq         ~> ceiling_le_zero
  2700   ceiling_le_eq_one          ~> ceiling_le_one
  2701   less_ceiling_eq_number_of  ~> number_of_less_ceiling
  2702   less_ceiling_eq_zero       ~> zero_less_ceiling
  2703   less_ceiling_eq_one        ~> one_less_ceiling
  2704   ceiling_less_eq_number_of  ~> ceiling_less_number_of
  2705   ceiling_less_eq_zero       ~> ceiling_less_zero
  2706   ceiling_less_eq_one        ~> ceiling_less_one
  2707   le_ceiling_eq_number_of    ~> number_of_le_ceiling
  2708   le_ceiling_eq_zero         ~> zero_le_ceiling
  2709   le_ceiling_eq_one          ~> one_le_ceiling
  2710   ceiling_subtract_number_of ~> ceiling_diff_number_of
  2711   ceiling_subtract_one       ~> ceiling_diff_one
  2712 
  2713 * Theory "Finite_Set": various folding_XXX locales facilitate the
  2714 application of the various fold combinators on finite sets.
  2715 
  2716 * Library theory "RBT" renamed to "RBT_Impl"; new library theory "RBT"
  2717 provides abstract red-black tree type which is backed by "RBT_Impl" as
  2718 implementation.  INCOMPATIBILTY.
  2719 
  2720 * Theory Library/Coinductive_List has been removed -- superseded by
  2721 AFP/thys/Coinductive.
  2722 
  2723 * Theory PReal, including the type "preal" and related operations, has
  2724 been removed.  INCOMPATIBILITY.
  2725 
  2726 * Real: new development using Cauchy Sequences.
  2727 
  2728 * Split off theory "Big_Operators" containing setsum, setprod,
  2729 Inf_fin, Sup_fin, Min, Max from theory Finite_Set.  INCOMPATIBILITY.
  2730 
  2731 * Theory "Rational" renamed to "Rat", for consistency with "Nat",
  2732 "Int" etc.  INCOMPATIBILITY.
  2733 
  2734 * Constant Rat.normalize needs to be qualified.  INCOMPATIBILITY.
  2735 
  2736 * New set of rules "ac_simps" provides combined assoc / commute
  2737 rewrites for all interpretations of the appropriate generic locales.
  2738 
  2739 * Renamed theory "OrderedGroup" to "Groups" and split theory
  2740 "Ring_and_Field" into theories "Rings" and "Fields"; for more
  2741 appropriate and more consistent names suitable for name prefixes
  2742 within the HOL theories.  INCOMPATIBILITY.
  2743 
  2744 * Some generic constants have been put to appropriate theories:
  2745   - less_eq, less: Orderings
  2746   - zero, one, plus, minus, uminus, times, abs, sgn: Groups
  2747   - inverse, divide: Rings
  2748 INCOMPATIBILITY.
  2749 
  2750 * More consistent naming of type classes involving orderings (and
  2751 lattices):
  2752 
  2753     lower_semilattice                   ~> semilattice_inf
  2754     upper_semilattice                   ~> semilattice_sup
  2755 
  2756     dense_linear_order                  ~> dense_linorder
  2757 
  2758     pordered_ab_group_add               ~> ordered_ab_group_add
  2759     pordered_ab_group_add_abs           ~> ordered_ab_group_add_abs
  2760     pordered_ab_semigroup_add           ~> ordered_ab_semigroup_add
  2761     pordered_ab_semigroup_add_imp_le    ~> ordered_ab_semigroup_add_imp_le
  2762     pordered_cancel_ab_semigroup_add    ~> ordered_cancel_ab_semigroup_add
  2763     pordered_cancel_comm_semiring       ~> ordered_cancel_comm_semiring
  2764     pordered_cancel_semiring            ~> ordered_cancel_semiring
  2765     pordered_comm_monoid_add            ~> ordered_comm_monoid_add
  2766     pordered_comm_ring                  ~> ordered_comm_ring
  2767     pordered_comm_semiring              ~> ordered_comm_semiring
  2768     pordered_ring                       ~> ordered_ring
  2769     pordered_ring_abs                   ~> ordered_ring_abs
  2770     pordered_semiring                   ~> ordered_semiring
  2771 
  2772     ordered_ab_group_add                ~> linordered_ab_group_add
  2773     ordered_ab_semigroup_add            ~> linordered_ab_semigroup_add
  2774     ordered_cancel_ab_semigroup_add     ~> linordered_cancel_ab_semigroup_add
  2775     ordered_comm_semiring_strict        ~> linordered_comm_semiring_strict
  2776     ordered_field                       ~> linordered_field
  2777     ordered_field_no_lb                 ~> linordered_field_no_lb
  2778     ordered_field_no_ub                 ~> linordered_field_no_ub
  2779     ordered_field_dense_linear_order    ~> dense_linordered_field
  2780     ordered_idom                        ~> linordered_idom
  2781     ordered_ring                        ~> linordered_ring
  2782     ordered_ring_le_cancel_factor       ~> linordered_ring_le_cancel_factor
  2783     ordered_ring_less_cancel_factor     ~> linordered_ring_less_cancel_factor
  2784     ordered_ring_strict                 ~> linordered_ring_strict
  2785     ordered_semidom                     ~> linordered_semidom
  2786     ordered_semiring                    ~> linordered_semiring
  2787     ordered_semiring_1                  ~> linordered_semiring_1
  2788     ordered_semiring_1_strict           ~> linordered_semiring_1_strict
  2789     ordered_semiring_strict             ~> linordered_semiring_strict
  2790 
  2791   The following slightly odd type classes have been moved to a
  2792   separate theory Library/Lattice_Algebras:
  2793 
  2794     lordered_ab_group_add               ~> lattice_ab_group_add
  2795     lordered_ab_group_add_abs           ~> lattice_ab_group_add_abs
  2796     lordered_ab_group_add_meet          ~> semilattice_inf_ab_group_add
  2797     lordered_ab_group_add_join          ~> semilattice_sup_ab_group_add
  2798     lordered_ring                       ~> lattice_ring
  2799 
  2800 INCOMPATIBILITY.
  2801 
  2802 * Refined field classes:
  2803   - classes division_ring_inverse_zero, field_inverse_zero,
  2804     linordered_field_inverse_zero include rule inverse 0 = 0 --
  2805     subsumes former division_by_zero class;
  2806   - numerous lemmas have been ported from field to division_ring.
  2807 INCOMPATIBILITY.
  2808 
  2809 * Refined algebra theorem collections:
  2810   - dropped theorem group group_simps, use algebra_simps instead;
  2811   - dropped theorem group ring_simps, use field_simps instead;
  2812   - proper theorem collection field_simps subsumes former theorem
  2813     groups field_eq_simps and field_simps;
  2814   - dropped lemma eq_minus_self_iff which is a duplicate for
  2815     equal_neg_zero.
  2816 INCOMPATIBILITY.
  2817 
  2818 * Theory Finite_Set and List: some lemmas have been generalized from
  2819 sets to lattices:
  2820 
  2821   fun_left_comm_idem_inter      ~> fun_left_comm_idem_inf
  2822   fun_left_comm_idem_union      ~> fun_left_comm_idem_sup
  2823   inter_Inter_fold_inter        ~> inf_Inf_fold_inf
  2824   union_Union_fold_union        ~> sup_Sup_fold_sup
  2825   Inter_fold_inter              ~> Inf_fold_inf
  2826   Union_fold_union              ~> Sup_fold_sup
  2827   inter_INTER_fold_inter        ~> inf_INFI_fold_inf
  2828   union_UNION_fold_union        ~> sup_SUPR_fold_sup
  2829   INTER_fold_inter              ~> INFI_fold_inf
  2830   UNION_fold_union              ~> SUPR_fold_sup
  2831 
  2832 * Theory "Complete_Lattice": lemmas top_def and bot_def have been
  2833 replaced by the more convenient lemmas Inf_empty and Sup_empty.
  2834 Dropped lemmas Inf_insert_simp and Sup_insert_simp, which are subsumed
  2835 by Inf_insert and Sup_insert.  Lemmas Inf_UNIV and Sup_UNIV replace
  2836 former Inf_Univ and Sup_Univ.  Lemmas inf_top_right and sup_bot_right
  2837 subsume inf_top and sup_bot respectively.  INCOMPATIBILITY.
  2838 
  2839 * Reorganized theory Multiset: swapped notation of pointwise and
  2840 multiset order:
  2841 
  2842   - pointwise ordering is instance of class order with standard syntax
  2843     <= and <;
  2844   - multiset ordering has syntax <=# and <#; partial order properties
  2845     are provided by means of interpretation with prefix
  2846     multiset_order;
  2847   - less duplication, less historical organization of sections,
  2848     conversion from associations lists to multisets, rudimentary code
  2849     generation;
  2850   - use insert_DiffM2 [symmetric] instead of elem_imp_eq_diff_union,
  2851     if needed.
  2852 
  2853 Renamed:
  2854 
  2855   multiset_eq_conv_count_eq  ~>  multiset_ext_iff
  2856   multi_count_ext  ~>  multiset_ext
  2857   diff_union_inverse2  ~>  diff_union_cancelR
  2858 
  2859 INCOMPATIBILITY.
  2860 
  2861 * Theory Permutation: replaced local "remove" by List.remove1.
  2862 
  2863 * Code generation: ML and OCaml code is decorated with signatures.
  2864 
  2865 * Theory List: added transpose.
  2866 
  2867 * Library/Nat_Bijection.thy is a collection of bijective functions
  2868 between nat and other types, which supersedes the older libraries
  2869 Library/Nat_Int_Bij.thy and HOLCF/NatIso.thy.  INCOMPATIBILITY.
  2870 
  2871   Constants:
  2872   Nat_Int_Bij.nat2_to_nat         ~> prod_encode
  2873   Nat_Int_Bij.nat_to_nat2         ~> prod_decode
  2874   Nat_Int_Bij.int_to_nat_bij      ~> int_encode
  2875   Nat_Int_Bij.nat_to_int_bij      ~> int_decode
  2876   Countable.pair_encode           ~> prod_encode
  2877   NatIso.prod2nat                 ~> prod_encode
  2878   NatIso.nat2prod                 ~> prod_decode
  2879   NatIso.sum2nat                  ~> sum_encode
  2880   NatIso.nat2sum                  ~> sum_decode
  2881   NatIso.list2nat                 ~> list_encode
  2882   NatIso.nat2list                 ~> list_decode
  2883   NatIso.set2nat                  ~> set_encode
  2884   NatIso.nat2set                  ~> set_decode
  2885 
  2886   Lemmas:
  2887   Nat_Int_Bij.bij_nat_to_int_bij  ~> bij_int_decode
  2888   Nat_Int_Bij.nat2_to_nat_inj     ~> inj_prod_encode
  2889   Nat_Int_Bij.nat2_to_nat_surj    ~> surj_prod_encode
  2890   Nat_Int_Bij.nat_to_nat2_inj     ~> inj_prod_decode
  2891   Nat_Int_Bij.nat_to_nat2_surj    ~> surj_prod_decode
  2892   Nat_Int_Bij.i2n_n2i_id          ~> int_encode_inverse
  2893   Nat_Int_Bij.n2i_i2n_id          ~> int_decode_inverse
  2894   Nat_Int_Bij.surj_nat_to_int_bij ~> surj_int_encode
  2895   Nat_Int_Bij.surj_int_to_nat_bij ~> surj_int_decode
  2896   Nat_Int_Bij.inj_nat_to_int_bij  ~> inj_int_encode
  2897   Nat_Int_Bij.inj_int_to_nat_bij  ~> inj_int_decode
  2898   Nat_Int_Bij.bij_nat_to_int_bij  ~> bij_int_encode
  2899   Nat_Int_Bij.bij_int_to_nat_bij  ~> bij_int_decode
  2900 
  2901 * Sledgehammer:
  2902   - Renamed ATP commands:
  2903     atp_info     ~> sledgehammer running_atps
  2904     atp_kill     ~> sledgehammer kill_atps
  2905     atp_messages ~> sledgehammer messages
  2906     atp_minimize ~> sledgehammer minimize
  2907     print_atps   ~> sledgehammer available_atps
  2908     INCOMPATIBILITY.
  2909   - Added user's manual ("isabelle doc sledgehammer").
  2910   - Added option syntax and "sledgehammer_params" to customize
  2911     Sledgehammer's behavior.  See the manual for details.
  2912   - Modified the Isar proof reconstruction code so that it produces
  2913     direct proofs rather than proofs by contradiction.  (This feature
  2914     is still experimental.)
  2915   - Made Isar proof reconstruction work for SPASS, remote ATPs, and in
  2916     full-typed mode.
  2917   - Added support for TPTP syntax for SPASS via the "spass_tptp" ATP.
  2918 
  2919 * Nitpick:
  2920   - Added and implemented "binary_ints" and "bits" options.
  2921   - Added "std" option and implemented support for nonstandard models.
  2922   - Added and implemented "finitize" option to improve the precision
  2923     of infinite datatypes based on a monotonicity analysis.
  2924   - Added support for quotient types.
  2925   - Added support for "specification" and "ax_specification"
  2926     constructs.
  2927   - Added support for local definitions (for "function" and
  2928     "termination" proofs).
  2929   - Added support for term postprocessors.
  2930   - Optimized "Multiset.multiset" and "FinFun.finfun".
  2931   - Improved efficiency of "destroy_constrs" optimization.
  2932   - Fixed soundness bugs related to "destroy_constrs" optimization and
  2933     record getters.
  2934   - Fixed soundness bug related to higher-order constructors.
  2935   - Fixed soundness bug when "full_descrs" is enabled.
  2936   - Improved precision of set constructs.
  2937   - Added "atoms" option.
  2938   - Added cache to speed up repeated Kodkod invocations on the same
  2939     problems.
  2940   - Renamed "MiniSatJNI", "zChaffJNI", "BerkMinAlloy", and
  2941     "SAT4JLight" to "MiniSat_JNI", "zChaff_JNI", "BerkMin_Alloy", and
  2942     "SAT4J_Light".  INCOMPATIBILITY.
  2943   - Removed "skolemize", "uncurry", "sym_break", "flatten_prop",
  2944     "sharing_depth", and "show_skolems" options.  INCOMPATIBILITY.
  2945   - Removed "nitpick_intro" attribute.  INCOMPATIBILITY.
  2946 
  2947 * Method "induct" now takes instantiations of the form t, where t is not
  2948   a variable, as a shorthand for "x == t", where x is a fresh variable.
  2949   If this is not intended, t has to be enclosed in parentheses.
  2950   By default, the equalities generated by definitional instantiations
  2951   are pre-simplified, which may cause parameters of inductive cases
  2952   to disappear, or may even delete some of the inductive cases.
  2953   Use "induct (no_simp)" instead of "induct" to restore the old
  2954   behaviour. The (no_simp) option is also understood by the "cases"
  2955   and "nominal_induct" methods, which now perform pre-simplification, too.
  2956   INCOMPATIBILITY.
  2957 
  2958 
  2959 *** HOLCF ***
  2960 
  2961 * Variable names in lemmas generated by the domain package have
  2962 changed; the naming scheme is now consistent with the HOL datatype
  2963 package.  Some proof scripts may be affected, INCOMPATIBILITY.
  2964 
  2965 * The domain package no longer defines the function "foo_copy" for
  2966 recursive domain "foo".  The reach lemma is now stated directly in
  2967 terms of "foo_take".  Lemmas and proofs that mention "foo_copy" must
  2968 be reformulated in terms of "foo_take", INCOMPATIBILITY.
  2969 
  2970 * Most definedness lemmas generated by the domain package (previously
  2971 of the form "x ~= UU ==> foo$x ~= UU") now have an if-and-only-if form
  2972 like "foo$x = UU <-> x = UU", which works better as a simp rule.
  2973 Proofs that used definedness lemmas as intro rules may break,
  2974 potential INCOMPATIBILITY.
  2975 
  2976 * Induction and casedist rules generated by the domain package now
  2977 declare proper case_names (one called "bottom", and one named for each
  2978 constructor).  INCOMPATIBILITY.
  2979 
  2980 * For mutually-recursive domains, separate "reach" and "take_lemma"
  2981 rules are generated for each domain, INCOMPATIBILITY.
  2982 
  2983   foo_bar.reach       ~> foo.reach  bar.reach
  2984   foo_bar.take_lemmas ~> foo.take_lemma  bar.take_lemma
  2985 
  2986 * Some lemmas generated by the domain package have been renamed for
  2987 consistency with the datatype package, INCOMPATIBILITY.
  2988 
  2989   foo.ind        ~> foo.induct
  2990   foo.finite_ind ~> foo.finite_induct
  2991   foo.coind      ~> foo.coinduct
  2992   foo.casedist   ~> foo.exhaust
  2993   foo.exhaust    ~> foo.nchotomy
  2994 
  2995 * For consistency with other definition packages, the fixrec package
  2996 now generates qualified theorem names, INCOMPATIBILITY.
  2997 
  2998   foo_simps  ~> foo.simps
  2999   foo_unfold ~> foo.unfold
  3000   foo_induct ~> foo.induct
  3001 
  3002 * The "fixrec_simp" attribute has been removed.  The "fixrec_simp"
  3003 method and internal fixrec proofs now use the default simpset instead.
  3004 INCOMPATIBILITY.
  3005 
  3006 * The "contlub" predicate has been removed.  Proof scripts should use
  3007 lemma contI2 in place of monocontlub2cont, INCOMPATIBILITY.
  3008 
  3009 * The "admw" predicate has been removed, INCOMPATIBILITY.
  3010 
  3011 * The constants cpair, cfst, and csnd have been removed in favor of
  3012 Pair, fst, and snd from Isabelle/HOL, INCOMPATIBILITY.
  3013 
  3014 
  3015 *** ML ***
  3016 
  3017 * Antiquotations for basic formal entities:
  3018 
  3019     @{class NAME}         -- type class
  3020     @{class_syntax NAME}  -- syntax representation of the above
  3021 
  3022     @{type_name NAME}     -- logical type
  3023     @{type_abbrev NAME}   -- type abbreviation
  3024     @{nonterminal NAME}   -- type of concrete syntactic category
  3025     @{type_syntax NAME}   -- syntax representation of any of the above
  3026 
  3027     @{const_name NAME}    -- logical constant (INCOMPATIBILITY)
  3028     @{const_abbrev NAME}  -- abbreviated constant
  3029     @{const_syntax NAME}  -- syntax representation of any of the above
  3030 
  3031 * Antiquotation @{syntax_const NAME} ensures that NAME refers to a raw
  3032 syntax constant (cf. 'syntax' command).
  3033 
  3034 * Antiquotation @{make_string} inlines a function to print arbitrary
  3035 values similar to the ML toplevel.  The result is compiler dependent
  3036 and may fall back on "?" in certain situations.
  3037 
  3038 * Diagnostic commands 'ML_val' and 'ML_command' may refer to
  3039 antiquotations @{Isar.state} and @{Isar.goal}.  This replaces impure
  3040 Isar.state() and Isar.goal(), which belong to the old TTY loop and do
  3041 not work with the asynchronous Isar document model.
  3042 
  3043 * Configuration options now admit dynamic default values, depending on
  3044 the context or even global references.
  3045 
  3046 * SHA1.digest digests strings according to SHA-1 (see RFC 3174).  It
  3047 uses an efficient external library if available (for Poly/ML).
  3048 
  3049 * Renamed some important ML structures, while keeping the old names
  3050 for some time as aliases within the structure Legacy:
  3051 
  3052   OuterKeyword  ~>  Keyword
  3053   OuterLex      ~>  Token
  3054   OuterParse    ~>  Parse
  3055   OuterSyntax   ~>  Outer_Syntax
  3056   PrintMode     ~>  Print_Mode
  3057   SpecParse     ~>  Parse_Spec
  3058   ThyInfo       ~>  Thy_Info
  3059   ThyLoad       ~>  Thy_Load
  3060   ThyOutput     ~>  Thy_Output
  3061   TypeInfer     ~>  Type_Infer
  3062 
  3063 Note that "open Legacy" simplifies porting of sources, but forgetting
  3064 to remove it again will complicate porting again in the future.
  3065 
  3066 * Most operations that refer to a global context are named
  3067 accordingly, e.g. Simplifier.global_context or
  3068 ProofContext.init_global.  There are some situations where a global
  3069 context actually works, but under normal circumstances one needs to
  3070 pass the proper local context through the code!
  3071 
  3072 * Discontinued old TheoryDataFun with its copy/init operation -- data
  3073 needs to be pure.  Functor Theory_Data_PP retains the traditional
  3074 Pretty.pp argument to merge, which is absent in the standard
  3075 Theory_Data version.
  3076 
  3077 * Sorts.certify_sort and derived "cert" operations for types and terms
  3078 no longer minimize sorts.  Thus certification at the boundary of the
  3079 inference kernel becomes invariant under addition of class relations,
  3080 which is an important monotonicity principle.  Sorts are now minimized
  3081 in the syntax layer only, at the boundary between the end-user and the
  3082 system.  Subtle INCOMPATIBILITY, may have to use Sign.minimize_sort
  3083 explicitly in rare situations.
  3084 
  3085 * Renamed old-style Drule.standard to Drule.export_without_context, to
  3086 emphasize that this is in no way a standard operation.
  3087 INCOMPATIBILITY.
  3088 
  3089 * Subgoal.FOCUS (and variants): resulting goal state is normalized as
  3090 usual for resolution.  Rare INCOMPATIBILITY.
  3091 
  3092 * Renamed varify/unvarify operations to varify_global/unvarify_global
  3093 to emphasize that these only work in a global situation (which is
  3094 quite rare).
  3095 
  3096 * Curried take and drop in library.ML; negative length is interpreted
  3097 as infinity (as in chop).  Subtle INCOMPATIBILITY.
  3098 
  3099 * Proof terms: type substitutions on proof constants now use canonical
  3100 order of type variables.  INCOMPATIBILITY for tools working with proof
  3101 terms.
  3102 
  3103 * Raw axioms/defs may no longer carry sort constraints, and raw defs
  3104 may no longer carry premises.  User-level specifications are
  3105 transformed accordingly by Thm.add_axiom/add_def.
  3106 
  3107 
  3108 *** System ***
  3109 
  3110 * Discontinued special HOL_USEDIR_OPTIONS for the main HOL image;
  3111 ISABELLE_USEDIR_OPTIONS applies uniformly to all sessions.  Note that
  3112 proof terms are enabled unconditionally in the new HOL-Proofs image.
  3113 
  3114 * Discontinued old ISABELLE and ISATOOL environment settings (legacy
  3115 feature since Isabelle2009).  Use ISABELLE_PROCESS and ISABELLE_TOOL,
  3116 respectively.
  3117 
  3118 * Old lib/scripts/polyml-platform is superseded by the
  3119 ISABELLE_PLATFORM setting variable, which defaults to the 32 bit
  3120 variant, even on a 64 bit machine.  The following example setting
  3121 prefers 64 bit if available:
  3122 
  3123   ML_PLATFORM="${ISABELLE_PLATFORM64:-$ISABELLE_PLATFORM}"
  3124 
  3125 * The preliminary Isabelle/jEdit application demonstrates the emerging
  3126 Isabelle/Scala layer for advanced prover interaction and integration.
  3127 See src/Tools/jEdit or "isabelle jedit" provided by the properly built
  3128 component.
  3129 
  3130 * "IsabelleText" is a Unicode font derived from Bitstream Vera Mono
  3131 and Bluesky TeX fonts.  It provides the usual Isabelle symbols,
  3132 similar to the default assignment of the document preparation system
  3133 (cf. isabellesym.sty).  The Isabelle/Scala class Isabelle_System
  3134 provides some operations for direct access to the font without asking
  3135 the user for manual installation.
  3136 
  3137 
  3138 
  3139 New in Isabelle2009-1 (December 2009)
  3140 -------------------------------------
  3141 
  3142 *** General ***
  3143 
  3144 * Discontinued old form of "escaped symbols" such as \\<forall>.  Only
  3145 one backslash should be used, even in ML sources.
  3146 
  3147 
  3148 *** Pure ***
  3149 
  3150 * Locale interpretation propagates mixins along the locale hierarchy.
  3151 The currently only available mixins are the equations used to map
  3152 local definitions to terms of the target domain of an interpretation.
  3153 
  3154 * Reactivated diagnostic command 'print_interps'.  Use "print_interps
  3155 loc" to print all interpretations of locale "loc" in the theory.
  3156 Interpretations in proofs are not shown.
  3157 
  3158 * Thoroughly revised locales tutorial.  New section on conditional
  3159 interpretation.
  3160 
  3161 * On instantiation of classes, remaining undefined class parameters
  3162 are formally declared.  INCOMPATIBILITY.
  3163 
  3164 
  3165 *** Document preparation ***
  3166 
  3167 * New generalized style concept for printing terms: @{foo (style) ...}
  3168 instead of @{foo_style style ...}  (old form is still retained for
  3169 backward compatibility).  Styles can be also applied for
  3170 antiquotations prop, term_type and typeof.
  3171 
  3172 
  3173 *** HOL ***
  3174 
  3175 * New proof method "smt" for a combination of first-order logic with
  3176 equality, linear and nonlinear (natural/integer/real) arithmetic, and
  3177 fixed-size bitvectors; there is also basic support for higher-order
  3178 features (esp. lambda abstractions).  It is an incomplete decision
  3179 procedure based on external SMT solvers using the oracle mechanism;
  3180 for the SMT solver Z3, this method is proof-producing.  Certificates
  3181 are provided to avoid calling the external solvers solely for
  3182 re-checking proofs.  Due to a remote SMT service there is no need for
  3183 installing SMT solvers locally.  See src/HOL/SMT.
  3184 
  3185 * New commands to load and prove verification conditions generated by
  3186 the Boogie program verifier or derived systems (e.g. the Verifying C
  3187 Compiler (VCC) or Spec#).  See src/HOL/Boogie.
  3188 
  3189 * New counterexample generator tool 'nitpick' based on the Kodkod
  3190 relational model finder.  See src/HOL/Tools/Nitpick and
  3191 src/HOL/Nitpick_Examples.
  3192 
  3193 * New commands 'code_pred' and 'values' to invoke the predicate
  3194 compiler and to enumerate values of inductive predicates.
  3195 
  3196 * A tabled implementation of the reflexive transitive closure.
  3197 
  3198 * New implementation of quickcheck uses generic code generator;
  3199 default generators are provided for all suitable HOL types, records
  3200 and datatypes.  Old quickcheck can be re-activated importing theory
  3201 Library/SML_Quickcheck.
  3202 
  3203 * New testing tool Mirabelle for automated proof tools.  Applies
  3204 several tools and tactics like sledgehammer, metis, or quickcheck, to
  3205 every proof step in a theory.  To be used in batch mode via the
  3206 "mirabelle" utility.
  3207 
  3208 * New proof method "sos" (sum of squares) for nonlinear real
  3209 arithmetic (originally due to John Harison). It requires theory
  3210 Library/Sum_Of_Squares.  It is not a complete decision procedure but
  3211 works well in practice on quantifier-free real arithmetic with +, -,
  3212 *, ^, =, <= and <, i.e. boolean combinations of equalities and
  3213 inequalities between polynomials.  It makes use of external
  3214 semidefinite programming solvers.  Method "sos" generates a
  3215 certificate that can be pasted into the proof thus avoiding the need
  3216 to call an external tool every time the proof is checked.  See
  3217 src/HOL/Library/Sum_Of_Squares.
  3218 
  3219 * New method "linarith" invokes existing linear arithmetic decision
  3220 procedure only.
  3221 
  3222 * New command 'atp_minimal' reduces result produced by Sledgehammer.
  3223 
  3224 * New Sledgehammer option "Full Types" in Proof General settings menu.
  3225 Causes full type information to be output to the ATPs.  This slows
  3226 ATPs down considerably but eliminates a source of unsound "proofs"
  3227 that fail later.
  3228 
  3229 * New method "metisFT": A version of metis that uses full type
  3230 information in order to avoid failures of proof reconstruction.
  3231 
  3232 * New evaluator "approximate" approximates an real valued term using
  3233 the same method as the approximation method.
  3234 
  3235 * Method "approximate" now supports arithmetic expressions as
  3236 boundaries of intervals and implements interval splitting and Taylor
  3237 series expansion.
  3238 
  3239 * ML antiquotation @{code_datatype} inserts definition of a datatype
  3240 generated by the code generator; e.g. see src/HOL/Predicate.thy.
  3241 
  3242 * New theory SupInf of the supremum and infimum operators for sets of
  3243 reals.
  3244 
  3245 * New theory Probability, which contains a development of measure
  3246 theory, eventually leading to Lebesgue integration and probability.
  3247 
  3248 * Extended Multivariate Analysis to include derivation and Brouwer's
  3249 fixpoint theorem.
  3250 
  3251 * Reorganization of number theory, INCOMPATIBILITY:
  3252   - new number theory development for nat and int, in theories Divides
  3253     and GCD as well as in new session Number_Theory
  3254   - some constants and facts now suffixed with _nat and _int
  3255     accordingly
  3256   - former session NumberTheory now named Old_Number_Theory, including
  3257     theories Legacy_GCD and Primes (prefer Number_Theory if possible)
  3258   - moved theory Pocklington from src/HOL/Library to
  3259     src/HOL/Old_Number_Theory
  3260 
  3261 * Theory GCD includes functions Gcd/GCD and Lcm/LCM for the gcd and
  3262 lcm of finite and infinite sets. It is shown that they form a complete
  3263 lattice.
  3264 
  3265 * Class semiring_div requires superclass no_zero_divisors and proof of
  3266 div_mult_mult1; theorems div_mult_mult1, div_mult_mult2,
  3267 div_mult_mult1_if, div_mult_mult1 and div_mult_mult2 have been
  3268 generalized to class semiring_div, subsuming former theorems
  3269 zdiv_zmult_zmult1, zdiv_zmult_zmult1_if, zdiv_zmult_zmult1 and
  3270 zdiv_zmult_zmult2.  div_mult_mult1 is now [simp] by default.
  3271 INCOMPATIBILITY.
  3272 
  3273 * Refinements to lattice classes and sets:
  3274   - less default intro/elim rules in locale variant, more default
  3275     intro/elim rules in class variant: more uniformity
  3276   - lemma ge_sup_conv renamed to le_sup_iff, in accordance with
  3277     le_inf_iff
  3278   - dropped lemma alias inf_ACI for inf_aci (same for sup_ACI and
  3279     sup_aci)
  3280   - renamed ACI to inf_sup_aci
  3281   - new class "boolean_algebra"
  3282   - class "complete_lattice" moved to separate theory
  3283     "Complete_Lattice"; corresponding constants (and abbreviations)
  3284     renamed and with authentic syntax:
  3285     Set.Inf ~>    Complete_Lattice.Inf
  3286     Set.Sup ~>    Complete_Lattice.Sup
  3287     Set.INFI ~>   Complete_Lattice.INFI
  3288     Set.SUPR ~>   Complete_Lattice.SUPR
  3289     Set.Inter ~>  Complete_Lattice.Inter
  3290     Set.Union ~>  Complete_Lattice.Union
  3291     Set.INTER ~>  Complete_Lattice.INTER
  3292     Set.UNION ~>  Complete_Lattice.UNION
  3293   - authentic syntax for
  3294     Set.Pow
  3295     Set.image
  3296   - mere abbreviations:
  3297     Set.empty               (for bot)
  3298     Set.UNIV                (for top)
  3299     Set.inter               (for inf, formerly Set.Int)
  3300     Set.union               (for sup, formerly Set.Un)
  3301     Complete_Lattice.Inter  (for Inf)
  3302     Complete_Lattice.Union  (for Sup)
  3303     Complete_Lattice.INTER  (for INFI)
  3304     Complete_Lattice.UNION  (for SUPR)
  3305   - object-logic definitions as far as appropriate
  3306 
  3307 INCOMPATIBILITY.  Care is required when theorems Int_subset_iff or
  3308 Un_subset_iff are explicitly deleted as default simp rules; then also
  3309 their lattice counterparts le_inf_iff and le_sup_iff have to be
  3310 deleted to achieve the desired effect.
  3311 
  3312 * Rules inf_absorb1, inf_absorb2, sup_absorb1, sup_absorb2 are no simp
  3313 rules by default any longer; the same applies to min_max.inf_absorb1
  3314 etc.  INCOMPATIBILITY.
  3315 
  3316 * Rules sup_Int_eq and sup_Un_eq are no longer declared as
  3317 pred_set_conv by default.  INCOMPATIBILITY.
  3318 
  3319 * Power operations on relations and functions are now one dedicated
  3320 constant "compow" with infix syntax "^^".  Power operation on
  3321 multiplicative monoids retains syntax "^" and is now defined generic
  3322 in class power.  INCOMPATIBILITY.
  3323 
  3324 * Relation composition "R O S" now has a more standard argument order:
  3325 "R O S = {(x, z). EX y. (x, y) : R & (y, z) : S}".  INCOMPATIBILITY,
  3326 rewrite propositions with "S O R" --> "R O S". Proofs may occasionally
  3327 break, since the O_assoc rule was not rewritten like this.  Fix using
  3328 O_assoc[symmetric].  The same applies to the curried version "R OO S".
  3329 
  3330 * Function "Inv" is renamed to "inv_into" and function "inv" is now an
  3331 abbreviation for "inv_into UNIV".  Lemmas are renamed accordingly.
  3332 INCOMPATIBILITY.
  3333 
  3334 * Most rules produced by inductive and datatype package have mandatory
  3335 prefixes.  INCOMPATIBILITY.
  3336 
  3337 * Changed "DERIV_intros" to a dynamic fact, which can be augmented by
  3338 the attribute of the same name.  Each of the theorems in the list
  3339 DERIV_intros assumes composition with an additional function and
  3340 matches a variable to the derivative, which has to be solved by the
  3341 Simplifier.  Hence (auto intro!: DERIV_intros) computes the derivative
  3342 of most elementary terms.  Former Maclauren.DERIV_tac and
  3343 Maclauren.deriv_tac should be replaced by (auto intro!: DERIV_intros).
  3344 INCOMPATIBILITY.
  3345 
  3346 * Code generator attributes follow the usual underscore convention:
  3347     code_unfold     replaces    code unfold
  3348     code_post       replaces    code post
  3349     etc.
  3350   INCOMPATIBILITY.
  3351 
  3352 * Renamed methods:
  3353     sizechange -> size_change
  3354     induct_scheme -> induction_schema
  3355   INCOMPATIBILITY.
  3356 
  3357 * Discontinued abbreviation "arbitrary" of constant "undefined".
  3358 INCOMPATIBILITY, use "undefined" directly.
  3359 
  3360 * Renamed theorems:
  3361     Suc_eq_add_numeral_1 -> Suc_eq_plus1
  3362     Suc_eq_add_numeral_1_left -> Suc_eq_plus1_left
  3363     Suc_plus1 -> Suc_eq_plus1
  3364     *anti_sym -> *antisym*
  3365     vector_less_eq_def -> vector_le_def
  3366   INCOMPATIBILITY.
  3367 
  3368 * Added theorem List.map_map as [simp].  Removed List.map_compose.
  3369 INCOMPATIBILITY.
  3370 
  3371 * Removed predicate "M hassize n" (<--> card M = n & finite M).
  3372 INCOMPATIBILITY.
  3373 
  3374 
  3375 *** HOLCF ***
  3376 
  3377 * Theory Representable defines a class "rep" of domains that are
  3378 representable (via an ep-pair) in the universal domain type "udom".
  3379 Instances are provided for all type constructors defined in HOLCF.
  3380 
  3381 * The 'new_domain' command is a purely definitional version of the
  3382 domain package, for representable domains.  Syntax is identical to the
  3383 old domain package.  The 'new_domain' package also supports indirect
  3384 recursion using previously-defined type constructors.  See
  3385 src/HOLCF/ex/New_Domain.thy for examples.
  3386 
  3387 * Method "fixrec_simp" unfolds one step of a fixrec-defined constant
  3388 on the left-hand side of an equation, and then performs
  3389 simplification.  Rewriting is done using rules declared with the
  3390 "fixrec_simp" attribute.  The "fixrec_simp" method is intended as a
  3391 replacement for "fixpat"; see src/HOLCF/ex/Fixrec_ex.thy for examples.
  3392 
  3393 * The pattern-match compiler in 'fixrec' can now handle constructors
  3394 with HOL function types.  Pattern-match combinators for the Pair
  3395 constructor are pre-configured.
  3396 
  3397 * The 'fixrec' package now produces better fixed-point induction rules
  3398 for mutually-recursive definitions:  Induction rules have conclusions
  3399 of the form "P foo bar" instead of "P <foo, bar>".
  3400 
  3401 * The constant "sq_le" (with infix syntax "<<" or "\<sqsubseteq>") has
  3402 been renamed to "below".  The name "below" now replaces "less" in many
  3403 theorem names.  (Legacy theorem names using "less" are still supported
  3404 as well.)
  3405 
  3406 * The 'fixrec' package now supports "bottom patterns".  Bottom
  3407 patterns can be used to generate strictness rules, or to make
  3408 functions more strict (much like the bang-patterns supported by the
  3409 Glasgow Haskell Compiler).  See src/HOLCF/ex/Fixrec_ex.thy for
  3410 examples.
  3411 
  3412 
  3413 *** ML ***
  3414 
  3415 * Support for Poly/ML 5.3.0, with improved reporting of compiler
  3416 errors and run-time exceptions, including detailed source positions.
  3417 
  3418 * Structure Name_Space (formerly NameSpace) now manages uniquely
  3419 identified entries, with some additional information such as source
  3420 position, logical grouping etc.
  3421 
  3422 * Theory and context data is now introduced by the simplified and
  3423 modernized functors Theory_Data, Proof_Data, Generic_Data.  Data needs
  3424 to be pure, but the old TheoryDataFun for mutable data (with explicit
  3425 copy operation) is still available for some time.
  3426 
  3427 * Structure Synchronized (cf. src/Pure/Concurrent/synchronized.ML)
  3428 provides a high-level programming interface to synchronized state
  3429 variables with atomic update.  This works via pure function
  3430 application within a critical section -- its runtime should be as
  3431 short as possible; beware of deadlocks if critical code is nested,
  3432 either directly or indirectly via other synchronized variables!
  3433 
  3434 * Structure Unsynchronized (cf. src/Pure/ML-Systems/unsynchronized.ML)
  3435 wraps raw ML references, explicitly indicating their non-thread-safe
  3436 behaviour.  The Isar toplevel keeps this structure open, to
  3437 accommodate Proof General as well as quick and dirty interactive
  3438 experiments with references.
  3439 
  3440 * PARALLEL_CHOICE and PARALLEL_GOALS provide basic support for
  3441 parallel tactical reasoning.
  3442 
  3443 * Tacticals Subgoal.FOCUS, Subgoal.FOCUS_PREMS, Subgoal.FOCUS_PARAMS
  3444 are similar to SUBPROOF, but are slightly more flexible: only the
  3445 specified parts of the subgoal are imported into the context, and the
  3446 body tactic may introduce new subgoals and schematic variables.
  3447 
  3448 * Old tactical METAHYPS, which does not observe the proof context, has
  3449 been renamed to Old_Goals.METAHYPS and awaits deletion.  Use SUBPROOF
  3450 or Subgoal.FOCUS etc.
  3451 
  3452 * Renamed functor TableFun to Table, and GraphFun to Graph.  (Since
  3453 functors have their own ML name space there is no point to mark them
  3454 separately.)  Minor INCOMPATIBILITY.
  3455 
  3456 * Renamed NamedThmsFun to Named_Thms.  INCOMPATIBILITY.
  3457 
  3458 * Renamed several structures FooBar to Foo_Bar.  Occasional,
  3459 INCOMPATIBILITY.
  3460 
  3461 * Operations of structure Skip_Proof no longer require quick_and_dirty
  3462 mode, which avoids critical setmp.
  3463 
  3464 * Eliminated old Attrib.add_attributes, Method.add_methods and related
  3465 combinators for "args".  INCOMPATIBILITY, need to use simplified
  3466 Attrib/Method.setup introduced in Isabelle2009.
  3467 
  3468 * Proper context for simpset_of, claset_of, clasimpset_of.  May fall
  3469 back on global_simpset_of, global_claset_of, global_clasimpset_of as
  3470 last resort.  INCOMPATIBILITY.
  3471 
  3472 * Display.pretty_thm now requires a proper context (cf. former
  3473 ProofContext.pretty_thm).  May fall back on Display.pretty_thm_global
  3474 or even Display.pretty_thm_without_context as last resort.
  3475 INCOMPATIBILITY.
  3476 
  3477 * Discontinued Display.pretty_ctyp/cterm etc.  INCOMPATIBILITY, use
  3478 Syntax.pretty_typ/term directly, preferably with proper context
  3479 instead of global theory.
  3480 
  3481 
  3482 *** System ***
  3483 
  3484 * Further fine tuning of parallel proof checking, scales up to 8 cores
  3485 (max. speedup factor 5.0).  See also Goal.parallel_proofs in ML and
  3486 usedir option -q.
  3487 
  3488 * Support for additional "Isabelle components" via etc/components, see
  3489 also the system manual.
  3490 
  3491 * The isabelle makeall tool now operates on all components with
  3492 IsaMakefile, not just hardwired "logics".
  3493 
  3494 * Removed "compress" option from isabelle-process and isabelle usedir;
  3495 this is always enabled.
  3496 
  3497 * Discontinued support for Poly/ML 4.x versions.
  3498 
  3499 * Isabelle tool "wwwfind" provides web interface for 'find_theorems'
  3500 on a given logic image.  This requires the lighttpd webserver and is
  3501 currently supported on Linux only.
  3502 
  3503 
  3504 
  3505 New in Isabelle2009 (April 2009)
  3506 --------------------------------
  3507 
  3508 *** General ***
  3509 
  3510 * Simplified main Isabelle executables, with less surprises on
  3511 case-insensitive file-systems (such as Mac OS).
  3512 
  3513   - The main Isabelle tool wrapper is now called "isabelle" instead of
  3514     "isatool."
  3515 
  3516   - The former "isabelle" alias for "isabelle-process" has been
  3517     removed (should rarely occur to regular users).
  3518 
  3519   - The former "isabelle-interface" and its alias "Isabelle" have been
  3520     removed (interfaces are now regular Isabelle tools).
  3521 
  3522 Within scripts and make files, the Isabelle environment variables
  3523 ISABELLE_TOOL and ISABELLE_PROCESS replace old ISATOOL and ISABELLE,
  3524 respectively.  (The latter are still available as legacy feature.)
  3525 
  3526 The old isabelle-interface wrapper could react in confusing ways if
  3527 the interface was uninstalled or changed otherwise.  Individual
  3528 interface tool configuration is now more explicit, see also the
  3529 Isabelle system manual.  In particular, Proof General is now available
  3530 via "isabelle emacs".
  3531 
  3532 INCOMPATIBILITY, need to adapt derivative scripts.  Users may need to
  3533 purge installed copies of Isabelle executables and re-run "isabelle
  3534 install -p ...", or use symlinks.
  3535 
  3536 * The default for ISABELLE_HOME_USER is now ~/.isabelle instead of the
  3537 old ~/isabelle, which was slightly non-standard and apt to cause
  3538 surprises on case-insensitive file-systems (such as Mac OS).
  3539 
  3540 INCOMPATIBILITY, need to move existing ~/isabelle/etc,
  3541 ~/isabelle/heaps, ~/isabelle/browser_info to the new place.  Special
  3542 care is required when using older releases of Isabelle.  Note that
  3543 ISABELLE_HOME_USER can be changed in Isabelle/etc/settings of any
  3544 Isabelle distribution, in order to use the new ~/.isabelle uniformly.
  3545 
  3546 * Proofs of fully specified statements are run in parallel on
  3547 multi-core systems.  A speedup factor of 2.5 to 3.2 can be expected on
  3548 a regular 4-core machine, if the initial heap space is made reasonably
  3549 large (cf. Poly/ML option -H).  (Requires Poly/ML 5.2.1 or later.)
  3550 
  3551 * The main reference manuals ("isar-ref", "implementation", and
  3552 "system") have been updated and extended.  Formally checked references
  3553 as hyperlinks are now available uniformly.
  3554 
  3555 
  3556 *** Pure ***
  3557 
  3558 * Complete re-implementation of locales.  INCOMPATIBILITY in several
  3559 respects.  The most important changes are listed below.  See the
  3560 Tutorial on Locales ("locales" manual) for details.
  3561 
  3562 - In locale expressions, instantiation replaces renaming.  Parameters
  3563 must be declared in a for clause.  To aid compatibility with previous
  3564 parameter inheritance, in locale declarations, parameters that are not
  3565 'touched' (instantiation position "_" or omitted) are implicitly added
  3566 with their syntax at the beginning of the for clause.
  3567 
  3568 - Syntax from abbreviations and definitions in locales is available in
  3569 locale expressions and context elements.  The latter is particularly
  3570 useful in locale declarations.
  3571 
  3572 - More flexible mechanisms to qualify names generated by locale
  3573 expressions.  Qualifiers (prefixes) may be specified in locale
  3574 expressions, and can be marked as mandatory (syntax: "name!:") or
  3575 optional (syntax "name?:").  The default depends for plain "name:"
  3576 depends on the situation where a locale expression is used: in
  3577 commands 'locale' and 'sublocale' prefixes are optional, in
  3578 'interpretation' and 'interpret' prefixes are mandatory.  The old
  3579 implicit qualifiers derived from the parameter names of a locale are
  3580 no longer generated.
  3581 
  3582 - Command "sublocale l < e" replaces "interpretation l < e".  The
  3583 instantiation clause in "interpretation" and "interpret" (square
  3584 brackets) is no longer available.  Use locale expressions.
  3585 
  3586 - When converting proof scripts, mandatory qualifiers in
  3587 'interpretation' and 'interpret' should be retained by default, even
  3588 if this is an INCOMPATIBILITY compared to former behavior.  In the
  3589 worst case, use the "name?:" form for non-mandatory ones.  Qualifiers
  3590 in locale expressions range over a single locale instance only.
  3591 
  3592 - Dropped locale element "includes".  This is a major INCOMPATIBILITY.
  3593 In existing theorem specifications replace the includes element by the
  3594 respective context elements of the included locale, omitting those
  3595 that are already present in the theorem specification.  Multiple
  3596 assume elements of a locale should be replaced by a single one
  3597 involving the locale predicate.  In the proof body, declarations (most
  3598 notably theorems) may be regained by interpreting the respective
  3599 locales in the proof context as required (command "interpret").
  3600 
  3601 If using "includes" in replacement of a target solely because the
  3602 parameter types in the theorem are not as general as in the target,
  3603 consider declaring a new locale with additional type constraints on
  3604 the parameters (context element "constrains").
  3605 
  3606 - Discontinued "locale (open)".  INCOMPATIBILITY.
  3607 
  3608 - Locale interpretation commands no longer attempt to simplify goal.
  3609 INCOMPATIBILITY: in rare situations the generated goal differs.  Use
  3610 methods intro_locales and unfold_locales to clarify.
  3611 
  3612 - Locale interpretation commands no longer accept interpretation
  3613 attributes.  INCOMPATIBILITY.
  3614 
  3615 * Class declaration: so-called "base sort" must not be given in import
  3616 list any longer, but is inferred from the specification.  Particularly
  3617 in HOL, write
  3618 
  3619     class foo = ...
  3620 
  3621 instead of
  3622 
  3623     class foo = type + ...
  3624 
  3625 * Class target: global versions of theorems stemming do not carry a
  3626 parameter prefix any longer.  INCOMPATIBILITY.
  3627 
  3628 * Class 'instance' command no longer accepts attached definitions.
  3629 INCOMPATIBILITY, use proper 'instantiation' target instead.
  3630 
  3631 * Recovered hiding of consts, which was accidentally broken in
  3632 Isabelle2007.  Potential INCOMPATIBILITY, ``hide const c'' really
  3633 makes c inaccessible; consider using ``hide (open) const c'' instead.
  3634 
  3635 * Slightly more coherent Pure syntax, with updated documentation in
  3636 isar-ref manual.  Removed locales meta_term_syntax and
  3637 meta_conjunction_syntax: TERM and &&& (formerly &&) are now permanent,
  3638 INCOMPATIBILITY in rare situations.  Note that &&& should not be used
  3639 directly in regular applications.
  3640 
  3641 * There is a new syntactic category "float_const" for signed decimal
  3642 fractions (e.g. 123.45 or -123.45).
  3643 
  3644 * Removed exotic 'token_translation' command.  INCOMPATIBILITY, use ML
  3645 interface with 'setup' command instead.
  3646 
  3647 * Command 'local_setup' is similar to 'setup', but operates on a local
  3648 theory context.
  3649 
  3650 * The 'axiomatization' command now only works within a global theory
  3651 context.  INCOMPATIBILITY.
  3652 
  3653 * Goal-directed proof now enforces strict proof irrelevance wrt. sort
  3654 hypotheses.  Sorts required in the course of reasoning need to be
  3655 covered by the constraints in the initial statement, completed by the
  3656 type instance information of the background theory.  Non-trivial sort
  3657 hypotheses, which rarely occur in practice, may be specified via
  3658 vacuous propositions of the form SORT_CONSTRAINT('a::c).  For example:
  3659 
  3660   lemma assumes "SORT_CONSTRAINT('a::empty)" shows False ...
  3661 
  3662 The result contains an implicit sort hypotheses as before --
  3663 SORT_CONSTRAINT premises are eliminated as part of the canonical rule
  3664 normalization.
  3665 
  3666 * Generalized Isar history, with support for linear undo, direct state
  3667 addressing etc.
  3668 
  3669 * Changed defaults for unify configuration options:
  3670 
  3671   unify_trace_bound = 50 (formerly 25)
  3672   unify_search_bound = 60 (formerly 30)
  3673 
  3674 * Different bookkeeping for code equations (INCOMPATIBILITY):
  3675 
  3676   a) On theory merge, the last set of code equations for a particular
  3677      constant is taken (in accordance with the policy applied by other
  3678      parts of the code generator framework).
  3679 
  3680   b) Code equations stemming from explicit declarations (e.g. code
  3681      attribute) gain priority over default code equations stemming
  3682      from definition, primrec, fun etc.
  3683 
  3684 * Keyword 'code_exception' now named 'code_abort'.  INCOMPATIBILITY.
  3685 
  3686 * Unified theorem tables for both code generators.  Thus [code
  3687 func] has disappeared and only [code] remains.  INCOMPATIBILITY.
  3688 
  3689 * Command 'find_consts' searches for constants based on type and name
  3690 patterns, e.g.
  3691 
  3692     find_consts "_ => bool"
  3693 
  3694 By default, matching is against subtypes, but it may be restricted to
  3695 the whole type.  Searching by name is possible.  Multiple queries are
  3696 conjunctive and queries may be negated by prefixing them with a
  3697 hyphen:
  3698 
  3699     find_consts strict: "_ => bool" name: "Int" -"int => int"
  3700 
  3701 * New 'find_theorems' criterion "solves" matches theorems that
  3702 directly solve the current goal (modulo higher-order unification).
  3703 
  3704 * Auto solve feature for main theorem statements: whenever a new goal
  3705 is stated, "find_theorems solves" is called; any theorems that could
  3706 solve the lemma directly are listed as part of the goal state.
  3707 Cf. associated options in Proof General Isabelle settings menu,
  3708 enabled by default, with reasonable timeout for pathological cases of
  3709 higher-order unification.
  3710 
  3711 
  3712 *** Document preparation ***
  3713 
  3714 * Antiquotation @{lemma} now imitates a regular terminal proof,
  3715 demanding keyword 'by' and supporting the full method expression
  3716 syntax just like the Isar command 'by'.
  3717 
  3718 
  3719 *** HOL ***
  3720 
  3721 * Integrated main parts of former image HOL-Complex with HOL.  Entry
  3722 points Main and Complex_Main remain as before.
  3723 
  3724 * Logic image HOL-Plain provides a minimal HOL with the most important
  3725 tools available (inductive, datatype, primrec, ...).  This facilitates
  3726 experimentation and tool development.  Note that user applications
  3727 (and library theories) should never refer to anything below theory
  3728 Main, as before.
  3729 
  3730 * Logic image HOL-Main stops at theory Main, and thus facilitates
  3731 experimentation due to shorter build times.
  3732 
  3733 * Logic image HOL-NSA contains theories of nonstandard analysis which
  3734 were previously part of former HOL-Complex.  Entry point Hyperreal
  3735 remains valid, but theories formerly using Complex_Main should now use
  3736 new entry point Hypercomplex.
  3737 
  3738 * Generic ATP manager for Sledgehammer, based on ML threads instead of
  3739 Posix processes.  Avoids potentially expensive forking of the ML
  3740 process.  New thread-based implementation also works on non-Unix
  3741 platforms (Cygwin).  Provers are no longer hardwired, but defined
  3742 within the theory via plain ML wrapper functions.  Basic Sledgehammer
  3743 commands are covered in the isar-ref manual.
  3744 
  3745 * Wrapper scripts for remote SystemOnTPTP service allows to use
  3746 sledgehammer without local ATP installation (Vampire etc.). Other
  3747 provers may be included via suitable ML wrappers, see also
  3748 src/HOL/ATP_Linkup.thy.
  3749 
  3750 * ATP selection (E/Vampire/Spass) is now via Proof General's settings
  3751 menu.
  3752 
  3753 * The metis method no longer fails because the theorem is too trivial
  3754 (contains the empty clause).
  3755 
  3756 * The metis method now fails in the usual manner, rather than raising
  3757 an exception, if it determines that it cannot prove the theorem.
  3758 
  3759 * Method "coherent" implements a prover for coherent logic (see also
  3760 src/Tools/coherent.ML).
  3761 
  3762 * Constants "undefined" and "default" replace "arbitrary".  Usually
  3763 "undefined" is the right choice to replace "arbitrary", though
  3764 logically there is no difference.  INCOMPATIBILITY.
  3765 
  3766 * Command "value" now integrates different evaluation mechanisms.  The
  3767 result of the first successful evaluation mechanism is printed.  In
  3768 square brackets a particular named evaluation mechanisms may be
  3769 specified (currently, [SML], [code] or [nbe]).  See further
  3770 src/HOL/ex/Eval_Examples.thy.
  3771 
  3772 * Normalization by evaluation now allows non-leftlinear equations.
  3773 Declare with attribute [code nbe].
  3774 
  3775 * Methods "case_tac" and "induct_tac" now refer to the very same rules
  3776 as the structured Isar versions "cases" and "induct", cf. the
  3777 corresponding "cases" and "induct" attributes.  Mutual induction rules
  3778 are now presented as a list of individual projections
  3779 (e.g. foo_bar.inducts for types foo and bar); the old format with
  3780 explicit HOL conjunction is no longer supported.  INCOMPATIBILITY, in
  3781 rare situations a different rule is selected --- notably nested tuple
  3782 elimination instead of former prod.exhaust: use explicit (case_tac t
  3783 rule: prod.exhaust) here.
  3784 
  3785 * Attributes "cases", "induct", "coinduct" support "del" option.
  3786 
  3787 * Removed fact "case_split_thm", which duplicates "case_split".
  3788 
  3789 * The option datatype has been moved to a new theory Option.  Renamed
  3790 option_map to Option.map, and o2s to Option.set, INCOMPATIBILITY.
  3791 
  3792 * New predicate "strict_mono" classifies strict functions on partial
  3793 orders.  With strict functions on linear orders, reasoning about
  3794 (in)equalities is facilitated by theorems "strict_mono_eq",
  3795 "strict_mono_less_eq" and "strict_mono_less".
  3796 
  3797 * Some set operations are now proper qualified constants with
  3798 authentic syntax.  INCOMPATIBILITY:
  3799 
  3800     op Int ~>   Set.Int
  3801     op Un ~>    Set.Un
  3802     INTER ~>    Set.INTER
  3803     UNION ~>    Set.UNION
  3804     Inter ~>    Set.Inter
  3805     Union ~>    Set.Union
  3806     {} ~>       Set.empty
  3807     UNIV ~>     Set.UNIV
  3808 
  3809 * Class complete_lattice with operations Inf, Sup, INFI, SUPR now in
  3810 theory Set.
  3811 
  3812 * Auxiliary class "itself" has disappeared -- classes without any
  3813 parameter are treated as expected by the 'class' command.
  3814 
  3815 * Leibnitz's Series for Pi and the arcus tangens and logarithm series.
  3816 
  3817 * Common decision procedures (Cooper, MIR, Ferrack, Approximation,
  3818 Dense_Linear_Order) are now in directory HOL/Decision_Procs.
  3819 
  3820 * Theory src/HOL/Decision_Procs/Approximation provides the new proof
  3821 method "approximation".  It proves formulas on real values by using
  3822 interval arithmetic.  In the formulas are also the transcendental
  3823 functions sin, cos, tan, atan, ln, exp and the constant pi are
  3824 allowed. For examples see
  3825 src/HOL/Descision_Procs/ex/Approximation_Ex.thy.
  3826 
  3827 * Theory "Reflection" now resides in HOL/Library.
  3828 
  3829 * Entry point to Word library now simply named "Word".
  3830 INCOMPATIBILITY.
  3831 
  3832 * Made source layout more coherent with logical distribution
  3833 structure:
  3834 
  3835     src/HOL/Library/RType.thy ~> src/HOL/Typerep.thy
  3836     src/HOL/Library/Code_Message.thy ~> src/HOL/
  3837     src/HOL/Library/GCD.thy ~> src/HOL/
  3838     src/HOL/Library/Order_Relation.thy ~> src/HOL/
  3839     src/HOL/Library/Parity.thy ~> src/HOL/
  3840     src/HOL/Library/Univ_Poly.thy ~> src/HOL/
  3841     src/HOL/Real/ContNotDenum.thy ~> src/HOL/Library/
  3842     src/HOL/Real/Lubs.thy ~> src/HOL/
  3843     src/HOL/Real/PReal.thy ~> src/HOL/
  3844     src/HOL/Real/Rational.thy ~> src/HOL/
  3845     src/HOL/Real/RComplete.thy ~> src/HOL/
  3846     src/HOL/Real/RealDef.thy ~> src/HOL/
  3847     src/HOL/Real/RealPow.thy ~> src/HOL/
  3848     src/HOL/Real/Real.thy ~> src/HOL/
  3849     src/HOL/Complex/Complex_Main.thy ~> src/HOL/
  3850     src/HOL/Complex/Complex.thy ~> src/HOL/
  3851     src/HOL/Complex/FrechetDeriv.thy ~> src/HOL/Library/
  3852     src/HOL/Complex/Fundamental_Theorem_Algebra.thy ~> src/HOL/Library/
  3853     src/HOL/Hyperreal/Deriv.thy ~> src/HOL/
  3854     src/HOL/Hyperreal/Fact.thy ~> src/HOL/
  3855     src/HOL/Hyperreal/Integration.thy ~> src/HOL/
  3856     src/HOL/Hyperreal/Lim.thy ~> src/HOL/
  3857     src/HOL/Hyperreal/Ln.thy ~> src/HOL/
  3858     src/HOL/Hyperreal/Log.thy ~> src/HOL/
  3859     src/HOL/Hyperreal/MacLaurin.thy ~> src/HOL/
  3860     src/HOL/Hyperreal/NthRoot.thy ~> src/HOL/
  3861     src/HOL/Hyperreal/Series.thy ~> src/HOL/
  3862     src/HOL/Hyperreal/SEQ.thy ~> src/HOL/
  3863     src/HOL/Hyperreal/Taylor.thy ~> src/HOL/
  3864     src/HOL/Hyperreal/Transcendental.thy ~> src/HOL/
  3865     src/HOL/Real/Float ~> src/HOL/Library/
  3866     src/HOL/Real/HahnBanach ~> src/HOL/HahnBanach
  3867     src/HOL/Real/RealVector.thy ~> src/HOL/
  3868 
  3869     src/HOL/arith_data.ML ~> src/HOL/Tools
  3870     src/HOL/hologic.ML ~> src/HOL/Tools
  3871     src/HOL/simpdata.ML ~> src/HOL/Tools
  3872     src/HOL/int_arith1.ML ~> src/HOL/Tools/int_arith.ML
  3873     src/HOL/int_factor_simprocs.ML ~> src/HOL/Tools
  3874     src/HOL/nat_simprocs.ML ~> src/HOL/Tools
  3875     src/HOL/Real/float_arith.ML ~> src/HOL/Tools
  3876     src/HOL/Real/float_syntax.ML ~> src/HOL/Tools
  3877     src/HOL/Real/rat_arith.ML ~> src/HOL/Tools
  3878     src/HOL/Real/real_arith.ML ~> src/HOL/Tools
  3879 
  3880     src/HOL/Library/Array.thy ~> src/HOL/Imperative_HOL
  3881     src/HOL/Library/Heap_Monad.thy ~> src/HOL/Imperative_HOL
  3882     src/HOL/Library/Heap.thy ~> src/HOL/Imperative_HOL
  3883     src/HOL/Library/Imperative_HOL.thy ~> src/HOL/Imperative_HOL
  3884     src/HOL/Library/Ref.thy ~> src/HOL/Imperative_HOL
  3885     src/HOL/Library/Relational.thy ~> src/HOL/Imperative_HOL
  3886 
  3887 * If methods "eval" and "evaluation" encounter a structured proof
  3888 state with !!/==>, only the conclusion is evaluated to True (if
  3889 possible), avoiding strange error messages.
  3890 
  3891 * Method "sizechange" automates termination proofs using (a
  3892 modification of) the size-change principle.  Requires SAT solver.  See
  3893 src/HOL/ex/Termination.thy for examples.
  3894 
  3895 * Simplifier: simproc for let expressions now unfolds if bound
  3896 variable occurs at most once in let expression body.  INCOMPATIBILITY.
  3897 
  3898 * Method "arith": Linear arithmetic now ignores all inequalities when
  3899 fast_arith_neq_limit is exceeded, instead of giving up entirely.
  3900 
  3901 * New attribute "arith" for facts that should always be used
  3902 automatically by arithmetic. It is intended to be used locally in
  3903 proofs, e.g.
  3904 
  3905   assumes [arith]: "x > 0"
  3906 
  3907 Global usage is discouraged because of possible performance impact.
  3908 
  3909 * New classes "top" and "bot" with corresponding operations "top" and
  3910 "bot" in theory Orderings; instantiation of class "complete_lattice"
  3911 requires instantiation of classes "top" and "bot".  INCOMPATIBILITY.
  3912 
  3913 * Changed definition lemma "less_fun_def" in order to provide an
  3914 instance for preorders on functions; use lemma "less_le" instead.
  3915 INCOMPATIBILITY.
  3916 
  3917 * Theory Orderings: class "wellorder" moved here, with explicit
  3918 induction rule "less_induct" as assumption.  For instantiation of
  3919 "wellorder" by means of predicate "wf", use rule wf_wellorderI.
  3920 INCOMPATIBILITY.
  3921 
  3922 * Theory Orderings: added class "preorder" as superclass of "order".
  3923 INCOMPATIBILITY: Instantiation proofs for order, linorder
  3924 etc. slightly changed.  Some theorems named order_class.* now named
  3925 preorder_class.*.
  3926 
  3927 * Theory Relation: renamed "refl" to "refl_on", "reflexive" to "refl,
  3928 "diag" to "Id_on".
  3929 
  3930 * Theory Finite_Set: added a new fold combinator of type
  3931 
  3932   ('a => 'b => 'b) => 'b => 'a set => 'b
  3933 
  3934 Occasionally this is more convenient than the old fold combinator
  3935 which is now defined in terms of the new one and renamed to
  3936 fold_image.
  3937 
  3938 * Theories Ring_and_Field and OrderedGroup: The lemmas "group_simps"
  3939 and "ring_simps" have been replaced by "algebra_simps" (which can be
  3940 extended with further lemmas!).  At the moment both still exist but
  3941 the former will disappear at some point.
  3942 
  3943 * Theory Power: Lemma power_Suc is now declared as a simp rule in
  3944 class recpower.  Type-specific simp rules for various recpower types
  3945 have been removed.  INCOMPATIBILITY, rename old lemmas as follows:
  3946 
  3947 rat_power_0    -> power_0
  3948 rat_power_Suc  -> power_Suc
  3949 realpow_0      -> power_0
  3950 realpow_Suc    -> power_Suc
  3951 complexpow_0   -> power_0
  3952 complexpow_Suc -> power_Suc
  3953 power_poly_0   -> power_0
  3954 power_poly_Suc -> power_Suc
  3955 
  3956 * Theories Ring_and_Field and Divides: Definition of "op dvd" has been
  3957 moved to separate class dvd in Ring_and_Field; a couple of lemmas on
  3958 dvd has been generalized to class comm_semiring_1.  Likewise a bunch
  3959 of lemmas from Divides has been generalized from nat to class
  3960 semiring_div.  INCOMPATIBILITY.  This involves the following theorem
  3961 renames resulting from duplicate elimination:
  3962 
  3963     dvd_def_mod ~>          dvd_eq_mod_eq_0
  3964     zero_dvd_iff ~>         dvd_0_left_iff
  3965     dvd_0 ~>                dvd_0_right
  3966     DIVISION_BY_ZERO_DIV ~> div_by_0
  3967     DIVISION_BY_ZERO_MOD ~> mod_by_0
  3968     mult_div ~>             div_mult_self2_is_id
  3969     mult_mod ~>             mod_mult_self2_is_0
  3970 
  3971 * Theory IntDiv: removed many lemmas that are instances of class-based
  3972 generalizations (from Divides and Ring_and_Field).  INCOMPATIBILITY,
  3973 rename old lemmas as follows:
  3974 
  3975 dvd_diff               -> nat_dvd_diff
  3976 dvd_zminus_iff         -> dvd_minus_iff
  3977 mod_add1_eq            -> mod_add_eq
  3978 mod_mult1_eq           -> mod_mult_right_eq
  3979 mod_mult1_eq'          -> mod_mult_left_eq
  3980 mod_mult_distrib_mod   -> mod_mult_eq
  3981 nat_mod_add_left_eq    -> mod_add_left_eq
  3982 nat_mod_add_right_eq   -> mod_add_right_eq
  3983 nat_mod_div_trivial    -> mod_div_trivial
  3984 nat_mod_mod_trivial    -> mod_mod_trivial
  3985 zdiv_zadd_self1        -> div_add_self1
  3986 zdiv_zadd_self2        -> div_add_self2
  3987 zdiv_zmult_self1       -> div_mult_self2_is_id
  3988 zdiv_zmult_self2       -> div_mult_self1_is_id
  3989 zdvd_triv_left         -> dvd_triv_left
  3990 zdvd_triv_right        -> dvd_triv_right
  3991 zdvd_zmult_cancel_disj -> dvd_mult_cancel_left
  3992 zmod_eq0_zdvd_iff      -> dvd_eq_mod_eq_0[symmetric]
  3993 zmod_zadd_left_eq      -> mod_add_left_eq
  3994 zmod_zadd_right_eq     -> mod_add_right_eq
  3995 zmod_zadd_self1        -> mod_add_self1
  3996 zmod_zadd_self2        -> mod_add_self2
  3997 zmod_zadd1_eq          -> mod_add_eq
  3998 zmod_zdiff1_eq         -> mod_diff_eq
  3999 zmod_zdvd_zmod         -> mod_mod_cancel
  4000 zmod_zmod_cancel       -> mod_mod_cancel
  4001 zmod_zmult_self1       -> mod_mult_self2_is_0
  4002 zmod_zmult_self2       -> mod_mult_self1_is_0
  4003 zmod_1                 -> mod_by_1
  4004 zdiv_1                 -> div_by_1
  4005 zdvd_abs1              -> abs_dvd_iff
  4006 zdvd_abs2              -> dvd_abs_iff
  4007 zdvd_refl              -> dvd_refl
  4008 zdvd_trans             -> dvd_trans
  4009 zdvd_zadd              -> dvd_add
  4010 zdvd_zdiff             -> dvd_diff
  4011 zdvd_zminus_iff        -> dvd_minus_iff
  4012 zdvd_zminus2_iff       -> minus_dvd_iff
  4013 zdvd_zmultD            -> dvd_mult_right
  4014 zdvd_zmultD2           -> dvd_mult_left
  4015 zdvd_zmult_mono        -> mult_dvd_mono
  4016 zdvd_0_right           -> dvd_0_right
  4017 zdvd_0_left            -> dvd_0_left_iff
  4018 zdvd_1_left            -> one_dvd
  4019 zminus_dvd_iff         -> minus_dvd_iff
  4020 
  4021 * Theory Rational: 'Fract k 0' now equals '0'.  INCOMPATIBILITY.
  4022 
  4023 * The real numbers offer decimal input syntax: 12.34 is translated
  4024 into 1234/10^2. This translation is not reversed upon output.
  4025 
  4026 * Theory Library/Polynomial defines an abstract type 'a poly of
  4027 univariate polynomials with coefficients of type 'a.  In addition to
  4028 the standard ring operations, it also supports div and mod.  Code
  4029 generation is also supported, using list-style constructors.
  4030 
  4031 * Theory Library/Inner_Product defines a class of real_inner for real
  4032 inner product spaces, with an overloaded operation inner :: 'a => 'a
  4033 => real.  Class real_inner is a subclass of real_normed_vector from
  4034 theory RealVector.
  4035 
  4036 * Theory Library/Product_Vector provides instances for the product
  4037 type 'a * 'b of several classes from RealVector and Inner_Product.
  4038 Definitions of addition, subtraction, scalar multiplication, norms,
  4039 and inner products are included.
  4040 
  4041 * Theory Library/Bit defines the field "bit" of integers modulo 2.  In
  4042 addition to the field operations, numerals and case syntax are also
  4043 supported.
  4044 
  4045 * Theory Library/Diagonalize provides constructive version of Cantor's
  4046 first diagonalization argument.
  4047 
  4048 * Theory Library/GCD: Curried operations gcd, lcm (for nat) and zgcd,
  4049 zlcm (for int); carried together from various gcd/lcm developements in
  4050 the HOL Distribution.  Constants zgcd and zlcm replace former igcd and
  4051 ilcm; corresponding theorems renamed accordingly.  INCOMPATIBILITY,
  4052 may recover tupled syntax as follows:
  4053 
  4054     hide (open) const gcd
  4055     abbreviation gcd where
  4056       "gcd == (%(a, b). GCD.gcd a b)"
  4057     notation (output)
  4058       GCD.gcd ("gcd '(_, _')")
  4059 
  4060 The same works for lcm, zgcd, zlcm.
  4061 
  4062 * Theory Library/Nat_Infinity: added addition, numeral syntax and more
  4063 instantiations for algebraic structures.  Removed some duplicate
  4064 theorems.  Changes in simp rules.  INCOMPATIBILITY.
  4065 
  4066 * ML antiquotation @{code} takes a constant as argument and generates
  4067 corresponding code in background and inserts name of the corresponding
  4068 resulting ML value/function/datatype constructor binding in place.
  4069 All occurrences of @{code} with a single ML block are generated
  4070 simultaneously.  Provides a generic and safe interface for
  4071 instrumentalizing code generation.  See
  4072 src/HOL/Decision_Procs/Ferrack.thy for a more ambitious application.
  4073 In future you ought to refrain from ad-hoc compiling generated SML
  4074 code on the ML toplevel.  Note that (for technical reasons) @{code}
  4075 cannot refer to constants for which user-defined serializations are
  4076 set.  Refer to the corresponding ML counterpart directly in that
  4077 cases.
  4078 
  4079 * Command 'rep_datatype': instead of theorem names the command now
  4080 takes a list of terms denoting the constructors of the type to be
  4081 represented as datatype.  The characteristic theorems have to be
  4082 proven.  INCOMPATIBILITY.  Also observe that the following theorems
  4083 have disappeared in favour of existing ones:
  4084 
  4085     unit_induct                 ~> unit.induct
  4086     prod_induct                 ~> prod.induct
  4087     sum_induct                  ~> sum.induct
  4088     Suc_Suc_eq                  ~> nat.inject
  4089     Suc_not_Zero Zero_not_Suc   ~> nat.distinct
  4090 
  4091 
  4092 *** HOL-Algebra ***
  4093 
  4094 * New locales for orders and lattices where the equivalence relation
  4095 is not restricted to equality.  INCOMPATIBILITY: all order and lattice
  4096 locales use a record structure with field eq for the equivalence.
  4097 
  4098 * New theory of factorial domains.
  4099 
  4100 * Units_l_inv and Units_r_inv are now simp rules by default.
  4101 INCOMPATIBILITY.  Simplifier proof that require deletion of l_inv
  4102 and/or r_inv will now also require deletion of these lemmas.
  4103 
  4104 * Renamed the following theorems, INCOMPATIBILITY:
  4105 
  4106 UpperD ~> Upper_memD
  4107 LowerD ~> Lower_memD
  4108 least_carrier ~> least_closed
  4109 greatest_carrier ~> greatest_closed
  4110 greatest_Lower_above ~> greatest_Lower_below
  4111 one_zero ~> carrier_one_zero
  4112 one_not_zero ~> carrier_one_not_zero  (collision with assumption)
  4113 
  4114 
  4115 *** HOL-Nominal ***
  4116 
  4117 * Nominal datatypes can now contain type-variables.
  4118 
  4119 * Commands 'nominal_inductive' and 'equivariance' work with local
  4120 theory targets.
  4121 
  4122 * Nominal primrec can now works with local theory targets and its
  4123 specification syntax now conforms to the general format as seen in
  4124 'inductive' etc.
  4125 
  4126 * Method "perm_simp" honours the standard simplifier attributes
  4127 (no_asm), (no_asm_use) etc.
  4128 
  4129 * The new predicate #* is defined like freshness, except that on the
  4130 left hand side can be a set or list of atoms.
  4131 
  4132 * Experimental command 'nominal_inductive2' derives strong induction
  4133 principles for inductive definitions.  In contrast to
  4134 'nominal_inductive', which can only deal with a fixed number of
  4135 binders, it can deal with arbitrary expressions standing for sets of
  4136 atoms to be avoided.  The only inductive definition we have at the
  4137 moment that needs this generalisation is the typing rule for Lets in
  4138 the algorithm W:
  4139 
  4140  Gamma |- t1 : T1   (x,close Gamma T1)::Gamma |- t2 : T2   x#Gamma
  4141  -----------------------------------------------------------------
  4142          Gamma |- Let x be t1 in t2 : T2
  4143 
  4144 In this rule one wants to avoid all the binders that are introduced by
  4145 "close Gamma T1".  We are looking for other examples where this
  4146 feature might be useful.  Please let us know.
  4147 
  4148 
  4149 *** HOLCF ***
  4150 
  4151 * Reimplemented the simplification procedure for proving continuity
  4152 subgoals.  The new simproc is extensible; users can declare additional
  4153 continuity introduction rules with the attribute [cont2cont].
  4154 
  4155 * The continuity simproc now uses a different introduction rule for
  4156 solving continuity subgoals on terms with lambda abstractions.  In
  4157 some rare cases the new simproc may fail to solve subgoals that the
  4158 old one could solve, and "simp add: cont2cont_LAM" may be necessary.
  4159 Potential INCOMPATIBILITY.
  4160 
  4161 * Command 'fixrec': specification syntax now conforms to the general
  4162 format as seen in 'inductive' etc.  See src/HOLCF/ex/Fixrec_ex.thy for
  4163 examples.  INCOMPATIBILITY.
  4164 
  4165 
  4166 *** ZF ***
  4167 
  4168 * Proof of Zorn's Lemma for partial orders.
  4169 
  4170 
  4171 *** ML ***
  4172 
  4173 * Multithreading for Poly/ML 5.1/5.2 is no longer supported, only for
  4174 Poly/ML 5.2.1 or later.  Important note: the TimeLimit facility
  4175 depends on multithreading, so timouts will not work before Poly/ML
  4176 5.2.1!
  4177 
  4178 * High-level support for concurrent ML programming, see
  4179 src/Pure/Cuncurrent.  The data-oriented model of "future values" is
  4180 particularly convenient to organize independent functional
  4181 computations.  The concept of "synchronized variables" provides a
  4182 higher-order interface for components with shared state, avoiding the
  4183 delicate details of mutexes and condition variables.  (Requires
  4184 Poly/ML 5.2.1 or later.)
  4185 
  4186 * ML bindings produced via Isar commands are stored within the Isar
  4187 context (theory or proof).  Consequently, commands like 'use' and 'ML'
  4188 become thread-safe and work with undo as expected (concerning
  4189 top-level bindings, not side-effects on global references).
  4190 INCOMPATIBILITY, need to provide proper Isar context when invoking the
  4191 compiler at runtime; really global bindings need to be given outside a
  4192 theory.  (Requires Poly/ML 5.2 or later.)
  4193 
  4194 * Command 'ML_prf' is analogous to 'ML' but works within a proof
  4195 context.  Top-level ML bindings are stored within the proof context in
  4196 a purely sequential fashion, disregarding the nested proof structure.
  4197 ML bindings introduced by 'ML_prf' are discarded at the end of the
  4198 proof.  (Requires Poly/ML 5.2 or later.)
  4199 
  4200 * Simplified ML attribute and method setup, cf. functions Attrib.setup
  4201 and Method.setup, as well as Isar commands 'attribute_setup' and
  4202 'method_setup'.  INCOMPATIBILITY for 'method_setup', need to simplify
  4203 existing code accordingly, or use plain 'setup' together with old
  4204 Method.add_method.
  4205 
  4206 * Simplified ML oracle interface Thm.add_oracle promotes 'a -> cterm
  4207 to 'a -> thm, while results are always tagged with an authentic oracle
  4208 name.  The Isar command 'oracle' is now polymorphic, no argument type
  4209 is specified.  INCOMPATIBILITY, need to simplify existing oracle code
  4210 accordingly.  Note that extra performance may be gained by producing
  4211 the cterm carefully, avoiding slow Thm.cterm_of.
  4212 
  4213 * Simplified interface for defining document antiquotations via
  4214 ThyOutput.antiquotation, ThyOutput.output, and optionally
  4215 ThyOutput.maybe_pretty_source.  INCOMPATIBILITY, need to simplify user
  4216 antiquotations accordingly, see src/Pure/Thy/thy_output.ML for common
  4217 examples.
  4218 
  4219 * More systematic treatment of long names, abstract name bindings, and
  4220 name space operations.  Basic operations on qualified names have been
  4221 move from structure NameSpace to Long_Name, e.g. Long_Name.base_name,
  4222 Long_Name.append.  Old type bstring has been mostly replaced by
  4223 abstract type binding (see structure Binding), which supports precise
  4224 qualification by packages and local theory targets, as well as proper
  4225 tracking of source positions.  INCOMPATIBILITY, need to wrap old
  4226 bstring values into Binding.name, or better pass through abstract
  4227 bindings everywhere.  See further src/Pure/General/long_name.ML,
  4228 src/Pure/General/binding.ML and src/Pure/General/name_space.ML
  4229 
  4230 * Result facts (from PureThy.note_thms, ProofContext.note_thms,
  4231 LocalTheory.note etc.) now refer to the *full* internal name, not the
  4232 bstring as before.  INCOMPATIBILITY, not detected by ML type-checking!
  4233 
  4234 * Disposed old type and term read functions (Sign.read_def_typ,
  4235 Sign.read_typ, Sign.read_def_terms, Sign.read_term,
  4236 Thm.read_def_cterms, Thm.read_cterm etc.).  INCOMPATIBILITY, should
  4237 use regular Syntax.read_typ, Syntax.read_term, Syntax.read_typ_global,
  4238 Syntax.read_term_global etc.; see also OldGoals.read_term as last
  4239 resort for legacy applications.
  4240 
  4241 * Disposed old declarations, tactics, tactic combinators that refer to
  4242 the simpset or claset of an implicit theory (such as Addsimps,
  4243 Simp_tac, SIMPSET).  INCOMPATIBILITY, should use @{simpset} etc. in
  4244 embedded ML text, or local_simpset_of with a proper context passed as
  4245 explicit runtime argument.
  4246 
  4247 * Rules and tactics that read instantiations (read_instantiate,
  4248 res_inst_tac, thin_tac, subgoal_tac etc.) now demand a proper proof
  4249 context, which is required for parsing and type-checking.  Moreover,
  4250 the variables are specified as plain indexnames, not string encodings
  4251 thereof.  INCOMPATIBILITY.
  4252 
  4253 * Generic Toplevel.add_hook interface allows to analyze the result of
  4254 transactions.  E.g. see src/Pure/ProofGeneral/proof_general_pgip.ML
  4255 for theorem dependency output of transactions resulting in a new
  4256 theory state.
  4257 
  4258 * ML antiquotations: block-structured compilation context indicated by
  4259 \<lbrace> ... \<rbrace>; additional antiquotation forms:
  4260 
  4261   @{binding name}                         - basic name binding
  4262   @{let ?pat = term}                      - term abbreviation (HO matching)
  4263   @{note name = fact}                     - fact abbreviation
  4264   @{thm fact}                             - singleton fact (with attributes)
  4265   @{thms fact}                            - general fact (with attributes)
  4266   @{lemma prop by method}                 - singleton goal
  4267   @{lemma prop by meth1 meth2}            - singleton goal
  4268   @{lemma prop1 ... propN by method}      - general goal
  4269   @{lemma prop1 ... propN by meth1 meth2} - general goal
  4270   @{lemma (open) ...}                     - open derivation
  4271 
  4272 
  4273 *** System ***
  4274 
  4275 * The Isabelle "emacs" tool provides a specific interface to invoke
  4276 Proof General / Emacs, with more explicit failure if that is not
  4277 installed (the old isabelle-interface script silently falls back on
  4278 isabelle-process).  The PROOFGENERAL_HOME setting determines the
  4279 installation location of the Proof General distribution.
  4280 
  4281 * Isabelle/lib/classes/Pure.jar provides basic support to integrate
  4282 the Isabelle process into a JVM/Scala application.  See
  4283 Isabelle/lib/jedit/plugin for a minimal example.  (The obsolete Java
  4284 process wrapper has been discontinued.)
  4285 
  4286 * Added homegrown Isabelle font with unicode layout, see lib/fonts.
  4287 
  4288 * Various status messages (with exact source position information) are
  4289 emitted, if proper markup print mode is enabled.  This allows
  4290 user-interface components to provide detailed feedback on internal
  4291 prover operations.
  4292 
  4293 
  4294 
  4295 New in Isabelle2008 (June 2008)
  4296 -------------------------------
  4297 
  4298 *** General ***
  4299 
  4300 * The Isabelle/Isar Reference Manual (isar-ref) has been reorganized
  4301 and updated, with formally checked references as hyperlinks.
  4302 
  4303 * Theory loader: use_thy (and similar operations) no longer set the
  4304 implicit ML context, which was occasionally hard to predict and in
  4305 conflict with concurrency.  INCOMPATIBILITY, use ML within Isar which
  4306 provides a proper context already.
  4307 
  4308 * Theory loader: old-style ML proof scripts being *attached* to a thy
  4309 file are no longer supported.  INCOMPATIBILITY, regular 'uses' and
  4310 'use' within a theory file will do the job.
  4311 
  4312 * Name space merge now observes canonical order, i.e. the second space
  4313 is inserted into the first one, while existing entries in the first
  4314 space take precedence.  INCOMPATIBILITY in rare situations, may try to
  4315 swap theory imports.
  4316 
  4317 * Syntax: symbol \<chi> is now considered a letter.  Potential
  4318 INCOMPATIBILITY in identifier syntax etc.
  4319 
  4320 * Outer syntax: string tokens no longer admit escaped white space,
  4321 which was an accidental (undocumented) feature.  INCOMPATIBILITY, use
  4322 white space without escapes.
  4323 
  4324 * Outer syntax: string tokens may contain arbitrary character codes
  4325 specified via 3 decimal digits (as in SML).  E.g. "foo\095bar" for
  4326 "foo_bar".
  4327 
  4328 
  4329 *** Pure ***
  4330 
  4331 * Context-dependent token translations.  Default setup reverts locally
  4332 fixed variables, and adds hilite markup for undeclared frees.
  4333 
  4334 * Unused theorems can be found using the new command 'unused_thms'.
  4335 There are three ways of invoking it:
  4336 
  4337 (1) unused_thms
  4338      Only finds unused theorems in the current theory.
  4339 
  4340 (2) unused_thms thy_1 ... thy_n -
  4341      Finds unused theorems in the current theory and all of its ancestors,
  4342      excluding the theories thy_1 ... thy_n and all of their ancestors.
  4343 
  4344 (3) unused_thms thy_1 ... thy_n - thy'_1 ... thy'_m
  4345      Finds unused theorems in the theories thy'_1 ... thy'_m and all of
  4346      their ancestors, excluding the theories thy_1 ... thy_n and all of
  4347      their ancestors.
  4348 
  4349 In order to increase the readability of the list produced by
  4350 unused_thms, theorems that have been created by a particular instance
  4351 of a theory command such as 'inductive' or 'function' are considered
  4352 to belong to the same "group", meaning that if at least one theorem in
  4353 this group is used, the other theorems in the same group are no longer
  4354 reported as unused.  Moreover, if all theorems in the group are
  4355 unused, only one theorem in the group is displayed.
  4356 
  4357 Note that proof objects have to be switched on in order for
  4358 unused_thms to work properly (i.e. !proofs must be >= 1, which is
  4359 usually the case when using Proof General with the default settings).
  4360 
  4361 * Authentic naming of facts disallows ad-hoc overwriting of previous
  4362 theorems within the same name space.  INCOMPATIBILITY, need to remove
  4363 duplicate fact bindings, or even accidental fact duplications.  Note
  4364 that tools may maintain dynamically scoped facts systematically, using
  4365 PureThy.add_thms_dynamic.
  4366 
  4367 * Command 'hide' now allows to hide from "fact" name space as well.
  4368 
  4369 * Eliminated destructive theorem database, simpset, claset, and
  4370 clasimpset.  Potential INCOMPATIBILITY, really need to observe linear
  4371 update of theories within ML code.
  4372 
  4373 * Eliminated theory ProtoPure and CPure, leaving just one Pure theory.
  4374 INCOMPATIBILITY, object-logics depending on former Pure require
  4375 additional setup PureThy.old_appl_syntax_setup; object-logics
  4376 depending on former CPure need to refer to Pure.
  4377 
  4378 * Commands 'use' and 'ML' are now purely functional, operating on
  4379 theory/local_theory.  Removed former 'ML_setup' (on theory), use 'ML'
  4380 instead.  Added 'ML_val' as mere diagnostic replacement for 'ML'.
  4381 INCOMPATIBILITY.
  4382 
  4383 * Command 'setup': discontinued implicit version with ML reference.
  4384 
  4385 * Instantiation target allows for simultaneous specification of class
  4386 instance operations together with an instantiation proof.
  4387 Type-checking phase allows to refer to class operations uniformly.
  4388 See src/HOL/Complex/Complex.thy for an Isar example and
  4389 src/HOL/Library/Eval.thy for an ML example.
  4390 
  4391 * Indexing of literal facts: be more serious about including only
  4392 facts from the visible specification/proof context, but not the
  4393 background context (locale etc.).  Affects `prop` notation and method
  4394 "fact".  INCOMPATIBILITY: need to name facts explicitly in rare
  4395 situations.
  4396 
  4397 * Method "cases", "induct", "coinduct": removed obsolete/undocumented
  4398 "(open)" option, which used to expose internal bound variables to the
  4399 proof text.
  4400 
  4401 * Isar statements: removed obsolete case "rule_context".
  4402 INCOMPATIBILITY, better use explicit fixes/assumes.
  4403 
  4404 * Locale proofs: default proof step now includes 'unfold_locales';
  4405 hence 'proof' without argument may be used to unfold locale
  4406 predicates.
  4407 
  4408 
  4409 *** Document preparation ***
  4410 
  4411 * Simplified pdfsetup.sty: color/hyperref is used unconditionally for
  4412 both pdf and dvi (hyperlinks usually work in xdvi as well); removed
  4413 obsolete thumbpdf setup (contemporary PDF viewers do this on the
  4414 spot); renamed link color from "darkblue" to "linkcolor" (default
  4415 value unchanged, can be redefined via \definecolor); no longer sets
  4416 "a4paper" option (unnecessary or even intrusive).
  4417 
  4418 * Antiquotation @{lemma A method} proves proposition A by the given
  4419 method (either a method name or a method name plus (optional) method
  4420 arguments in parentheses) and prints A just like @{prop A}.
  4421 
  4422 
  4423 *** HOL ***
  4424 
  4425 * New primrec package.  Specification syntax conforms in style to
  4426 definition/function/....  No separate induction rule is provided.  The
  4427 "primrec" command distinguishes old-style and new-style specifications
  4428 by syntax.  The former primrec package is now named OldPrimrecPackage.
  4429 When adjusting theories, beware: constants stemming from new-style
  4430 primrec specifications have authentic syntax.
  4431 
  4432 * Metis prover is now an order of magnitude faster, and also works
  4433 with multithreading.
  4434 
  4435 * Metis: the maximum number of clauses that can be produced from a
  4436 theorem is now given by the attribute max_clauses.  Theorems that
  4437 exceed this number are ignored, with a warning printed.
  4438 
  4439 * Sledgehammer no longer produces structured proofs by default. To
  4440 enable, declare [[sledgehammer_full = true]].  Attributes
  4441 reconstruction_modulus, reconstruction_sorts renamed
  4442 sledgehammer_modulus, sledgehammer_sorts.  INCOMPATIBILITY.
  4443 
  4444 * Method "induct_scheme" derives user-specified induction rules
  4445 from well-founded induction and completeness of patterns. This factors
  4446 out some operations that are done internally by the function package
  4447 and makes them available separately.  See
  4448 src/HOL/ex/Induction_Scheme.thy for examples.
  4449 
  4450 * More flexible generation of measure functions for termination
  4451 proofs: Measure functions can be declared by proving a rule of the
  4452 form "is_measure f" and giving it the [measure_function] attribute.
  4453 The "is_measure" predicate is logically meaningless (always true), and
  4454 just guides the heuristic.  To find suitable measure functions, the
  4455 termination prover sets up the goal "is_measure ?f" of the appropriate
  4456 type and generates all solutions by prolog-style backwards proof using
  4457 the declared rules.
  4458 
  4459 This setup also deals with rules like 
  4460 
  4461   "is_measure f ==> is_measure (list_size f)"
  4462 
  4463 which accommodates nested datatypes that recurse through lists.
  4464 Similar rules are predeclared for products and option types.
  4465 
  4466 * Turned the type of sets "'a set" into an abbreviation for "'a => bool"
  4467 
  4468   INCOMPATIBILITIES:
  4469 
  4470   - Definitions of overloaded constants on sets have to be replaced by
  4471     definitions on => and bool.
  4472 
  4473   - Some definitions of overloaded operators on sets can now be proved
  4474     using the definitions of the operators on => and bool.  Therefore,
  4475     the following theorems have been renamed:
  4476 
  4477       subset_def   -> subset_eq
  4478       psubset_def  -> psubset_eq
  4479       set_diff_def -> set_diff_eq
  4480       Compl_def    -> Compl_eq
  4481       Sup_set_def  -> Sup_set_eq
  4482       Inf_set_def  -> Inf_set_eq
  4483       sup_set_def  -> sup_set_eq
  4484       inf_set_def  -> inf_set_eq
  4485 
  4486   - Due to the incompleteness of the HO unification algorithm, some
  4487     rules such as subst may require manual instantiation, if some of
  4488     the unknowns in the rule is a set.
  4489 
  4490   - Higher order unification and forward proofs:
  4491     The proof pattern
  4492 
  4493       have "P (S::'a set)" <...>
  4494       then have "EX S. P S" ..
  4495 
  4496     no longer works (due to the incompleteness of the HO unification
  4497     algorithm) and must be replaced by the pattern
  4498 
  4499       have "EX S. P S"
  4500       proof
  4501         show "P S" <...>
  4502       qed
  4503 
  4504   - Calculational reasoning with subst (or similar rules):
  4505     The proof pattern
  4506 
  4507       have "P (S::'a set)" <...>
  4508       also have "S = T" <...>
  4509       finally have "P T" .
  4510 
  4511     no longer works (for similar reasons as the previous example) and
  4512     must be replaced by something like
  4513 
  4514       have "P (S::'a set)" <...>
  4515       moreover have "S = T" <...>
  4516       ultimately have "P T" by simp
  4517 
  4518   - Tactics or packages written in ML code:
  4519     Code performing pattern matching on types via
  4520 
  4521       Type ("set", [T]) => ...
  4522 
  4523     must be rewritten. Moreover, functions like strip_type or
  4524     binder_types no longer return the right value when applied to a
  4525     type of the form
  4526 
  4527       T1 => ... => Tn => U => bool
  4528 
  4529     rather than
  4530 
  4531       T1 => ... => Tn => U set
  4532 
  4533 * Merged theories Wellfounded_Recursion, Accessible_Part and
  4534 Wellfounded_Relations to theory Wellfounded.
  4535 
  4536 * Explicit class "eq" for executable equality.  INCOMPATIBILITY.
  4537 
  4538 * Class finite no longer treats UNIV as class parameter.  Use class
  4539 enum from theory Library/Enum instead to achieve a similar effect.
  4540 INCOMPATIBILITY.
  4541 
  4542 * Theory List: rule list_induct2 now has explicitly named cases "Nil"
  4543 and "Cons".  INCOMPATIBILITY.
  4544 
  4545 * HOL (and FOL): renamed variables in rules imp_elim and swap.
  4546 Potential INCOMPATIBILITY.
  4547 
  4548 * Theory Product_Type: duplicated lemmas split_Pair_apply and
  4549 injective_fst_snd removed, use split_eta and prod_eqI instead.
  4550 Renamed upd_fst to apfst and upd_snd to apsnd.  INCOMPATIBILITY.
  4551 
  4552 * Theory Nat: removed redundant lemmas that merely duplicate lemmas of
  4553 the same name in theory Orderings:
  4554 
  4555   less_trans
  4556   less_linear
  4557   le_imp_less_or_eq
  4558   le_less_trans
  4559   less_le_trans
  4560   less_not_sym
  4561   less_asym
  4562 
  4563 Renamed less_imp_le to less_imp_le_nat, and less_irrefl to
  4564 less_irrefl_nat.  Potential INCOMPATIBILITY due to more general types
  4565 and different variable names.
  4566 
  4567 * Library/Option_ord.thy: Canonical order on option type.
  4568 
  4569 * Library/RBT.thy: Red-black trees, an efficient implementation of
  4570 finite maps.
  4571 
  4572 * Library/Countable.thy: Type class for countable types.
  4573 
  4574 * Theory Int: The representation of numerals has changed.  The infix
  4575 operator BIT and the bit datatype with constructors B0 and B1 have
  4576 disappeared.  INCOMPATIBILITY, use "Int.Bit0 x" and "Int.Bit1 y" in
  4577 place of "x BIT bit.B0" and "y BIT bit.B1", respectively.  Theorems
  4578 involving BIT, B0, or B1 have been renamed with "Bit0" or "Bit1"
  4579 accordingly.
  4580 
  4581 * Theory Nat: definition of <= and < on natural numbers no longer
  4582 depend on well-founded relations.  INCOMPATIBILITY.  Definitions
  4583 le_def and less_def have disappeared.  Consider lemmas not_less
  4584 [symmetric, where ?'a = nat] and less_eq [symmetric] instead.
  4585 
  4586 * Theory Finite_Set: locales ACf, ACe, ACIf, ACIfSL and ACIfSLlin
  4587 (whose purpose mainly is for various fold_set functionals) have been
  4588 abandoned in favor of the existing algebraic classes
  4589 ab_semigroup_mult, comm_monoid_mult, ab_semigroup_idem_mult,
  4590 lower_semilattice (resp. upper_semilattice) and linorder.
  4591 INCOMPATIBILITY.
  4592 
  4593 * Theory Transitive_Closure: induct and cases rules now declare proper
  4594 case_names ("base" and "step").  INCOMPATIBILITY.
  4595 
  4596 * Theorem Inductive.lfp_ordinal_induct generalized to complete
  4597 lattices.  The form set-specific version is available as
  4598 Inductive.lfp_ordinal_induct_set.
  4599 
  4600 * Renamed theorems "power.simps" to "power_int.simps".
  4601 INCOMPATIBILITY.
  4602 
  4603 * Class semiring_div provides basic abstract properties of semirings
  4604 with division and modulo operations.  Subsumes former class dvd_mod.
  4605 
  4606 * Merged theories IntDef, Numeral and IntArith into unified theory
  4607 Int.  INCOMPATIBILITY.
  4608 
  4609 * Theory Library/Code_Index: type "index" now represents natural
  4610 numbers rather than integers.  INCOMPATIBILITY.
  4611 
  4612 * New class "uminus" with operation "uminus" (split of from class
  4613 "minus" which now only has operation "minus", binary).
  4614 INCOMPATIBILITY.
  4615 
  4616 * Constants "card", "internal_split", "option_map" now with authentic
  4617 syntax.  INCOMPATIBILITY.
  4618 
  4619 * Definitions subset_def, psubset_def, set_diff_def, Compl_def,
  4620 le_bool_def, less_bool_def, le_fun_def, less_fun_def, inf_bool_def,
  4621 sup_bool_def, Inf_bool_def, Sup_bool_def, inf_fun_def, sup_fun_def,
  4622 Inf_fun_def, Sup_fun_def, inf_set_def, sup_set_def, Inf_set_def,
  4623 Sup_set_def, le_def, less_def, option_map_def now with object
  4624 equality.  INCOMPATIBILITY.
  4625 
  4626 * Records. Removed K_record, and replaced it by pure lambda term
  4627 %x. c. The simplifier setup is now more robust against eta expansion.
  4628 INCOMPATIBILITY: in cases explicitly referring to K_record.
  4629 
  4630 * Library/Multiset: {#a, b, c#} abbreviates {#a#} + {#b#} + {#c#}.
  4631 
  4632 * Library/ListVector: new theory of arithmetic vector operations.
  4633 
  4634 * Library/Order_Relation: new theory of various orderings as sets of
  4635 pairs.  Defines preorders, partial orders, linear orders and
  4636 well-orders on sets and on types.
  4637 
  4638 
  4639 *** ZF ***
  4640 
  4641 * Renamed some theories to allow to loading both ZF and HOL in the
  4642 same session:
  4643 
  4644   Datatype  -> Datatype_ZF
  4645   Inductive -> Inductive_ZF
  4646   Int       -> Int_ZF
  4647   IntDiv    -> IntDiv_ZF
  4648   Nat       -> Nat_ZF
  4649   List      -> List_ZF
  4650   Main      -> Main_ZF
  4651 
  4652 INCOMPATIBILITY: ZF theories that import individual theories below
  4653 Main might need to be adapted.  Regular theory Main is still
  4654 available, as trivial extension of Main_ZF.
  4655 
  4656 
  4657 *** ML ***
  4658 
  4659 * ML within Isar: antiquotation @{const name} or @{const
  4660 name(typargs)} produces statically-checked Const term.
  4661 
  4662 * Functor NamedThmsFun: data is available to the user as dynamic fact
  4663 (of the same name).  Removed obsolete print command.
  4664 
  4665 * Removed obsolete "use_legacy_bindings" function.
  4666 
  4667 * The ``print mode'' is now a thread-local value derived from a global
  4668 template (the former print_mode reference), thus access becomes
  4669 non-critical.  The global print_mode reference is for session
  4670 management only; user-code should use print_mode_value,
  4671 print_mode_active, PrintMode.setmp etc.  INCOMPATIBILITY.
  4672 
  4673 * Functions system/system_out provide a robust way to invoke external
  4674 shell commands, with propagation of interrupts (requires Poly/ML
  4675 5.2.1).  Do not use OS.Process.system etc. from the basis library!
  4676 
  4677 
  4678 *** System ***
  4679 
  4680 * Default settings: PROOFGENERAL_OPTIONS no longer impose xemacs ---
  4681 in accordance with Proof General 3.7, which prefers GNU emacs.
  4682 
  4683 * isatool tty runs Isabelle process with plain tty interaction;
  4684 optional line editor may be specified via ISABELLE_LINE_EDITOR
  4685 setting, the default settings attempt to locate "ledit" and "rlwrap".
  4686 
  4687 * isatool browser now works with Cygwin as well, using general
  4688 "javapath" function defined in Isabelle process environment.
  4689 
  4690 * YXML notation provides a simple and efficient alternative to
  4691 standard XML transfer syntax.  See src/Pure/General/yxml.ML and
  4692 isatool yxml as described in the Isabelle system manual.
  4693 
  4694 * JVM class isabelle.IsabelleProcess (located in Isabelle/lib/classes)
  4695 provides general wrapper for managing an Isabelle process in a robust
  4696 fashion, with ``cooked'' output from stdin/stderr.
  4697 
  4698 * Rudimentary Isabelle plugin for jEdit (see Isabelle/lib/jedit),
  4699 based on Isabelle/JVM process wrapper (see Isabelle/lib/classes).
  4700 
  4701 * Removed obsolete THIS_IS_ISABELLE_BUILD feature.  NB: the documented
  4702 way of changing the user's settings is via
  4703 ISABELLE_HOME_USER/etc/settings, which is a fully featured bash
  4704 script.
  4705 
  4706 * Multithreading.max_threads := 0 refers to the number of actual CPU
  4707 cores of the underlying machine, which is a good starting point for
  4708 optimal performance tuning.  The corresponding usedir option -M allows
  4709 "max" as an alias for "0".  WARNING: does not work on certain versions
  4710 of Mac OS (with Poly/ML 5.1).
  4711 
  4712 * isabelle-process: non-ML sessions are run with "nice", to reduce the
  4713 adverse effect of Isabelle flooding interactive front-ends (notably
  4714 ProofGeneral / XEmacs).
  4715 
  4716 
  4717 
  4718 New in Isabelle2007 (November 2007)
  4719 -----------------------------------
  4720 
  4721 *** General ***
  4722 
  4723 * More uniform information about legacy features, notably a
  4724 warning/error of "Legacy feature: ...", depending on the state of the
  4725 tolerate_legacy_features flag (default true). FUTURE INCOMPATIBILITY:
  4726 legacy features will disappear eventually.
  4727 
  4728 * Theory syntax: the header format ``theory A = B + C:'' has been
  4729 discontinued in favour of ``theory A imports B C begin''.  Use isatool
  4730 fixheaders to convert existing theory files.  INCOMPATIBILITY.
  4731 
  4732 * Theory syntax: the old non-Isar theory file format has been
  4733 discontinued altogether.  Note that ML proof scripts may still be used
  4734 with Isar theories; migration is usually quite simple with the ML
  4735 function use_legacy_bindings.  INCOMPATIBILITY.
  4736 
  4737 * Theory syntax: some popular names (e.g. 'class', 'declaration',
  4738 'fun', 'help', 'if') are now keywords.  INCOMPATIBILITY, use double
  4739 quotes.
  4740 
  4741 * Theory loader: be more serious about observing the static theory
  4742 header specifications (including optional directories), but not the
  4743 accidental file locations of previously successful loads.  The strict
  4744 update policy of former update_thy is now already performed by
  4745 use_thy, so the former has been removed; use_thys updates several
  4746 theories simultaneously, just as 'imports' within a theory header
  4747 specification, but without merging the results.  Potential
  4748 INCOMPATIBILITY: may need to refine theory headers and commands
  4749 ROOT.ML which depend on load order.
  4750 
  4751 * Theory loader: optional support for content-based file
  4752 identification, instead of the traditional scheme of full physical
  4753 path plus date stamp; configured by the ISABELLE_FILE_IDENT setting
  4754 (cf. the system manual).  The new scheme allows to work with
  4755 non-finished theories in persistent session images, such that source
  4756 files may be moved later on without requiring reloads.
  4757 
  4758 * Theory loader: old-style ML proof scripts being *attached* to a thy
  4759 file (with the same base name as the theory) are considered a legacy
  4760 feature, which will disappear eventually. Even now, the theory loader
  4761 no longer maintains dependencies on such files.
  4762 
  4763 * Syntax: the scope for resolving ambiguities via type-inference is
  4764 now limited to individual terms, instead of whole simultaneous
  4765 specifications as before. This greatly reduces the complexity of the
  4766 syntax module and improves flexibility by separating parsing and
  4767 type-checking. INCOMPATIBILITY: additional type-constraints (explicit
  4768 'fixes' etc.) are required in rare situations.
  4769 
  4770 * Syntax: constants introduced by new-style packages ('definition',
  4771 'abbreviation' etc.) are passed through the syntax module in
  4772 ``authentic mode''. This means that associated mixfix annotations
  4773 really stick to such constants, independently of potential name space
  4774 ambiguities introduced later on. INCOMPATIBILITY: constants in parse
  4775 trees are represented slightly differently, may need to adapt syntax
  4776 translations accordingly. Use CONST marker in 'translations' and
  4777 @{const_syntax} antiquotation in 'parse_translation' etc.
  4778 
  4779 * Legacy goal package: reduced interface to the bare minimum required
  4780 to keep existing proof scripts running.  Most other user-level
  4781 functions are now part of the OldGoals structure, which is *not* open
  4782 by default (consider isatool expandshort before open OldGoals).
  4783 Removed top_sg, prin, printyp, pprint_term/typ altogether, because
  4784 these tend to cause confusion about the actual goal (!) context being
  4785 used here, which is not necessarily the same as the_context().
  4786 
  4787 * Command 'find_theorems': supports "*" wild-card in "name:"
  4788 criterion; "with_dups" option.  Certain ProofGeneral versions might
  4789 support a specific search form (see ProofGeneral/CHANGES).
  4790 
  4791 * The ``prems limit'' option (cf. ProofContext.prems_limit) is now -1
  4792 by default, which means that "prems" (and also "fixed variables") are
  4793 suppressed from proof state output.  Note that the ProofGeneral
  4794 settings mechanism allows to change and save options persistently, but
  4795 older versions of Isabelle will fail to start up if a negative prems
  4796 limit is imposed.
  4797 
  4798 * Local theory targets may be specified by non-nested blocks of
  4799 ``context/locale/class ... begin'' followed by ``end''.  The body may
  4800 contain definitions, theorems etc., including any derived mechanism
  4801 that has been implemented on top of these primitives.  This concept
  4802 generalizes the existing ``theorem (in ...)'' towards more versatility
  4803 and scalability.
  4804 
  4805 * Proof General interface: proper undo of final 'end' command;
  4806 discontinued Isabelle/classic mode (ML proof scripts).
  4807 
  4808 
  4809 *** Document preparation ***
  4810 
  4811 * Added antiquotation @{theory name} which prints the given name,
  4812 after checking that it refers to a valid ancestor theory in the
  4813 current context.
  4814 
  4815 * Added antiquotations @{ML_type text} and @{ML_struct text} which
  4816 check the given source text as ML type/structure, printing verbatim.
  4817 
  4818 * Added antiquotation @{abbrev "c args"} which prints the abbreviation
  4819 "c args == rhs" given in the current context.  (Any number of
  4820 arguments may be given on the LHS.)
  4821 
  4822 
  4823 *** Pure ***
  4824 
  4825 * The 'class' package offers a combination of axclass and locale to
  4826 achieve Haskell-like type classes in Isabelle.  Definitions and
  4827 theorems within a class context produce both relative results (with
  4828 implicit parameters according to the locale context), and polymorphic
  4829 constants with qualified polymorphism (according to the class
  4830 context).  Within the body context of a 'class' target, a separate
  4831 syntax layer ("user space type system") takes care of converting
  4832 between global polymorphic consts and internal locale representation.
  4833 See src/HOL/ex/Classpackage.thy for examples (as well as main HOL).
  4834 "isatool doc classes" provides a tutorial.
  4835 
  4836 * Generic code generator framework allows to generate executable
  4837 code for ML and Haskell (including Isabelle classes).  A short usage
  4838 sketch:
  4839 
  4840     internal compilation:
  4841         export_code <list of constants (term syntax)> in SML
  4842     writing SML code to a file:
  4843         export_code <list of constants (term syntax)> in SML <filename>
  4844     writing OCaml code to a file:
  4845         export_code <list of constants (term syntax)> in OCaml <filename>
  4846     writing Haskell code to a bunch of files:
  4847         export_code <list of constants (term syntax)> in Haskell <filename>
  4848 
  4849     evaluating closed propositions to True/False using code generation:
  4850         method ``eval''
  4851 
  4852 Reasonable default setup of framework in HOL.
  4853 
  4854 Theorem attributs for selecting and transforming function equations theorems:
  4855 
  4856     [code fun]:        select a theorem as function equation for a specific constant
  4857     [code fun del]:    deselect a theorem as function equation for a specific constant
  4858     [code inline]:     select an equation theorem for unfolding (inlining) in place
  4859     [code inline del]: deselect an equation theorem for unfolding (inlining) in place
  4860 
  4861 User-defined serializations (target in {SML, OCaml, Haskell}):
  4862 
  4863     code_const <and-list of constants (term syntax)>
  4864       {(target) <and-list of const target syntax>}+
  4865 
  4866     code_type <and-list of type constructors>
  4867       {(target) <and-list of type target syntax>}+
  4868 
  4869     code_instance <and-list of instances>
  4870       {(target)}+
  4871         where instance ::= <type constructor> :: <class>
  4872 
  4873     code_class <and_list of classes>
  4874       {(target) <and-list of class target syntax>}+
  4875         where class target syntax ::= <class name> {where {<classop> == <target syntax>}+}?
  4876 
  4877 code_instance and code_class only are effective to target Haskell.
  4878 
  4879 For example usage see src/HOL/ex/Codegenerator.thy and
  4880 src/HOL/ex/Codegenerator_Pretty.thy.  A separate tutorial on code
  4881 generation from Isabelle/HOL theories is available via "isatool doc
  4882 codegen".
  4883 
  4884 * Code generator: consts in 'consts_code' Isar commands are now
  4885 referred to by usual term syntax (including optional type
  4886 annotations).
  4887 
  4888 * Command 'no_translations' removes translation rules from theory
  4889 syntax.
  4890 
  4891 * Overloaded definitions are now actually checked for acyclic
  4892 dependencies.  The overloading scheme is slightly more general than
  4893 that of Haskell98, although Isabelle does not demand an exact
  4894 correspondence to type class and instance declarations.
  4895 INCOMPATIBILITY, use ``defs (unchecked overloaded)'' to admit more
  4896 exotic versions of overloading -- at the discretion of the user!
  4897 
  4898 Polymorphic constants are represented via type arguments, i.e. the
  4899 instantiation that matches an instance against the most general
  4900 declaration given in the signature.  For example, with the declaration
  4901 c :: 'a => 'a => 'a, an instance c :: nat => nat => nat is represented
  4902 as c(nat).  Overloading is essentially simultaneous structural
  4903 recursion over such type arguments.  Incomplete specification patterns
  4904 impose global constraints on all occurrences, e.g. c('a * 'a) on the
  4905 LHS means that more general c('a * 'b) will be disallowed on any RHS.
  4906 Command 'print_theory' outputs the normalized system of recursive
  4907 equations, see section "definitions".
  4908 
  4909 * Configuration options are maintained within the theory or proof
  4910 context (with name and type bool/int/string), providing a very simple
  4911 interface to a poor-man's version of general context data.  Tools may
  4912 declare options in ML (e.g. using Attrib.config_int) and then refer to
  4913 these values using Config.get etc.  Users may change options via an
  4914 associated attribute of the same name.  This form of context
  4915 declaration works particularly well with commands 'declare' or
  4916 'using', for example ``declare [[foo = 42]]''.  Thus it has become
  4917 very easy to avoid global references, which would not observe Isar
  4918 toplevel undo/redo and fail to work with multithreading.
  4919 
  4920 Various global ML references of Pure and HOL have been turned into
  4921 configuration options:
  4922 
  4923   Unify.search_bound		unify_search_bound
  4924   Unify.trace_bound		unify_trace_bound
  4925   Unify.trace_simp		unify_trace_simp
  4926   Unify.trace_types		unify_trace_types
  4927   Simplifier.simp_depth_limit	simp_depth_limit
  4928   Blast.depth_limit		blast_depth_limit
  4929   DatatypeProp.dtK		datatype_distinctness_limit
  4930   fast_arith_neq_limit  	fast_arith_neq_limit
  4931   fast_arith_split_limit	fast_arith_split_limit
  4932 
  4933 * Named collections of theorems may be easily installed as context
  4934 data using the functor NamedThmsFun (see also
  4935 src/Pure/Tools/named_thms.ML).  The user may add or delete facts via
  4936 attributes; there is also a toplevel print command.  This facility is
  4937 just a common case of general context data, which is the preferred way
  4938 for anything more complex than just a list of facts in canonical
  4939 order.
  4940 
  4941 * Isar: command 'declaration' augments a local theory by generic
  4942 declaration functions written in ML.  This enables arbitrary content
  4943 being added to the context, depending on a morphism that tells the
  4944 difference of the original declaration context wrt. the application
  4945 context encountered later on.
  4946 
  4947 * Isar: proper interfaces for simplification procedures.  Command
  4948 'simproc_setup' declares named simprocs (with match patterns, and body
  4949 text in ML).  Attribute "simproc" adds/deletes simprocs in the current
  4950 context.  ML antiquotation @{simproc name} retrieves named simprocs.
  4951 
  4952 * Isar: an extra pair of brackets around attribute declarations
  4953 abbreviates a theorem reference involving an internal dummy fact,
  4954 which will be ignored later --- only the effect of the attribute on
  4955 the background context will persist.  This form of in-place
  4956 declarations is particularly useful with commands like 'declare' and
  4957 'using', for example ``have A using [[simproc a]] by simp''.
  4958 
  4959 * Isar: method "assumption" (and implicit closing of subproofs) now
  4960 takes simple non-atomic goal assumptions into account: after applying
  4961 an assumption as a rule the resulting subgoals are solved by atomic
  4962 assumption steps.  This is particularly useful to finish 'obtain'
  4963 goals, such as "!!x. (!!x. P x ==> thesis) ==> P x ==> thesis",
  4964 without referring to the original premise "!!x. P x ==> thesis" in the
  4965 Isar proof context.  POTENTIAL INCOMPATIBILITY: method "assumption" is
  4966 more permissive.
  4967 
  4968 * Isar: implicit use of prems from the Isar proof context is
  4969 considered a legacy feature.  Common applications like ``have A .''
  4970 may be replaced by ``have A by fact'' or ``note `A`''.  In general,
  4971 referencing facts explicitly here improves readability and
  4972 maintainability of proof texts.
  4973 
  4974 * Isar: improper proof element 'guess' is like 'obtain', but derives
  4975 the obtained context from the course of reasoning!  For example:
  4976 
  4977   assume "EX x y. A x & B y"   -- "any previous fact"
  4978   then guess x and y by clarify
  4979 
  4980 This technique is potentially adventurous, depending on the facts and
  4981 proof tools being involved here.
  4982 
  4983 * Isar: known facts from the proof context may be specified as literal
  4984 propositions, using ASCII back-quote syntax.  This works wherever
  4985 named facts used to be allowed so far, in proof commands, proof
  4986 methods, attributes etc.  Literal facts are retrieved from the context
  4987 according to unification of type and term parameters.  For example,
  4988 provided that "A" and "A ==> B" and "!!x. P x ==> Q x" are known
  4989 theorems in the current context, then these are valid literal facts:
  4990 `A` and `A ==> B` and `!!x. P x ==> Q x" as well as `P a ==> Q a` etc.
  4991 
  4992 There is also a proof method "fact" which does the same composition
  4993 for explicit goal states, e.g. the following proof texts coincide with
  4994 certain special cases of literal facts:
  4995 
  4996   have "A" by fact                 ==  note `A`
  4997   have "A ==> B" by fact           ==  note `A ==> B`
  4998   have "!!x. P x ==> Q x" by fact  ==  note `!!x. P x ==> Q x`
  4999   have "P a ==> Q a" by fact       ==  note `P a ==> Q a`
  5000 
  5001 * Isar: ":" (colon) is no longer a symbolic identifier character in
  5002 outer syntax.  Thus symbolic identifiers may be used without
  5003 additional white space in declarations like this: ``assume *: A''.
  5004 
  5005 * Isar: 'print_facts' prints all local facts of the current context,
  5006 both named and unnamed ones.
  5007 
  5008 * Isar: 'def' now admits simultaneous definitions, e.g.:
  5009 
  5010   def x == "t" and y == "u"
  5011 
  5012 * Isar: added command 'unfolding', which is structurally similar to
  5013 'using', but affects both the goal state and facts by unfolding given
  5014 rewrite rules.  Thus many occurrences of the 'unfold' method or
  5015 'unfolded' attribute may be replaced by first-class proof text.
  5016 
  5017 * Isar: methods 'unfold' / 'fold', attributes 'unfolded' / 'folded',
  5018 and command 'unfolding' now all support object-level equalities
  5019 (potentially conditional).  The underlying notion of rewrite rule is
  5020 analogous to the 'rule_format' attribute, but *not* that of the
  5021 Simplifier (which is usually more generous).
  5022 
  5023 * Isar: the new attribute [rotated n] (default n = 1) rotates the
  5024 premises of a theorem by n. Useful in conjunction with drule.
  5025 
  5026 * Isar: the goal restriction operator [N] (default N = 1) evaluates a
  5027 method expression within a sandbox consisting of the first N
  5028 sub-goals, which need to exist.  For example, ``simp_all [3]''
  5029 simplifies the first three sub-goals, while (rule foo, simp_all)[]
  5030 simplifies all new goals that emerge from applying rule foo to the
  5031 originally first one.
  5032 
  5033 * Isar: schematic goals are no longer restricted to higher-order
  5034 patterns; e.g. ``lemma "?P(?x)" by (rule TrueI)'' now works as
  5035 expected.
  5036 
  5037 * Isar: the conclusion of a long theorem statement is now either
  5038 'shows' (a simultaneous conjunction, as before), or 'obtains'
  5039 (essentially a disjunction of cases with local parameters and
  5040 assumptions).  The latter allows to express general elimination rules
  5041 adequately; in this notation common elimination rules look like this:
  5042 
  5043   lemma exE:    -- "EX x. P x ==> (!!x. P x ==> thesis) ==> thesis"
  5044     assumes "EX x. P x"
  5045     obtains x where "P x"
  5046 
  5047   lemma conjE:  -- "A & B ==> (A ==> B ==> thesis) ==> thesis"
  5048     assumes "A & B"
  5049     obtains A and B
  5050 
  5051   lemma disjE:  -- "A | B ==> (A ==> thesis) ==> (B ==> thesis) ==> thesis"
  5052     assumes "A | B"
  5053     obtains
  5054       A
  5055     | B
  5056 
  5057 The subsequent classical rules even refer to the formal "thesis"
  5058 explicitly:
  5059 
  5060   lemma classical:     -- "(~ thesis ==> thesis) ==> thesis"
  5061     obtains "~ thesis"
  5062 
  5063   lemma Peirce's_Law:  -- "((thesis ==> something) ==> thesis) ==> thesis"
  5064     obtains "thesis ==> something"
  5065 
  5066 The actual proof of an 'obtains' statement is analogous to that of the
  5067 Isar proof element 'obtain', only that there may be several cases.
  5068 Optional case names may be specified in parentheses; these will be
  5069 available both in the present proof and as annotations in the
  5070 resulting rule, for later use with the 'cases' method (cf. attribute
  5071 case_names).
  5072 
  5073 * Isar: the assumptions of a long theorem statement are available as
  5074 "assms" fact in the proof context.  This is more appropriate than the
  5075 (historical) "prems", which refers to all assumptions of the current
  5076 context, including those from the target locale, proof body etc.
  5077 
  5078 * Isar: 'print_statement' prints theorems from the current theory or
  5079 proof context in long statement form, according to the syntax of a
  5080 top-level lemma.
  5081 
  5082 * Isar: 'obtain' takes an optional case name for the local context
  5083 introduction rule (default "that").
  5084 
  5085 * Isar: removed obsolete 'concl is' patterns.  INCOMPATIBILITY, use
  5086 explicit (is "_ ==> ?foo") in the rare cases where this still happens
  5087 to occur.
  5088 
  5089 * Pure: syntax "CONST name" produces a fully internalized constant
  5090 according to the current context.  This is particularly useful for
  5091 syntax translations that should refer to internal constant
  5092 representations independently of name spaces.
  5093 
  5094 * Pure: syntax constant for foo (binder "FOO ") is called "foo_binder"
  5095 instead of "FOO ". This allows multiple binder declarations to coexist
  5096 in the same context.  INCOMPATIBILITY.
  5097 
  5098 * Isar/locales: 'notation' provides a robust interface to the 'syntax'
  5099 primitive that also works in a locale context (both for constants and
  5100 fixed variables). Type declaration and internal syntactic representation
  5101 of given constants retrieved from the context. Likewise, the
  5102 'no_notation' command allows to remove given syntax annotations from the
  5103 current context.
  5104 
  5105 * Isar/locales: new derived specification elements 'axiomatization',
  5106 'definition', 'abbreviation', which support type-inference, admit
  5107 object-level specifications (equality, equivalence).  See also the
  5108 isar-ref manual.  Examples:
  5109 
  5110   axiomatization
  5111     eq  (infix "===" 50) where
  5112     eq_refl: "x === x" and eq_subst: "x === y ==> P x ==> P y"
  5113 
  5114   definition "f x y = x + y + 1"
  5115   definition g where "g x = f x x"
  5116 
  5117   abbreviation
  5118     neq  (infix "=!=" 50) where
  5119     "x =!= y == ~ (x === y)"
  5120 
  5121 These specifications may be also used in a locale context.  Then the
  5122 constants being introduced depend on certain fixed parameters, and the
  5123 constant name is qualified by the locale base name.  An internal
  5124 abbreviation takes care for convenient input and output, making the
  5125 parameters implicit and using the original short name.  See also
  5126 src/HOL/ex/Abstract_NAT.thy for an example of deriving polymorphic
  5127 entities from a monomorphic theory.
  5128 
  5129 Presently, abbreviations are only available 'in' a target locale, but
  5130 not inherited by general import expressions.  Also note that
  5131 'abbreviation' may be used as a type-safe replacement for 'syntax' +
  5132 'translations' in common applications.  The "no_abbrevs" print mode
  5133 prevents folding of abbreviations in term output.
  5134 
  5135 Concrete syntax is attached to specified constants in internal form,
  5136 independently of name spaces.  The parse tree representation is
  5137 slightly different -- use 'notation' instead of raw 'syntax', and
  5138 'translations' with explicit "CONST" markup to accommodate this.
  5139 
  5140 * Pure/Isar: unified syntax for new-style specification mechanisms
  5141 (e.g.  'definition', 'abbreviation', or 'inductive' in HOL) admits
  5142 full type inference and dummy patterns ("_").  For example:
  5143 
  5144   definition "K x _ = x"
  5145 
  5146   inductive conj for A B
  5147   where "A ==> B ==> conj A B"
  5148 
  5149 * Pure: command 'print_abbrevs' prints all constant abbreviations of
  5150 the current context.  Print mode "no_abbrevs" prevents inversion of
  5151 abbreviations on output.
  5152 
  5153 * Isar/locales: improved parameter handling: use of locales "var" and
  5154 "struct" no longer necessary; - parameter renamings are no longer
  5155 required to be injective.  For example, this allows to define
  5156 endomorphisms as locale endom = homom mult mult h.
  5157 
  5158 * Isar/locales: changed the way locales with predicates are defined.
  5159 Instead of accumulating the specification, the imported expression is
  5160 now an interpretation.  INCOMPATIBILITY: different normal form of
  5161 locale expressions.  In particular, in interpretations of locales with
  5162 predicates, goals repesenting already interpreted fragments are not
  5163 removed automatically.  Use methods `intro_locales' and
  5164 `unfold_locales'; see below.
  5165 
  5166 * Isar/locales: new methods `intro_locales' and `unfold_locales'
  5167 provide backward reasoning on locales predicates.  The methods are
  5168 aware of interpretations and discharge corresponding goals.
  5169 `intro_locales' is less aggressive then `unfold_locales' and does not
  5170 unfold predicates to assumptions.
  5171 
  5172 * Isar/locales: the order in which locale fragments are accumulated
  5173 has changed.  This enables to override declarations from fragments due
  5174 to interpretations -- for example, unwanted simp rules.
  5175 
  5176 * Isar/locales: interpretation in theories and proof contexts has been
  5177 extended.  One may now specify (and prove) equations, which are
  5178 unfolded in interpreted theorems.  This is useful for replacing
  5179 defined concepts (constants depending on locale parameters) by
  5180 concepts already existing in the target context.  Example:
  5181 
  5182   interpretation partial_order ["op <= :: [int, int] => bool"]
  5183     where "partial_order.less (op <=) (x::int) y = (x < y)"
  5184 
  5185 Typically, the constant `partial_order.less' is created by a
  5186 definition specification element in the context of locale
  5187 partial_order.
  5188 
  5189 * Method "induct": improved internal context management to support
  5190 local fixes and defines on-the-fly. Thus explicit meta-level
  5191 connectives !!  and ==> are rarely required anymore in inductive goals
  5192 (using object-logic connectives for this purpose has been long
  5193 obsolete anyway). Common proof patterns are explained in
  5194 src/HOL/Induct/Common_Patterns.thy, see also
  5195 src/HOL/Isar_examples/Puzzle.thy and src/HOL/Lambda for realistic
  5196 examples.
  5197 
  5198 * Method "induct": improved handling of simultaneous goals. Instead of
  5199 introducing object-level conjunction, the statement is now split into
  5200 several conclusions, while the corresponding symbolic cases are nested
  5201 accordingly. INCOMPATIBILITY, proofs need to be structured explicitly,
  5202 see src/HOL/Induct/Common_Patterns.thy, for example.
  5203 
  5204 * Method "induct": mutual induction rules are now specified as a list
  5205 of rule sharing the same induction cases. HOL packages usually provide
  5206 foo_bar.inducts for mutually defined items foo and bar (e.g. inductive
  5207 predicates/sets or datatypes). INCOMPATIBILITY, users need to specify
  5208 mutual induction rules differently, i.e. like this:
  5209 
  5210   (induct rule: foo_bar.inducts)
  5211   (induct set: foo bar)
  5212   (induct pred: foo bar)
  5213   (induct type: foo bar)
  5214 
  5215 The ML function ProjectRule.projections turns old-style rules into the
  5216 new format.
  5217 
  5218 * Method "coinduct": dual of induction, see
  5219 src/HOL/Library/Coinductive_List.thy for various examples.
  5220 
  5221 * Method "cases", "induct", "coinduct": the ``(open)'' option is
  5222 considered a legacy feature.
  5223 
  5224 * Attribute "symmetric" produces result with standardized schematic
  5225 variables (index 0).  Potential INCOMPATIBILITY.
  5226 
  5227 * Simplifier: by default the simplifier trace only shows top level
  5228 rewrites now. That is, trace_simp_depth_limit is set to 1 by
  5229 default. Thus there is less danger of being flooded by the trace. The
  5230 trace indicates where parts have been suppressed.
  5231   
  5232 * Provers/classical: removed obsolete classical version of elim_format
  5233 attribute; classical elim/dest rules are now treated uniformly when
  5234 manipulating the claset.
  5235 
  5236 * Provers/classical: stricter checks to ensure that supplied intro,
  5237 dest and elim rules are well-formed; dest and elim rules must have at
  5238 least one premise.
  5239 
  5240 * Provers/classical: attributes dest/elim/intro take an optional
  5241 weight argument for the rule (just as the Pure versions).  Weights are
  5242 ignored by automated tools, but determine the search order of single
  5243 rule steps.
  5244 
  5245 * Syntax: input syntax now supports dummy variable binding "%_. b",
  5246 where the body does not mention the bound variable.  Note that dummy
  5247 patterns implicitly depend on their context of bounds, which makes
  5248 "{_. _}" match any set comprehension as expected.  Potential
  5249 INCOMPATIBILITY -- parse translations need to cope with syntactic
  5250 constant "_idtdummy" in the binding position.
  5251 
  5252 * Syntax: removed obsolete syntactic constant "_K" and its associated
  5253 parse translation.  INCOMPATIBILITY -- use dummy abstraction instead,
  5254 for example "A -> B" => "Pi A (%_. B)".
  5255 
  5256 * Pure: 'class_deps' command visualizes the subclass relation, using
  5257 the graph browser tool.
  5258 
  5259 * Pure: 'print_theory' now suppresses certain internal declarations by
  5260 default; use '!' option for full details.
  5261 
  5262 
  5263 *** HOL ***
  5264 
  5265 * Method "metis" proves goals by applying the Metis general-purpose
  5266 resolution prover (see also http://gilith.com/software/metis/).
  5267 Examples are in the directory MetisExamples.  WARNING: the
  5268 Isabelle/HOL-Metis integration does not yet work properly with
  5269 multi-threading.
  5270   
  5271 * Command 'sledgehammer' invokes external automatic theorem provers as
  5272 background processes.  It generates calls to the "metis" method if
  5273 successful. These can be pasted into the proof.  Users do not have to
  5274 wait for the automatic provers to return.  WARNING: does not really
  5275 work with multi-threading.
  5276 
  5277 * New "auto_quickcheck" feature tests outermost goal statements for
  5278 potential counter-examples.  Controlled by ML references
  5279 auto_quickcheck (default true) and auto_quickcheck_time_limit (default
  5280 5000 milliseconds).  Fails silently if statements is outside of
  5281 executable fragment, or any other codgenerator problem occurs.
  5282 
  5283 * New constant "undefined" with axiom "undefined x = undefined".
  5284 
  5285 * Added class "HOL.eq", allowing for code generation with polymorphic
  5286 equality.
  5287 
  5288 * Some renaming of class constants due to canonical name prefixing in
  5289 the new 'class' package:
  5290 
  5291     HOL.abs ~> HOL.abs_class.abs
  5292     HOL.divide ~> HOL.divide_class.divide
  5293     0 ~> HOL.zero_class.zero
  5294     1 ~> HOL.one_class.one
  5295     op + ~> HOL.plus_class.plus
  5296     op - ~> HOL.minus_class.minus
  5297     uminus ~> HOL.minus_class.uminus
  5298     op * ~> HOL.times_class.times
  5299     op < ~> HOL.ord_class.less
  5300     op <= > HOL.ord_class.less_eq
  5301     Nat.power ~> Power.power_class.power
  5302     Nat.size ~> Nat.size_class.size
  5303     Numeral.number_of ~> Numeral.number_class.number_of
  5304     FixedPoint.Inf ~> Lattices.complete_lattice_class.Inf
  5305     FixedPoint.Sup ~> Lattices.complete_lattice_class.Sup
  5306     Orderings.min ~> Orderings.ord_class.min
  5307     Orderings.max ~> Orderings.ord_class.max
  5308     Divides.op div ~> Divides.div_class.div
  5309     Divides.op mod ~> Divides.div_class.mod
  5310     Divides.op dvd ~> Divides.div_class.dvd
  5311 
  5312 INCOMPATIBILITY.  Adaptions may be required in the following cases:
  5313 
  5314 a) User-defined constants using any of the names "plus", "minus",
  5315 "times", "less" or "less_eq". The standard syntax translations for
  5316 "+", "-" and "*" may go wrong.  INCOMPATIBILITY: use more specific
  5317 names.
  5318 
  5319 b) Variables named "plus", "minus", "times", "less", "less_eq"
  5320 INCOMPATIBILITY: use more specific names.
  5321 
  5322 c) Permutative equations (e.g. "a + b = b + a")
  5323 Since the change of names also changes the order of terms, permutative
  5324 rewrite rules may get applied in a different order. Experience shows
  5325 that this is rarely the case (only two adaptions in the whole Isabelle
  5326 distribution).  INCOMPATIBILITY: rewrite proofs
  5327 
  5328 d) ML code directly refering to constant names
  5329 This in general only affects hand-written proof tactics, simprocs and
  5330 so on.  INCOMPATIBILITY: grep your sourcecode and replace names.
  5331 Consider using @{const_name} antiquotation.
  5332 
  5333 * New class "default" with associated constant "default".
  5334 
  5335 * Function "sgn" is now overloaded and available on int, real, complex
  5336 (and other numeric types), using class "sgn".  Two possible defs of
  5337 sgn are given as equational assumptions in the classes sgn_if and
  5338 sgn_div_norm; ordered_idom now also inherits from sgn_if.
  5339 INCOMPATIBILITY.
  5340 
  5341 * Locale "partial_order" now unified with class "order" (cf. theory
  5342 Orderings), added parameter "less".  INCOMPATIBILITY.
  5343 
  5344 * Renamings in classes "order" and "linorder": facts "refl", "trans" and
  5345 "cases" to "order_refl", "order_trans" and "linorder_cases", to avoid
  5346 clashes with HOL "refl" and "trans".  INCOMPATIBILITY.
  5347 
  5348 * Classes "order" and "linorder": potential INCOMPATIBILITY due to
  5349 changed order of proof goals in instance proofs.
  5350 
  5351 * The transitivity reasoner for partial and linear orders is set up
  5352 for classes "order" and "linorder".  Instances of the reasoner are available
  5353 in all contexts importing or interpreting the corresponding locales.
  5354 Method "order" invokes the reasoner separately; the reasoner
  5355 is also integrated with the Simplifier as a solver.  Diagnostic
  5356 command 'print_orders' shows the available instances of the reasoner
  5357 in the current context.
  5358 
  5359 * Localized monotonicity predicate in theory "Orderings"; integrated
  5360 lemmas max_of_mono and min_of_mono with this predicate.
  5361 INCOMPATIBILITY.
  5362 
  5363 * Formulation of theorem "dense" changed slightly due to integration
  5364 with new class dense_linear_order.
  5365 
  5366 * Uniform lattice theory development in HOL.
  5367 
  5368     constants "meet" and "join" now named "inf" and "sup"
  5369     constant "Meet" now named "Inf"
  5370 
  5371     classes "meet_semilorder" and "join_semilorder" now named
  5372       "lower_semilattice" and "upper_semilattice"
  5373     class "lorder" now named "lattice"
  5374     class "comp_lat" now named "complete_lattice"
  5375 
  5376     Instantiation of lattice classes allows explicit definitions
  5377     for "inf" and "sup" operations (or "Inf" and "Sup" for complete lattices).
  5378 
  5379   INCOMPATIBILITY.  Theorem renames:
  5380 
  5381     meet_left_le            ~> inf_le1
  5382     meet_right_le           ~> inf_le2
  5383     join_left_le            ~> sup_ge1
  5384     join_right_le           ~> sup_ge2
  5385     meet_join_le            ~> inf_sup_ord
  5386     le_meetI                ~> le_infI
  5387     join_leI                ~> le_supI
  5388     le_meet                 ~> le_inf_iff
  5389     le_join                 ~> ge_sup_conv
  5390     meet_idempotent         ~> inf_idem
  5391     join_idempotent         ~> sup_idem
  5392     meet_comm               ~> inf_commute
  5393     join_comm               ~> sup_commute
  5394     meet_leI1               ~> le_infI1
  5395     meet_leI2               ~> le_infI2
  5396     le_joinI1               ~> le_supI1
  5397     le_joinI2               ~> le_supI2
  5398     meet_assoc              ~> inf_assoc
  5399     join_assoc              ~> sup_assoc
  5400     meet_left_comm          ~> inf_left_commute
  5401     meet_left_idempotent    ~> inf_left_idem
  5402     join_left_comm          ~> sup_left_commute
  5403     join_left_idempotent    ~> sup_left_idem
  5404     meet_aci                ~> inf_aci
  5405     join_aci                ~> sup_aci
  5406     le_def_meet             ~> le_iff_inf
  5407     le_def_join             ~> le_iff_sup
  5408     join_absorp2            ~> sup_absorb2
  5409     join_absorp1            ~> sup_absorb1
  5410     meet_absorp1            ~> inf_absorb1
  5411     meet_absorp2            ~> inf_absorb2
  5412     meet_join_absorp        ~> inf_sup_absorb
  5413     join_meet_absorp        ~> sup_inf_absorb
  5414     distrib_join_le         ~> distrib_sup_le
  5415     distrib_meet_le         ~> distrib_inf_le
  5416 
  5417     add_meet_distrib_left   ~> add_inf_distrib_left
  5418     add_join_distrib_left   ~> add_sup_distrib_left
  5419     is_join_neg_meet        ~> is_join_neg_inf
  5420     is_meet_neg_join        ~> is_meet_neg_sup
  5421     add_meet_distrib_right  ~> add_inf_distrib_right
  5422     add_join_distrib_right  ~> add_sup_distrib_right
  5423     add_meet_join_distribs  ~> add_sup_inf_distribs
  5424     join_eq_neg_meet        ~> sup_eq_neg_inf
  5425     meet_eq_neg_join        ~> inf_eq_neg_sup
  5426     add_eq_meet_join        ~> add_eq_inf_sup
  5427     meet_0_imp_0            ~> inf_0_imp_0
  5428     join_0_imp_0            ~> sup_0_imp_0
  5429     meet_0_eq_0             ~> inf_0_eq_0
  5430     join_0_eq_0             ~> sup_0_eq_0
  5431     neg_meet_eq_join        ~> neg_inf_eq_sup
  5432     neg_join_eq_meet        ~> neg_sup_eq_inf
  5433     join_eq_if              ~> sup_eq_if
  5434 
  5435     mono_meet               ~> mono_inf
  5436     mono_join               ~> mono_sup
  5437     meet_bool_eq            ~> inf_bool_eq
  5438     join_bool_eq            ~> sup_bool_eq
  5439     meet_fun_eq             ~> inf_fun_eq
  5440     join_fun_eq             ~> sup_fun_eq
  5441     meet_set_eq             ~> inf_set_eq
  5442     join_set_eq             ~> sup_set_eq
  5443     meet1_iff               ~> inf1_iff
  5444     meet2_iff               ~> inf2_iff
  5445     meet1I                  ~> inf1I
  5446     meet2I                  ~> inf2I
  5447     meet1D1                 ~> inf1D1
  5448     meet2D1                 ~> inf2D1
  5449     meet1D2                 ~> inf1D2
  5450     meet2D2                 ~> inf2D2
  5451     meet1E                  ~> inf1E
  5452     meet2E                  ~> inf2E
  5453     join1_iff               ~> sup1_iff
  5454     join2_iff               ~> sup2_iff
  5455     join1I1                 ~> sup1I1
  5456     join2I1                 ~> sup2I1
  5457     join1I1                 ~> sup1I1
  5458     join2I2                 ~> sup1I2
  5459     join1CI                 ~> sup1CI
  5460     join2CI                 ~> sup2CI
  5461     join1E                  ~> sup1E
  5462     join2E                  ~> sup2E
  5463 
  5464     is_meet_Meet            ~> is_meet_Inf
  5465     Meet_bool_def           ~> Inf_bool_def
  5466     Meet_fun_def            ~> Inf_fun_def
  5467     Meet_greatest           ~> Inf_greatest
  5468     Meet_lower              ~> Inf_lower
  5469     Meet_set_def            ~> Inf_set_def
  5470 
  5471     Sup_def                 ~> Sup_Inf
  5472     Sup_bool_eq             ~> Sup_bool_def
  5473     Sup_fun_eq              ~> Sup_fun_def
  5474     Sup_set_eq              ~> Sup_set_def
  5475 
  5476     listsp_meetI            ~> listsp_infI
  5477     listsp_meet_eq          ~> listsp_inf_eq
  5478 
  5479     meet_min                ~> inf_min
  5480     join_max                ~> sup_max
  5481 
  5482 * Added syntactic class "size"; overloaded constant "size" now has
  5483 type "'a::size ==> bool"
  5484 
  5485 * Internal reorganisation of `size' of datatypes: size theorems
  5486 "foo.size" are no longer subsumed by "foo.simps" (but are still
  5487 simplification rules by default!); theorems "prod.size" now named
  5488 "*.size".
  5489 
  5490 * Class "div" now inherits from class "times" rather than "type".
  5491 INCOMPATIBILITY.
  5492 
  5493 * HOL/Finite_Set: "name-space" locales Lattice, Distrib_lattice,
  5494 Linorder etc.  have disappeared; operations defined in terms of
  5495 fold_set now are named Inf_fin, Sup_fin.  INCOMPATIBILITY.
  5496 
  5497 * HOL/Nat: neq0_conv no longer declared as iff.  INCOMPATIBILITY.
  5498 
  5499 * HOL-Word: New extensive library and type for generic, fixed size
  5500 machine words, with arithmetic, bit-wise, shifting and rotating
  5501 operations, reflection into int, nat, and bool lists, automation for
  5502 linear arithmetic (by automatic reflection into nat or int), including
  5503 lemmas on overflow and monotonicity.  Instantiated to all appropriate
  5504 arithmetic type classes, supporting automatic simplification of
  5505 numerals on all operations.
  5506 
  5507 * Library/Boolean_Algebra: locales for abstract boolean algebras.
  5508 
  5509 * Library/Numeral_Type: numbers as types, e.g. TYPE(32).
  5510 
  5511 * Code generator library theories:
  5512   - Code_Integer represents HOL integers by big integer literals in target
  5513     languages.
  5514   - Code_Char represents HOL characters by character literals in target
  5515     languages.
  5516   - Code_Char_chr like Code_Char, but also offers treatment of character
  5517     codes; includes Code_Integer.
  5518   - Executable_Set allows to generate code for finite sets using lists.
  5519   - Executable_Rat implements rational numbers as triples (sign, enumerator,
  5520     denominator).
  5521   - Executable_Real implements a subset of real numbers, namly those
  5522     representable by rational numbers.
  5523   - Efficient_Nat implements natural numbers by integers, which in general will
  5524     result in higher efficency; pattern matching with 0/Suc is eliminated;
  5525     includes Code_Integer.
  5526   - Code_Index provides an additional datatype index which is mapped to
  5527     target-language built-in integers.
  5528   - Code_Message provides an additional datatype message_string which is isomorphic to
  5529     strings; messages are mapped to target-language strings.
  5530 
  5531 * New package for inductive predicates
  5532 
  5533   An n-ary predicate p with m parameters z_1, ..., z_m can now be defined via
  5534 
  5535     inductive
  5536       p :: "U_1 => ... => U_m => T_1 => ... => T_n => bool"
  5537       for z_1 :: U_1 and ... and z_n :: U_m
  5538     where
  5539       rule_1: "... ==> p z_1 ... z_m t_1_1 ... t_1_n"
  5540     | ...
  5541 
  5542   with full support for type-inference, rather than
  5543 
  5544     consts s :: "U_1 => ... => U_m => (T_1 * ... * T_n) set"
  5545 
  5546     abbreviation p :: "U_1 => ... => U_m => T_1 => ... => T_n => bool"
  5547     where "p z_1 ... z_m x_1 ... x_n == (x_1, ..., x_n) : s z_1 ... z_m"
  5548 
  5549     inductive "s z_1 ... z_m"
  5550     intros
  5551       rule_1: "... ==> (t_1_1, ..., t_1_n) : s z_1 ... z_m"
  5552       ...
  5553 
  5554   For backward compatibility, there is a wrapper allowing inductive
  5555   sets to be defined with the new package via
  5556 
  5557     inductive_set
  5558       s :: "U_1 => ... => U_m => (T_1 * ... * T_n) set"
  5559       for z_1 :: U_1 and ... and z_n :: U_m
  5560     where
  5561       rule_1: "... ==> (t_1_1, ..., t_1_n) : s z_1 ... z_m"
  5562     | ...
  5563 
  5564   or
  5565 
  5566     inductive_set
  5567       s :: "U_1 => ... => U_m => (T_1 * ... * T_n) set"
  5568       and p :: "U_1 => ... => U_m => T_1 => ... => T_n => bool"
  5569       for z_1 :: U_1 and ... and z_n :: U_m
  5570     where
  5571       "p z_1 ... z_m x_1 ... x_n == (x_1, ..., x_n) : s z_1 ... z_m"
  5572     | rule_1: "... ==> p z_1 ... z_m t_1_1 ... t_1_n"
  5573     | ...
  5574 
  5575   if the additional syntax "p ..." is required.
  5576 
  5577   Numerous examples can be found in the subdirectories src/HOL/Auth,
  5578   src/HOL/Bali, src/HOL/Induct, and src/HOL/MicroJava.
  5579 
  5580   INCOMPATIBILITIES:
  5581 
  5582   - Since declaration and definition of inductive sets or predicates
  5583     is no longer separated, abbreviations involving the newly
  5584     introduced sets or predicates must be specified together with the
  5585     introduction rules after the 'where' keyword (see above), rather
  5586     than before the actual inductive definition.
  5587 
  5588   - The variables in induction and elimination rules are now
  5589     quantified in the order of their occurrence in the introduction
  5590     rules, rather than in alphabetical order. Since this may break
  5591     some proofs, these proofs either have to be repaired, e.g. by
  5592     reordering the variables a_i_1 ... a_i_{k_i} in Isar 'case'
  5593     statements of the form
  5594 
  5595       case (rule_i a_i_1 ... a_i_{k_i})
  5596 
  5597     or the old order of quantification has to be restored by explicitly adding
  5598     meta-level quantifiers in the introduction rules, i.e.
  5599 
  5600       | rule_i: "!!a_i_1 ... a_i_{k_i}. ... ==> p z_1 ... z_m t_i_1 ... t_i_n"
  5601 
  5602   - The format of the elimination rules is now
  5603 
  5604       p z_1 ... z_m x_1 ... x_n ==>
  5605         (!!a_1_1 ... a_1_{k_1}. x_1 = t_1_1 ==> ... ==> x_n = t_1_n ==> ... ==> P)
  5606         ==> ... ==> P
  5607 
  5608     for predicates and
  5609 
  5610       (x_1, ..., x_n) : s z_1 ... z_m ==>
  5611         (!!a_1_1 ... a_1_{k_1}. x_1 = t_1_1 ==> ... ==> x_n = t_1_n ==> ... ==> P)
  5612         ==> ... ==> P
  5613 
  5614     for sets rather than
  5615 
  5616       x : s z_1 ... z_m ==>
  5617         (!!a_1_1 ... a_1_{k_1}. x = (t_1_1, ..., t_1_n) ==> ... ==> P)
  5618         ==> ... ==> P
  5619 
  5620     This may require terms in goals to be expanded to n-tuples
  5621     (e.g. using case_tac or simplification with the split_paired_all
  5622     rule) before the above elimination rule is applicable.
  5623 
  5624   - The elimination or case analysis rules for (mutually) inductive
  5625     sets or predicates are now called "p_1.cases" ... "p_k.cases". The
  5626     list of rules "p_1_..._p_k.elims" is no longer available.
  5627 
  5628 * New package "function"/"fun" for general recursive functions,
  5629 supporting mutual and nested recursion, definitions in local contexts,
  5630 more general pattern matching and partiality. See HOL/ex/Fundefs.thy
  5631 for small examples, and the separate tutorial on the function
  5632 package. The old recdef "package" is still available as before, but
  5633 users are encouraged to use the new package.
  5634 
  5635 * Method "lexicographic_order" automatically synthesizes termination
  5636 relations as lexicographic combinations of size measures. 
  5637 
  5638 * Case-expressions allow arbitrary constructor-patterns (including
  5639 "_") and take their order into account, like in functional
  5640 programming.  Internally, this is translated into nested
  5641 case-expressions; missing cases are added and mapped to the predefined
  5642 constant "undefined". In complicated cases printing may no longer show
  5643 the original input but the internal form. Lambda-abstractions allow
  5644 the same form of pattern matching: "% pat1 => e1 | ..." is an
  5645 abbreviation for "%x. case x of pat1 => e1 | ..." where x is a new
  5646 variable.
  5647 
  5648 * IntDef: The constant "int :: nat => int" has been removed; now "int"
  5649 is an abbreviation for "of_nat :: nat => int". The simplification
  5650 rules for "of_nat" have been changed to work like "int" did
  5651 previously.  Potential INCOMPATIBILITY:
  5652   - "of_nat (Suc m)" simplifies to "1 + of_nat m" instead of "of_nat m + 1"
  5653   - of_nat_diff and of_nat_mult are no longer default simp rules
  5654 
  5655 * Method "algebra" solves polynomial equations over (semi)rings using
  5656 Groebner bases. The (semi)ring structure is defined by locales and the
  5657 tool setup depends on that generic context. Installing the method for
  5658 a specific type involves instantiating the locale and possibly adding
  5659 declarations for computation on the coefficients.  The method is
  5660 already instantiated for natural numbers and for the axiomatic class
  5661 of idoms with numerals.  See also the paper by Chaieb and Wenzel at
  5662 CALCULEMUS 2007 for the general principles underlying this
  5663 architecture of context-aware proof-tools.
  5664 
  5665 * Method "ferrack" implements quantifier elimination over
  5666 special-purpose dense linear orders using locales (analogous to
  5667 "algebra"). The method is already installed for class
  5668 {ordered_field,recpower,number_ring} which subsumes real, hyperreal,
  5669 rat, etc.
  5670 
  5671 * Former constant "List.op @" now named "List.append".  Use ML
  5672 antiquotations @{const_name List.append} or @{term " ... @ ... "} to
  5673 circumvent possible incompatibilities when working on ML level.
  5674 
  5675 * primrec: missing cases mapped to "undefined" instead of "arbitrary".
  5676 
  5677 * New function listsum :: 'a list => 'a for arbitrary monoids.
  5678 Special syntax: "SUM x <- xs. f x" (and latex variants)
  5679 
  5680 * New syntax for Haskell-like list comprehension (input only), eg.
  5681 [(x,y). x <- xs, y <- ys, x ~= y], see also src/HOL/List.thy.
  5682 
  5683 * The special syntax for function "filter" has changed from [x :
  5684 xs. P] to [x <- xs. P] to avoid an ambiguity caused by list
  5685 comprehension syntax, and for uniformity.  INCOMPATIBILITY.
  5686 
  5687 * [a..b] is now defined for arbitrary linear orders.  It used to be
  5688 defined on nat only, as an abbreviation for [a..<Suc b]
  5689 INCOMPATIBILITY.
  5690 
  5691 * Renamed lemma "set_take_whileD"  to "set_takeWhileD".
  5692 
  5693 * New functions "sorted" and "sort" in src/HOL/List.thy.
  5694 
  5695 * New lemma collection field_simps (an extension of ring_simps) for
  5696 manipulating (in)equations involving division. Multiplies with all
  5697 denominators that can be proved to be non-zero (in equations) or
  5698 positive/negative (in inequations).
  5699 
  5700 * Lemma collections ring_eq_simps, group_eq_simps and ring_distrib
  5701 have been improved and renamed to ring_simps, group_simps and
  5702 ring_distribs.  Removed lemmas field_xyz in theory Ring_and_Field
  5703 because they were subsumed by lemmas xyz.  INCOMPATIBILITY.
  5704 
  5705 * Theory Library/Commutative_Ring: switched from recdef to function
  5706 package; constants add, mul, pow now curried.  Infix syntax for
  5707 algebraic operations.
  5708 
  5709 * Dropped redundant lemma def_imp_eq in favor of meta_eq_to_obj_eq.
  5710 INCOMPATIBILITY.
  5711 
  5712 * Dropped redundant lemma if_def2 in favor of if_bool_eq_conj.
  5713 INCOMPATIBILITY.
  5714 
  5715 * HOL/records: generalised field-update to take a function on the
  5716 field rather than the new value: r(|A := x|) is translated to A_update
  5717 (K x) r The K-combinator that is internally used is called K_record.
  5718 INCOMPATIBILITY: Usage of the plain update functions has to be
  5719 adapted.
  5720  
  5721 * Class "semiring_0" now contains annihilation axioms x * 0 = 0 and 0
  5722 * x = 0, which are required for a semiring.  Richer structures do not
  5723 inherit from semiring_0 anymore, because this property is a theorem
  5724 there, not an axiom.  INCOMPATIBILITY: In instances of semiring_0,
  5725 there is more to prove, but this is mostly trivial.
  5726 
  5727 * Class "recpower" is generalized to arbitrary monoids, not just
  5728 commutative semirings.  INCOMPATIBILITY: may need to incorporate
  5729 commutativity or semiring properties additionally.
  5730 
  5731 * Constant "List.list_all2" in List.thy now uses authentic syntax.
  5732 INCOMPATIBILITY: translations containing list_all2 may go wrong,
  5733 better use 'abbreviation'.
  5734 
  5735 * Renamed constant "List.op mem" to "List.member".  INCOMPATIBILITY.
  5736 
  5737 * Numeral syntax: type 'bin' which was a mere type copy of 'int' has
  5738 been abandoned in favour of plain 'int'.  INCOMPATIBILITY --
  5739 significant changes for setting up numeral syntax for types:
  5740   - New constants Numeral.pred and Numeral.succ instead
  5741       of former Numeral.bin_pred and Numeral.bin_succ.
  5742   - Use integer operations instead of bin_add, bin_mult and so on.
  5743   - Numeral simplification theorems named Numeral.numeral_simps instead of Bin_simps.
  5744   - ML structure Bin_Simprocs now named Int_Numeral_Base_Simprocs.
  5745 
  5746 See src/HOL/Integ/IntArith.thy for an example setup.
  5747 
  5748 * Command 'normal_form' computes the normal form of a term that may
  5749 contain free variables.  For example ``normal_form "rev [a, b, c]"''
  5750 produces ``[b, c, a]'' (without proof).  This command is suitable for
  5751 heavy-duty computations because the functions are compiled to ML
  5752 first.  Correspondingly, a method "normalization" is provided.  See
  5753 further src/HOL/ex/NormalForm.thy and src/Tools/nbe.ML.
  5754 
  5755 * Alternative iff syntax "A <-> B" for equality on bool (with priority
  5756 25 like -->); output depends on the "iff" print_mode, the default is
  5757 "A = B" (with priority 50).
  5758 
  5759 * Relations less (<) and less_eq (<=) are also available on type bool.
  5760 Modified syntax to disallow nesting without explicit parentheses,
  5761 e.g. "(x < y) < z" or "x < (y < z)", but NOT "x < y < z".  Potential
  5762 INCOMPATIBILITY.
  5763 
  5764 * "LEAST x:A. P" expands to "LEAST x. x:A & P" (input only).
  5765 
  5766 * Relation composition operator "op O" now has precedence 75 and binds
  5767 stronger than union and intersection. INCOMPATIBILITY.
  5768 
  5769 * The old set interval syntax "{m..n(}" (and relatives) has been
  5770 removed.  Use "{m..<n}" (and relatives) instead.
  5771 
  5772 * In the context of the assumption "~(s = t)" the Simplifier rewrites
  5773 "t = s" to False (by simproc "neq").  INCOMPATIBILITY, consider using
  5774 ``declare [[simproc del: neq]]''.
  5775 
  5776 * Simplifier: "m dvd n" where m and n are numbers is evaluated to
  5777 True/False.
  5778 
  5779 * Theorem Cons_eq_map_conv no longer declared as "simp".
  5780 
  5781 * Theorem setsum_mult renamed to setsum_right_distrib.
  5782 
  5783 * Prefer ex1I over ex_ex1I in single-step reasoning, e.g. by the
  5784 ``rule'' method.
  5785 
  5786 * Reimplemented methods "sat" and "satx", with several improvements:
  5787 goals no longer need to be stated as "<prems> ==> False", equivalences
  5788 (i.e. "=" on type bool) are handled, variable names of the form
  5789 "lit_<n>" are no longer reserved, significant speedup.
  5790 
  5791 * Methods "sat" and "satx" can now replay MiniSat proof traces.
  5792 zChaff is still supported as well.
  5793 
  5794 * 'inductive' and 'datatype': provide projections of mutual rules,
  5795 bundled as foo_bar.inducts;
  5796 
  5797 * Library: moved theories Parity, GCD, Binomial, Infinite_Set to
  5798 Library.
  5799 
  5800 * Library: moved theory Accessible_Part to main HOL.
  5801 
  5802 * Library: added theory Coinductive_List of potentially infinite lists
  5803 as greatest fixed-point.
  5804 
  5805 * Library: added theory AssocList which implements (finite) maps as
  5806 association lists.
  5807 
  5808 * Method "evaluation" solves goals (i.e. a boolean expression)
  5809 efficiently by compiling it to ML.  The goal is "proved" (via an
  5810 oracle) if it evaluates to True.
  5811 
  5812 * Linear arithmetic now splits certain operators (e.g. min, max, abs)
  5813 also when invoked by the simplifier.  This results in the Simplifier
  5814 being more powerful on arithmetic goals.  INCOMPATIBILITY.
  5815 Configuration option fast_arith_split_limit=0 recovers the old
  5816 behavior.
  5817 
  5818 * Support for hex (0x20) and binary (0b1001) numerals.
  5819 
  5820 * New method: reify eqs (t), where eqs are equations for an
  5821 interpretation I :: 'a list => 'b => 'c and t::'c is an optional
  5822 parameter, computes a term s::'b and a list xs::'a list and proves the
  5823 theorem I xs s = t. This is also known as reification or quoting. The
  5824 resulting theorem is applied to the subgoal to substitute t with I xs
  5825 s.  If t is omitted, the subgoal itself is reified.
  5826 
  5827 * New method: reflection corr_thm eqs (t). The parameters eqs and (t)
  5828 are as explained above. corr_thm is a theorem for I vs (f t) = I vs t,
  5829 where f is supposed to be a computable function (in the sense of code
  5830 generattion). The method uses reify to compute s and xs as above then
  5831 applies corr_thm and uses normalization by evaluation to "prove" f s =
  5832 r and finally gets the theorem t = r, which is again applied to the
  5833 subgoal. An Example is available in src/HOL/ex/ReflectionEx.thy.
  5834 
  5835 * Reflection: Automatic reification now handels binding, an example is
  5836 available in src/HOL/ex/ReflectionEx.thy
  5837 
  5838 * HOL-Statespace: ``State Spaces: The Locale Way'' introduces a
  5839 command 'statespace' that is similar to 'record', but introduces an
  5840 abstract specification based on the locale infrastructure instead of
  5841 HOL types.  This leads to extra flexibility in composing state spaces,
  5842 in particular multiple inheritance and renaming of components.
  5843 
  5844 
  5845 *** HOL-Complex ***
  5846 
  5847 * Hyperreal: Functions root and sqrt are now defined on negative real
  5848 inputs so that root n (- x) = - root n x and sqrt (- x) = - sqrt x.
  5849 Nonnegativity side conditions have been removed from many lemmas, so
  5850 that more subgoals may now be solved by simplification; potential
  5851 INCOMPATIBILITY.
  5852 
  5853 * Real: new type classes formalize real normed vector spaces and
  5854 algebras, using new overloaded constants scaleR :: real => 'a => 'a
  5855 and norm :: 'a => real.
  5856 
  5857 * Real: constant of_real :: real => 'a::real_algebra_1 injects from
  5858 reals into other types. The overloaded constant Reals :: 'a set is now
  5859 defined as range of_real; potential INCOMPATIBILITY.
  5860 
  5861 * Real: proper support for ML code generation, including 'quickcheck'.
  5862 Reals are implemented as arbitrary precision rationals.
  5863 
  5864 * Hyperreal: Several constants that previously worked only for the
  5865 reals have been generalized, so they now work over arbitrary vector
  5866 spaces. Type annotations may need to be added in some cases; potential
  5867 INCOMPATIBILITY.
  5868 
  5869   Infinitesimal  :: ('a::real_normed_vector) star set
  5870   HFinite        :: ('a::real_normed_vector) star set
  5871   HInfinite      :: ('a::real_normed_vector) star set
  5872   approx         :: ('a::real_normed_vector) star => 'a star => bool
  5873   monad          :: ('a::real_normed_vector) star => 'a star set
  5874   galaxy         :: ('a::real_normed_vector) star => 'a star set
  5875   (NS)LIMSEQ     :: [nat => 'a::real_normed_vector, 'a] => bool
  5876   (NS)convergent :: (nat => 'a::real_normed_vector) => bool
  5877   (NS)Bseq       :: (nat => 'a::real_normed_vector) => bool
  5878   (NS)Cauchy     :: (nat => 'a::real_normed_vector) => bool
  5879   (NS)LIM        :: ['a::real_normed_vector => 'b::real_normed_vector, 'a, 'b] => bool
  5880   is(NS)Cont     :: ['a::real_normed_vector => 'b::real_normed_vector, 'a] => bool
  5881   deriv          :: ['a::real_normed_field => 'a, 'a, 'a] => bool
  5882   sgn            :: 'a::real_normed_vector => 'a
  5883   exp            :: 'a::{recpower,real_normed_field,banach} => 'a
  5884 
  5885 * Complex: Some complex-specific constants are now abbreviations for
  5886 overloaded ones: complex_of_real = of_real, cmod = norm, hcmod =
  5887 hnorm.  Other constants have been entirely removed in favor of the
  5888 polymorphic versions (INCOMPATIBILITY):
  5889 
  5890   approx        <-- capprox
  5891   HFinite       <-- CFinite
  5892   HInfinite     <-- CInfinite
  5893   Infinitesimal <-- CInfinitesimal
  5894   monad         <-- cmonad
  5895   galaxy        <-- cgalaxy
  5896   (NS)LIM       <-- (NS)CLIM, (NS)CRLIM
  5897   is(NS)Cont    <-- is(NS)Contc, is(NS)contCR
  5898   (ns)deriv     <-- (ns)cderiv
  5899 
  5900 
  5901 *** HOL-Algebra ***
  5902 
  5903 * Formalisation of ideals and the quotient construction over rings.
  5904 
  5905 * Order and lattice theory no longer based on records.
  5906 INCOMPATIBILITY.
  5907 
  5908 * Renamed lemmas least_carrier -> least_closed and greatest_carrier ->
  5909 greatest_closed.  INCOMPATIBILITY.
  5910 
  5911 * Method algebra is now set up via an attribute.  For examples see
  5912 Ring.thy.  INCOMPATIBILITY: the method is now weaker on combinations
  5913 of algebraic structures.
  5914 
  5915 * Renamed theory CRing to Ring.
  5916 
  5917 
  5918 *** HOL-Nominal ***
  5919 
  5920 * Substantial, yet incomplete support for nominal datatypes (binding
  5921 structures) based on HOL-Nominal logic.  See src/HOL/Nominal and
  5922 src/HOL/Nominal/Examples.  Prospective users should consult
  5923 http://isabelle.in.tum.de/nominal/
  5924 
  5925 
  5926 *** ML ***
  5927 
  5928 * ML basics: just one true type int, which coincides with IntInf.int
  5929 (even on SML/NJ).
  5930 
  5931 * ML within Isar: antiquotations allow to embed statically-checked
  5932 formal entities in the source, referring to the context available at
  5933 compile-time.  For example:
  5934 
  5935 ML {* @{sort "{zero,one}"} *}
  5936 ML {* @{typ "'a => 'b"} *}
  5937 ML {* @{term "%x. x"} *}
  5938 ML {* @{prop "x == y"} *}
  5939 ML {* @{ctyp "'a => 'b"} *}
  5940 ML {* @{cterm "%x. x"} *}
  5941 ML {* @{cprop "x == y"} *}
  5942 ML {* @{thm asm_rl} *}
  5943 ML {* @{thms asm_rl} *}
  5944 ML {* @{type_name c} *}
  5945 ML {* @{type_syntax c} *}
  5946 ML {* @{const_name c} *}
  5947 ML {* @{const_syntax c} *}
  5948 ML {* @{context} *}
  5949 ML {* @{theory} *}
  5950 ML {* @{theory Pure} *}
  5951 ML {* @{theory_ref} *}
  5952 ML {* @{theory_ref Pure} *}
  5953 ML {* @{simpset} *}
  5954 ML {* @{claset} *}
  5955 ML {* @{clasimpset} *}
  5956 
  5957 The same works for sources being ``used'' within an Isar context.
  5958 
  5959 * ML in Isar: improved error reporting; extra verbosity with
  5960 ML_Context.trace enabled.
  5961 
  5962 * Pure/General/table.ML: the join operations now works via exceptions
  5963 DUP/SAME instead of type option. This is simpler in simple cases, and
  5964 admits slightly more efficient complex applications.
  5965 
  5966 * Pure: 'advanced' translation functions (parse_translation etc.) now
  5967 use Context.generic instead of just theory.
  5968 
  5969 * Pure: datatype Context.generic joins theory/Proof.context and
  5970 provides some facilities for code that works in either kind of
  5971 context, notably GenericDataFun for uniform theory and proof data.
  5972 
  5973 * Pure: simplified internal attribute type, which is now always
  5974 Context.generic * thm -> Context.generic * thm. Global (theory) vs.
  5975 local (Proof.context) attributes have been discontinued, while
  5976 minimizing code duplication. Thm.rule_attribute and
  5977 Thm.declaration_attribute build canonical attributes; see also structure
  5978 Context for further operations on Context.generic, notably
  5979 GenericDataFun. INCOMPATIBILITY, need to adapt attribute type
  5980 declarations and definitions.
  5981 
  5982 * Context data interfaces (Theory/Proof/GenericDataFun): removed
  5983 name/print, uninitialized data defaults to ad-hoc copy of empty value,
  5984 init only required for impure data. INCOMPATIBILITY: empty really need
  5985 to be empty (no dependencies on theory content!)
  5986 
  5987 * Pure/kernel: consts certification ignores sort constraints given in
  5988 signature declarations. (This information is not relevant to the
  5989 logic, but only for type inference.) SIGNIFICANT INTERNAL CHANGE,
  5990 potential INCOMPATIBILITY.
  5991 
  5992 * Pure: axiomatic type classes are now purely definitional, with
  5993 explicit proofs of class axioms and super class relations performed
  5994 internally. See Pure/axclass.ML for the main internal interfaces --
  5995 notably AxClass.define_class supercedes AxClass.add_axclass, and
  5996 AxClass.axiomatize_class/classrel/arity supersede
  5997 Sign.add_classes/classrel/arities.
  5998 
  5999 * Pure/Isar: Args/Attrib parsers operate on Context.generic --
  6000 global/local versions on theory vs. Proof.context have been
  6001 discontinued; Attrib.syntax and Method.syntax have been adapted
  6002 accordingly.  INCOMPATIBILITY, need to adapt parser expressions for
  6003 attributes, methods, etc.
  6004 
  6005 * Pure: several functions of signature "... -> theory -> theory * ..."
  6006 have been reoriented to "... -> theory -> ... * theory" in order to
  6007 allow natural usage in combination with the ||>, ||>>, |-> and
  6008 fold_map combinators.
  6009 
  6010 * Pure: official theorem names (closed derivations) and additional
  6011 comments (tags) are now strictly separate.  Name hints -- which are
  6012 maintained as tags -- may be attached any time without affecting the
  6013 derivation.
  6014 
  6015 * Pure: primitive rule lift_rule now takes goal cterm instead of an
  6016 actual goal state (thm).  Use Thm.lift_rule (Thm.cprem_of st i) to
  6017 achieve the old behaviour.
  6018 
  6019 * Pure: the "Goal" constant is now called "prop", supporting a
  6020 slightly more general idea of ``protecting'' meta-level rule
  6021 statements.
  6022 
  6023 * Pure: Logic.(un)varify only works in a global context, which is now
  6024 enforced instead of silently assumed.  INCOMPATIBILITY, may use
  6025 Logic.legacy_(un)varify as temporary workaround.
  6026 
  6027 * Pure: structure Name provides scalable operations for generating
  6028 internal variable names, notably Name.variants etc.  This replaces
  6029 some popular functions from term.ML:
  6030 
  6031   Term.variant		->  Name.variant
  6032   Term.variantlist	->  Name.variant_list
  6033   Term.invent_names	->  Name.invent_list
  6034 
  6035 Note that low-level renaming rarely occurs in new code -- operations
  6036 from structure Variable are used instead (see below).
  6037 
  6038 * Pure: structure Variable provides fundamental operations for proper
  6039 treatment of fixed/schematic variables in a context.  For example,
  6040 Variable.import introduces fixes for schematics of given facts and
  6041 Variable.export reverses the effect (up to renaming) -- this replaces
  6042 various freeze_thaw operations.
  6043 
  6044 * Pure: structure Goal provides simple interfaces for
  6045 init/conclude/finish and tactical prove operations (replacing former
  6046 Tactic.prove).  Goal.prove is the canonical way to prove results
  6047 within a given context; Goal.prove_global is a degraded version for
  6048 theory level goals, including a global Drule.standard.  Note that
  6049 OldGoals.prove_goalw_cterm has long been obsolete, since it is
  6050 ill-behaved in a local proof context (e.g. with local fixes/assumes or
  6051 in a locale context).
  6052 
  6053 * Pure/Syntax: generic interfaces for parsing (Syntax.parse_term etc.)
  6054 and type checking (Syntax.check_term etc.), with common combinations
  6055 (Syntax.read_term etc.). These supersede former Sign.read_term etc.
  6056 which are considered legacy and await removal.
  6057 
  6058 * Pure/Syntax: generic interfaces for type unchecking
  6059 (Syntax.uncheck_terms etc.) and unparsing (Syntax.unparse_term etc.),
  6060 with common combinations (Syntax.pretty_term, Syntax.string_of_term
  6061 etc.).  Former Sign.pretty_term, Sign.string_of_term etc. are still
  6062 available for convenience, but refer to the very same operations using
  6063 a mere theory instead of a full context.
  6064 
  6065 * Isar: simplified treatment of user-level errors, using exception
  6066 ERROR of string uniformly.  Function error now merely raises ERROR,
  6067 without any side effect on output channels.  The Isar toplevel takes
  6068 care of proper display of ERROR exceptions.  ML code may use plain
  6069 handle/can/try; cat_error may be used to concatenate errors like this:
  6070 
  6071   ... handle ERROR msg => cat_error msg "..."
  6072 
  6073 Toplevel ML code (run directly or through the Isar toplevel) may be
  6074 embedded into the Isar toplevel with exception display/debug like
  6075 this:
  6076 
  6077   Isar.toplevel (fn () => ...)
  6078 
  6079 INCOMPATIBILITY, removed special transform_error facilities, removed
  6080 obsolete variants of user-level exceptions (ERROR_MESSAGE,
  6081 Context.PROOF, ProofContext.CONTEXT, Proof.STATE, ProofHistory.FAIL)
  6082 -- use plain ERROR instead.
  6083 
  6084 * Isar: theory setup now has type (theory -> theory), instead of a
  6085 list.  INCOMPATIBILITY, may use #> to compose setup functions.
  6086 
  6087 * Isar: ML toplevel pretty printer for type Proof.context, subject to
  6088 ProofContext.debug/verbose flags.
  6089 
  6090 * Isar: Toplevel.theory_to_proof admits transactions that modify the
  6091 theory before entering a proof state.  Transactions now always see a
  6092 quasi-functional intermediate checkpoint, both in interactive and
  6093 batch mode.
  6094 
  6095 * Isar: simplified interfaces for outer syntax.  Renamed
  6096 OuterSyntax.add_keywords to OuterSyntax.keywords.  Removed
  6097 OuterSyntax.add_parsers -- this functionality is now included in
  6098 OuterSyntax.command etc.  INCOMPATIBILITY.
  6099 
  6100 * Simplifier: the simpset of a running simplification process now
  6101 contains a proof context (cf. Simplifier.the_context), which is the
  6102 very context that the initial simpset has been retrieved from (by
  6103 simpset_of/local_simpset_of).  Consequently, all plug-in components
  6104 (solver, looper etc.) may depend on arbitrary proof data.
  6105 
  6106 * Simplifier.inherit_context inherits the proof context (plus the
  6107 local bounds) of the current simplification process; any simproc
  6108 etc. that calls the Simplifier recursively should do this!  Removed
  6109 former Simplifier.inherit_bounds, which is already included here --
  6110 INCOMPATIBILITY.  Tools based on low-level rewriting may even have to
  6111 specify an explicit context using Simplifier.context/theory_context.
  6112 
  6113 * Simplifier/Classical Reasoner: more abstract interfaces
  6114 change_simpset/claset for modifying the simpset/claset reference of a
  6115 theory; raw versions simpset/claset_ref etc. have been discontinued --
  6116 INCOMPATIBILITY.
  6117 
  6118 * Provers: more generic wrt. syntax of object-logics, avoid hardwired
  6119 "Trueprop" etc.
  6120 
  6121 
  6122 *** System ***
  6123 
  6124 * settings: the default heap location within ISABELLE_HOME_USER now
  6125 includes ISABELLE_IDENTIFIER.  This simplifies use of multiple
  6126 Isabelle installations.
  6127 
  6128 * isabelle-process: option -S (secure mode) disables some critical
  6129 operations, notably runtime compilation and evaluation of ML source
  6130 code.
  6131 
  6132 * Basic Isabelle mode for jEdit, see Isabelle/lib/jedit/.
  6133 
  6134 * Support for parallel execution, using native multicore support of
  6135 Poly/ML 5.1.  The theory loader exploits parallelism when processing
  6136 independent theories, according to the given theory header
  6137 specifications. The maximum number of worker threads is specified via
  6138 usedir option -M or the "max-threads" setting in Proof General. A
  6139 speedup factor of 1.5--3.5 can be expected on a 4-core machine, and up
  6140 to 6 on a 8-core machine.  User-code needs to observe certain
  6141 guidelines for thread-safe programming, see appendix A in the Isar
  6142 Implementation manual.
  6143 
  6144 
  6145 
  6146 New in Isabelle2005 (October 2005)
  6147 ----------------------------------
  6148 
  6149 *** General ***
  6150 
  6151 * Theory headers: the new header syntax for Isar theories is
  6152 
  6153   theory <name>
  6154   imports <theory1> ... <theoryN>
  6155   uses <file1> ... <fileM>
  6156   begin
  6157 
  6158 where the 'uses' part is optional.  The previous syntax
  6159 
  6160   theory <name> = <theory1> + ... + <theoryN>:
  6161 
  6162 will disappear in the next release.  Use isatool fixheaders to convert
  6163 existing theory files.  Note that there is no change in ancient
  6164 non-Isar theories now, but these will disappear soon.
  6165 
  6166 * Theory loader: parent theories can now also be referred to via
  6167 relative and absolute paths.
  6168 
  6169 * Command 'find_theorems' searches for a list of criteria instead of a
  6170 list of constants. Known criteria are: intro, elim, dest, name:string,
  6171 simp:term, and any term. Criteria can be preceded by '-' to select
  6172 theorems that do not match. Intro, elim, dest select theorems that
  6173 match the current goal, name:s selects theorems whose fully qualified
  6174 name contain s, and simp:term selects all simplification rules whose
  6175 lhs match term.  Any other term is interpreted as pattern and selects
  6176 all theorems matching the pattern. Available in ProofGeneral under
  6177 'ProofGeneral -> Find Theorems' or C-c C-f.  Example:
  6178 
  6179   C-c C-f (100) "(_::nat) + _ + _" intro -name: "HOL."
  6180 
  6181 prints the last 100 theorems matching the pattern "(_::nat) + _ + _",
  6182 matching the current goal as introduction rule and not having "HOL."
  6183 in their name (i.e. not being defined in theory HOL).
  6184 
  6185 * Command 'thms_containing' has been discontinued in favour of
  6186 'find_theorems'; INCOMPATIBILITY.
  6187 
  6188 * Communication with Proof General is now 8bit clean, which means that
  6189 Unicode text in UTF-8 encoding may be used within theory texts (both
  6190 formal and informal parts).  Cf. option -U of the Isabelle Proof
  6191 General interface.  Here are some simple examples (cf. src/HOL/ex):
  6192 
  6193   http://isabelle.in.tum.de/library/HOL/ex/Hebrew.html
  6194   http://isabelle.in.tum.de/library/HOL/ex/Chinese.html
  6195 
  6196 * Improved efficiency of the Simplifier and, to a lesser degree, the
  6197 Classical Reasoner.  Typical big applications run around 2 times
  6198 faster.
  6199 
  6200 
  6201 *** Document preparation ***
  6202 
  6203 * Commands 'display_drafts' and 'print_drafts' perform simple output
  6204 of raw sources.  Only those symbols that do not require additional
  6205 LaTeX packages (depending on comments in isabellesym.sty) are
  6206 displayed properly, everything else is left verbatim.  isatool display
  6207 and isatool print are used as front ends (these are subject to the
  6208 DVI/PDF_VIEWER and PRINT_COMMAND settings, respectively).
  6209 
  6210 * Command tags control specific markup of certain regions of text,
  6211 notably folding and hiding.  Predefined tags include "theory" (for
  6212 theory begin and end), "proof" for proof commands, and "ML" for
  6213 commands involving ML code; the additional tags "visible" and
  6214 "invisible" are unused by default.  Users may give explicit tag
  6215 specifications in the text, e.g. ''by %invisible (auto)''.  The
  6216 interpretation of tags is determined by the LaTeX job during document
  6217 preparation: see option -V of isatool usedir, or options -n and -t of
  6218 isatool document, or even the LaTeX macros \isakeeptag, \isafoldtag,
  6219 \isadroptag.
  6220 
  6221 Several document versions may be produced at the same time via isatool
  6222 usedir (the generated index.html will link all of them).  Typical
  6223 specifications include ''-V document=theory,proof,ML'' to present
  6224 theory/proof/ML parts faithfully, ''-V outline=/proof,/ML'' to fold
  6225 proof and ML commands, and ''-V mutilated=-theory,-proof,-ML'' to omit
  6226 these parts without any formal replacement text.  The Isabelle site
  6227 default settings produce ''document'' and ''outline'' versions as
  6228 specified above.
  6229 
  6230 * Several new antiquotations:
  6231 
  6232   @{term_type term} prints a term with its type annotated;
  6233 
  6234   @{typeof term} prints the type of a term;
  6235 
  6236   @{const const} is the same as @{term const}, but checks that the