NEWS
author boehmes
Mon Jun 04 09:07:23 2012 +0200 (2012-06-04)
changeset 48069 e9b2782c4f99
parent 48013 44de84112a67
child 48094 c3d4f4d9e54c
permissions -rw-r--r--
restricted Z3 by default to a fragment where proof reconstruction should not fail (for better integration with Sledgehammer) -- the full set of supported Z3 features can still be used by enabling the configuration option "z3_with_extensions"
     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 * The SMT solver Z3 has now by default a restricted set of directly
    29 supported features. For the full set of features (div/mod, nonlinear
    30 arithmetic, datatypes/records) with potential proof reconstruction
    31 failures, enable the configuration option "z3_with_extensions".
    32 Minor INCOMPATIBILITY.
    33 
    34 New in Isabelle2012 (May 2012)
    35 ------------------------------
    36 
    37 *** General ***
    38 
    39 * Prover IDE (PIDE) improvements:
    40 
    41   - more robust Sledgehammer integration (as before the sledgehammer
    42     command-line needs to be typed into the source buffer)
    43   - markup for bound variables
    44   - markup for types of term variables (displayed as tooltips)
    45   - support for user-defined Isar commands within the running session
    46   - improved support for Unicode outside original 16bit range
    47     e.g. glyph for \<A> (thanks to jEdit 4.5.1)
    48 
    49 * Forward declaration of outer syntax keywords within the theory
    50 header -- minor INCOMPATIBILITY for user-defined commands.  Allow new
    51 commands to be used in the same theory where defined.
    52 
    53 * Auxiliary contexts indicate block structure for specifications with
    54 additional parameters and assumptions.  Such unnamed contexts may be
    55 nested within other targets, like 'theory', 'locale', 'class',
    56 'instantiation' etc.  Results from the local context are generalized
    57 accordingly and applied to the enclosing target context.  Example:
    58 
    59   context
    60     fixes x y z :: 'a
    61     assumes xy: "x = y" and yz: "y = z"
    62   begin
    63 
    64   lemma my_trans: "x = z" using xy yz by simp
    65 
    66   end
    67 
    68   thm my_trans
    69 
    70 The most basic application is to factor-out context elements of
    71 several fixes/assumes/shows theorem statements, e.g. see
    72 ~~/src/HOL/Isar_Examples/Group_Context.thy
    73 
    74 Any other local theory specification element works within the "context
    75 ... begin ... end" block as well.
    76 
    77 * Bundled declarations associate attributed fact expressions with a
    78 given name in the context.  These may be later included in other
    79 contexts.  This allows to manage context extensions casually, without
    80 the logical dependencies of locales and locale interpretation.  See
    81 commands 'bundle', 'include', 'including' etc. in the isar-ref manual.
    82 
    83 * Commands 'lemmas' and 'theorems' allow local variables using 'for'
    84 declaration, and results are standardized before being stored.  Thus
    85 old-style "standard" after instantiation or composition of facts
    86 becomes obsolete.  Minor INCOMPATIBILITY, due to potential change of
    87 indices of schematic variables.
    88 
    89 * Rule attributes in local theory declarations (e.g. locale or class)
    90 are now statically evaluated: the resulting theorem is stored instead
    91 of the original expression.  INCOMPATIBILITY in rare situations, where
    92 the historic accident of dynamic re-evaluation in interpretations
    93 etc. was exploited.
    94 
    95 * New tutorial "Programming and Proving in Isabelle/HOL"
    96 ("prog-prove").  It completely supersedes "A Tutorial Introduction to
    97 Structured Isar Proofs" ("isar-overview"), which has been removed.  It
    98 also supersedes "Isabelle/HOL, A Proof Assistant for Higher-Order
    99 Logic" as the recommended beginners tutorial, but does not cover all
   100 of the material of that old tutorial.
   101 
   102 * Updated and extended reference manuals: "isar-ref",
   103 "implementation", "system"; reduced remaining material in old "ref"
   104 manual.
   105 
   106 
   107 *** Pure ***
   108 
   109 * Command 'definition' no longer exports the foundational "raw_def"
   110 into the user context.  Minor INCOMPATIBILITY, may use the regular
   111 "def" result with attribute "abs_def" to imitate the old version.
   112 
   113 * Attribute "abs_def" turns an equation of the form "f x y == t" into
   114 "f == %x y. t", which ensures that "simp" or "unfold" steps always
   115 expand it.  This also works for object-logic equality.  (Formerly
   116 undocumented feature.)
   117 
   118 * Sort constraints are now propagated in simultaneous statements, just
   119 like type constraints.  INCOMPATIBILITY in rare situations, where
   120 distinct sorts used to be assigned accidentally.  For example:
   121 
   122   lemma "P (x::'a::foo)" and "Q (y::'a::bar)"  -- "now illegal"
   123 
   124   lemma "P (x::'a)" and "Q (y::'a::bar)"
   125     -- "now uniform 'a::bar instead of default sort for first occurrence (!)"
   126 
   127 * Rule composition via attribute "OF" (or ML functions OF/MRS) is more
   128 tolerant against multiple unifiers, as long as the final result is
   129 unique.  (As before, rules are composed in canonical right-to-left
   130 order to accommodate newly introduced premises.)
   131 
   132 * Renamed some inner syntax categories:
   133 
   134     num ~> num_token
   135     xnum ~> xnum_token
   136     xstr ~> str_token
   137 
   138 Minor INCOMPATIBILITY.  Note that in practice "num_const" or
   139 "num_position" etc. are mainly used instead (which also include
   140 position information via constraints).
   141 
   142 * Simplified configuration options for syntax ambiguity: see
   143 "syntax_ambiguity_warning" and "syntax_ambiguity_limit" in isar-ref
   144 manual.  Minor INCOMPATIBILITY.
   145 
   146 * Discontinued configuration option "syntax_positions": atomic terms
   147 in parse trees are always annotated by position constraints.
   148 
   149 * Old code generator for SML and its commands 'code_module',
   150 'code_library', 'consts_code', 'types_code' have been discontinued.
   151 Use commands of the generic code generator instead.  INCOMPATIBILITY.
   152 
   153 * Redundant attribute "code_inline" has been discontinued. Use
   154 "code_unfold" instead.  INCOMPATIBILITY.
   155 
   156 * Dropped attribute "code_unfold_post" in favor of the its dual
   157 "code_abbrev", which yields a common pattern in definitions like
   158 
   159   definition [code_abbrev]: "f = t"
   160 
   161 INCOMPATIBILITY.
   162 
   163 * Obsolete 'types' command has been discontinued.  Use 'type_synonym'
   164 instead.  INCOMPATIBILITY.
   165 
   166 * Discontinued old "prems" fact, which used to refer to the accidental
   167 collection of foundational premises in the context (already marked as
   168 legacy since Isabelle2011).
   169 
   170 
   171 *** HOL ***
   172 
   173 * Type 'a set is now a proper type constructor (just as before
   174 Isabelle2008).  Definitions mem_def and Collect_def have disappeared.
   175 Non-trivial INCOMPATIBILITY.  For developments keeping predicates and
   176 sets separate, it is often sufficient to rephrase some set S that has
   177 been accidentally used as predicates by "%x. x : S", and some
   178 predicate P that has been accidentally used as set by "{x. P x}".
   179 Corresponding proofs in a first step should be pruned from any
   180 tinkering with former theorems mem_def and Collect_def as far as
   181 possible.
   182 
   183 For developments which deliberately mix predicates and sets, a
   184 planning step is necessary to determine what should become a predicate
   185 and what a set.  It can be helpful to carry out that step in
   186 Isabelle2011-1 before jumping right into the current release.
   187 
   188 * Code generation by default implements sets as container type rather
   189 than predicates.  INCOMPATIBILITY.
   190 
   191 * New type synonym 'a rel = ('a * 'a) set
   192 
   193 * The representation of numerals has changed.  Datatype "num"
   194 represents strictly positive binary numerals, along with functions
   195 "numeral :: num => 'a" and "neg_numeral :: num => 'a" to represent
   196 positive and negated numeric literals, respectively.  See also
   197 definitions in ~~/src/HOL/Num.thy.  Potential INCOMPATIBILITY, some
   198 user theories may require adaptations as follows:
   199 
   200   - Theorems with number_ring or number_semiring constraints: These
   201     classes are gone; use comm_ring_1 or comm_semiring_1 instead.
   202 
   203   - Theories defining numeric types: Remove number, number_semiring,
   204     and number_ring instances. Defer all theorems about numerals until
   205     after classes one and semigroup_add have been instantiated.
   206 
   207   - Numeral-only simp rules: Replace each rule having a "number_of v"
   208     pattern with two copies, one for numeral and one for neg_numeral.
   209 
   210   - Theorems about subclasses of semiring_1 or ring_1: These classes
   211     automatically support numerals now, so more simp rules and
   212     simprocs may now apply within the proof.
   213 
   214   - Definitions and theorems using old constructors Pls/Min/Bit0/Bit1:
   215     Redefine using other integer operations.
   216 
   217 * Transfer: New package intended to generalize the existing
   218 "descending" method and related theorem attributes from the Quotient
   219 package.  (Not all functionality is implemented yet, but future
   220 development will focus on Transfer as an eventual replacement for the
   221 corresponding parts of the Quotient package.)
   222 
   223   - transfer_rule attribute: Maintains a collection of transfer rules,
   224     which relate constants at two different types. Transfer rules may
   225     relate different type instances of the same polymorphic constant,
   226     or they may relate an operation on a raw type to a corresponding
   227     operation on an abstract type (quotient or subtype). For example:
   228 
   229     ((A ===> B) ===> list_all2 A ===> list_all2 B) map map
   230     (cr_int ===> cr_int ===> cr_int) (%(x,y) (u,v). (x+u, y+v)) plus_int
   231 
   232   - transfer method: Replaces a subgoal on abstract types with an
   233     equivalent subgoal on the corresponding raw types. Constants are
   234     replaced with corresponding ones according to the transfer rules.
   235     Goals are generalized over all free variables by default; this is
   236     necessary for variables whose types change, but can be overridden
   237     for specific variables with e.g. "transfer fixing: x y z".  The
   238     variant transfer' method allows replacing a subgoal with one that
   239     is logically stronger (rather than equivalent).
   240 
   241   - relator_eq attribute: Collects identity laws for relators of
   242     various type constructors, e.g. "list_all2 (op =) = (op =)".  The
   243     transfer method uses these lemmas to infer transfer rules for
   244     non-polymorphic constants on the fly.
   245 
   246   - transfer_prover method: Assists with proving a transfer rule for a
   247     new constant, provided the constant is defined in terms of other
   248     constants that already have transfer rules. It should be applied
   249     after unfolding the constant definitions.
   250 
   251   - HOL/ex/Transfer_Int_Nat.thy: Example theory demonstrating transfer
   252     from type nat to type int.
   253 
   254 * Lifting: New package intended to generalize the quotient_definition
   255 facility of the Quotient package; designed to work with Transfer.
   256 
   257   - lift_definition command: Defines operations on an abstract type in
   258     terms of a corresponding operation on a representation
   259     type.  Example syntax:
   260 
   261     lift_definition dlist_insert :: "'a => 'a dlist => 'a dlist"
   262       is List.insert
   263 
   264     Users must discharge a respectfulness proof obligation when each
   265     constant is defined. (For a type copy, i.e. a typedef with UNIV,
   266     the proof is discharged automatically.) The obligation is
   267     presented in a user-friendly, readable form; a respectfulness
   268     theorem in the standard format and a transfer rule are generated
   269     by the package.
   270 
   271   - Integration with code_abstype: For typedefs (e.g. subtypes
   272     corresponding to a datatype invariant, such as dlist),
   273     lift_definition generates a code certificate theorem and sets up
   274     code generation for each constant.
   275 
   276   - setup_lifting command: Sets up the Lifting package to work with a
   277     user-defined type. The user must provide either a quotient theorem
   278     or a type_definition theorem.  The package configures transfer
   279     rules for equality and quantifiers on the type, and sets up the
   280     lift_definition command to work with the type.
   281 
   282   - Usage examples: See Quotient_Examples/Lift_DList.thy,
   283     Quotient_Examples/Lift_RBT.thy, Quotient_Examples/Lift_FSet.thy,
   284     Word/Word.thy and Library/Float.thy.
   285 
   286 * Quotient package:
   287 
   288   - The 'quotient_type' command now supports a 'morphisms' option with
   289     rep and abs functions, similar to typedef.
   290 
   291   - 'quotient_type' sets up new types to work with the Lifting and
   292     Transfer packages, as with 'setup_lifting'.
   293 
   294   - The 'quotient_definition' command now requires the user to prove a
   295     respectfulness property at the point where the constant is
   296     defined, similar to lift_definition; INCOMPATIBILITY.
   297 
   298   - Renamed predicate 'Quotient' to 'Quotient3', and renamed theorems
   299     accordingly, INCOMPATIBILITY.
   300 
   301 * New diagnostic command 'find_unused_assms' to find potentially
   302 superfluous assumptions in theorems using Quickcheck.
   303 
   304 * Quickcheck:
   305 
   306   - Quickcheck returns variable assignments as counterexamples, which
   307     allows to reveal the underspecification of functions under test.
   308     For example, refuting "hd xs = x", it presents the variable
   309     assignment xs = [] and x = a1 as a counterexample, assuming that
   310     any property is false whenever "hd []" occurs in it.
   311 
   312     These counterexample are marked as potentially spurious, as
   313     Quickcheck also returns "xs = []" as a counterexample to the
   314     obvious theorem "hd xs = hd xs".
   315 
   316     After finding a potentially spurious counterexample, Quickcheck
   317     continues searching for genuine ones.
   318 
   319     By default, Quickcheck shows potentially spurious and genuine
   320     counterexamples. The option "genuine_only" sets quickcheck to only
   321     show genuine counterexamples.
   322 
   323   - The command 'quickcheck_generator' creates random and exhaustive
   324     value generators for a given type and operations.
   325 
   326     It generates values by using the operations as if they were
   327     constructors of that type.
   328 
   329   - Support for multisets.
   330 
   331   - Added "use_subtype" options.
   332 
   333   - Added "quickcheck_locale" configuration to specify how to process
   334     conjectures in a locale context.
   335 
   336 * Nitpick: Fixed infinite loop caused by the 'peephole_optim' option
   337 and affecting 'rat' and 'real'.
   338 
   339 * Sledgehammer:
   340   - Integrated more tightly with SPASS, as described in the ITP 2012
   341     paper "More SPASS with Isabelle".
   342   - Made it try "smt" as a fallback if "metis" fails or times out.
   343   - Added support for the following provers: Alt-Ergo (via Why3 and
   344     TFF1), iProver, iProver-Eq.
   345   - Replaced remote E-SInE with remote Satallax in the default setup.
   346   - Sped up the minimizer.
   347   - Added "lam_trans", "uncurry_aliases", and "minimize" options.
   348   - Renamed "slicing" ("no_slicing") option to "slice" ("dont_slice").
   349   - Renamed "sound" option to "strict".
   350 
   351 * Metis: Added possibility to specify lambda translations scheme as a
   352 parenthesized argument (e.g., "by (metis (lifting) ...)").
   353 
   354 * SMT: Renamed "smt_fixed" option to "smt_read_only_certificates".
   355 
   356 * Command 'try0': Renamed from 'try_methods'. INCOMPATIBILITY.
   357 
   358 * New "case_product" attribute to generate a case rule doing multiple
   359 case distinctions at the same time.  E.g.
   360 
   361   list.exhaust [case_product nat.exhaust]
   362 
   363 produces a rule which can be used to perform case distinction on both
   364 a list and a nat.
   365 
   366 * New "eventually_elim" method as a generalized variant of the
   367 eventually_elim* rules.  Supports structured proofs.
   368 
   369 * Typedef with implicit set definition is considered legacy.  Use
   370 "typedef (open)" form instead, which will eventually become the
   371 default.
   372 
   373 * Record: code generation can be switched off manually with
   374 
   375   declare [[record_coden = false]]  -- "default true"
   376 
   377 * Datatype: type parameters allow explicit sort constraints.
   378 
   379 * Concrete syntax for case expressions includes constraints for source
   380 positions, and thus produces Prover IDE markup for its bindings.
   381 INCOMPATIBILITY for old-style syntax translations that augment the
   382 pattern notation; e.g. see src/HOL/HOLCF/One.thy for translations of
   383 one_case.
   384 
   385 * Clarified attribute "mono_set": pure declaration without modifying
   386 the result of the fact expression.
   387 
   388 * More default pred/set conversions on a couple of relation operations
   389 and predicates.  Added powers of predicate relations.  Consolidation
   390 of some relation theorems:
   391 
   392   converse_def ~> converse_unfold
   393   rel_comp_def ~> relcomp_unfold
   394   symp_def ~> (modified, use symp_def and sym_def instead)
   395   transp_def ~> transp_trans
   396   Domain_def ~> Domain_unfold
   397   Range_def ~> Domain_converse [symmetric]
   398 
   399 Generalized theorems INF_INT_eq, INF_INT_eq2, SUP_UN_eq, SUP_UN_eq2.
   400 
   401 See theory "Relation" for examples for making use of pred/set
   402 conversions by means of attributes "to_set" and "to_pred".
   403 
   404 INCOMPATIBILITY.
   405 
   406 * Renamed facts about the power operation on relations, i.e., relpow
   407 to match the constant's name:
   408 
   409   rel_pow_1 ~> relpow_1
   410   rel_pow_0_I ~> relpow_0_I
   411   rel_pow_Suc_I ~> relpow_Suc_I
   412   rel_pow_Suc_I2 ~> relpow_Suc_I2
   413   rel_pow_0_E ~> relpow_0_E
   414   rel_pow_Suc_E ~> relpow_Suc_E
   415   rel_pow_E ~> relpow_E
   416   rel_pow_Suc_D2 ~> relpow_Suc_D2
   417   rel_pow_Suc_E2 ~> relpow_Suc_E2
   418   rel_pow_Suc_D2' ~> relpow_Suc_D2'
   419   rel_pow_E2 ~> relpow_E2
   420   rel_pow_add ~> relpow_add
   421   rel_pow_commute ~> relpow
   422   rel_pow_empty ~> relpow_empty:
   423   rtrancl_imp_UN_rel_pow ~> rtrancl_imp_UN_relpow
   424   rel_pow_imp_rtrancl ~> relpow_imp_rtrancl
   425   rtrancl_is_UN_rel_pow ~> rtrancl_is_UN_relpow
   426   rtrancl_imp_rel_pow ~> rtrancl_imp_relpow
   427   rel_pow_fun_conv ~> relpow_fun_conv
   428   rel_pow_finite_bounded1 ~> relpow_finite_bounded1
   429   rel_pow_finite_bounded ~> relpow_finite_bounded
   430   rtrancl_finite_eq_rel_pow ~> rtrancl_finite_eq_relpow
   431   trancl_finite_eq_rel_pow ~> trancl_finite_eq_relpow
   432   single_valued_rel_pow ~> single_valued_relpow
   433 
   434 INCOMPATIBILITY.
   435 
   436 * Theory Relation: Consolidated constant name for relation composition
   437 and corresponding theorem names:
   438 
   439   - Renamed constant rel_comp to relcomp.
   440 
   441   - Dropped abbreviation pred_comp. Use relcompp instead.
   442 
   443   - Renamed theorems:
   444 
   445     rel_compI ~> relcompI
   446     rel_compEpair ~> relcompEpair
   447     rel_compE ~> relcompE
   448     pred_comp_rel_comp_eq ~> relcompp_relcomp_eq
   449     rel_comp_empty1 ~> relcomp_empty1
   450     rel_comp_mono ~> relcomp_mono
   451     rel_comp_subset_Sigma ~> relcomp_subset_Sigma
   452     rel_comp_distrib ~> relcomp_distrib
   453     rel_comp_distrib2 ~> relcomp_distrib2
   454     rel_comp_UNION_distrib ~> relcomp_UNION_distrib
   455     rel_comp_UNION_distrib2 ~> relcomp_UNION_distrib2
   456     single_valued_rel_comp ~> single_valued_relcomp
   457     rel_comp_def ~> relcomp_unfold
   458     converse_rel_comp ~> converse_relcomp
   459     pred_compI ~> relcomppI
   460     pred_compE ~> relcomppE
   461     pred_comp_bot1 ~> relcompp_bot1
   462     pred_comp_bot2 ~> relcompp_bot2
   463     transp_pred_comp_less_eq ~> transp_relcompp_less_eq
   464     pred_comp_mono ~> relcompp_mono
   465     pred_comp_distrib ~> relcompp_distrib
   466     pred_comp_distrib2 ~> relcompp_distrib2
   467     converse_pred_comp ~> converse_relcompp
   468 
   469     finite_rel_comp ~> finite_relcomp
   470 
   471     set_rel_comp ~> set_relcomp
   472 
   473 INCOMPATIBILITY.
   474 
   475 * Theory Divides: Discontinued redundant theorems about div and mod.
   476 INCOMPATIBILITY, use the corresponding generic theorems instead.
   477 
   478   DIVISION_BY_ZERO ~> div_by_0, mod_by_0
   479   zdiv_self ~> div_self
   480   zmod_self ~> mod_self
   481   zdiv_zero ~> div_0
   482   zmod_zero ~> mod_0
   483   zdiv_zmod_equality ~> div_mod_equality2
   484   zdiv_zmod_equality2 ~> div_mod_equality
   485   zmod_zdiv_trivial ~> mod_div_trivial
   486   zdiv_zminus_zminus ~> div_minus_minus
   487   zmod_zminus_zminus ~> mod_minus_minus
   488   zdiv_zminus2 ~> div_minus_right
   489   zmod_zminus2 ~> mod_minus_right
   490   zdiv_minus1_right ~> div_minus1_right
   491   zmod_minus1_right ~> mod_minus1_right
   492   zdvd_mult_div_cancel ~> dvd_mult_div_cancel
   493   zmod_zmult1_eq ~> mod_mult_right_eq
   494   zpower_zmod ~> power_mod
   495   zdvd_zmod ~> dvd_mod
   496   zdvd_zmod_imp_zdvd ~> dvd_mod_imp_dvd
   497   mod_mult_distrib ~> mult_mod_left
   498   mod_mult_distrib2 ~> mult_mod_right
   499 
   500 * Removed redundant theorems nat_mult_2 and nat_mult_2_right; use
   501 generic mult_2 and mult_2_right instead. INCOMPATIBILITY.
   502 
   503 * Finite_Set.fold now qualified.  INCOMPATIBILITY.
   504 
   505 * Consolidated theorem names concerning fold combinators:
   506 
   507   inf_INFI_fold_inf ~> inf_INF_fold_inf
   508   sup_SUPR_fold_sup ~> sup_SUP_fold_sup
   509   INFI_fold_inf ~> INF_fold_inf
   510   SUPR_fold_sup ~> SUP_fold_sup
   511   union_set ~> union_set_fold
   512   minus_set ~> minus_set_fold
   513   INFI_set_fold ~> INF_set_fold
   514   SUPR_set_fold ~> SUP_set_fold
   515   INF_code ~> INF_set_foldr
   516   SUP_code ~> SUP_set_foldr
   517   foldr.simps ~> foldr.simps (in point-free formulation)
   518   foldr_fold_rev ~> foldr_conv_fold
   519   foldl_fold ~> foldl_conv_fold
   520   foldr_foldr ~> foldr_conv_foldl
   521   foldl_foldr ~> foldl_conv_foldr
   522   fold_set_remdups ~> fold_set_fold_remdups
   523   fold_set ~> fold_set_fold
   524   fold1_set ~> fold1_set_fold
   525 
   526 INCOMPATIBILITY.
   527 
   528 * Dropped rarely useful theorems concerning fold combinators:
   529 foldl_apply, foldl_fun_comm, foldl_rev, fold_weak_invariant,
   530 rev_foldl_cons, fold_set_remdups, fold_set, fold_set1,
   531 concat_conv_foldl, foldl_weak_invariant, foldl_invariant,
   532 foldr_invariant, foldl_absorb0, foldl_foldr1_lemma, foldl_foldr1,
   533 listsum_conv_fold, listsum_foldl, sort_foldl_insort, foldl_assoc,
   534 foldr_conv_foldl, start_le_sum, elem_le_sum, sum_eq_0_conv.
   535 INCOMPATIBILITY.  For the common phrases "%xs. List.foldr plus xs 0"
   536 and "List.foldl plus 0", prefer "List.listsum".  Otherwise it can be
   537 useful to boil down "List.foldr" and "List.foldl" to "List.fold" by
   538 unfolding "foldr_conv_fold" and "foldl_conv_fold".
   539 
   540 * Dropped lemmas minus_set_foldr, union_set_foldr, union_coset_foldr,
   541 inter_coset_foldr, Inf_fin_set_foldr, Sup_fin_set_foldr,
   542 Min_fin_set_foldr, Max_fin_set_foldr, Inf_set_foldr, Sup_set_foldr,
   543 INF_set_foldr, SUP_set_foldr.  INCOMPATIBILITY.  Prefer corresponding
   544 lemmas over fold rather than foldr, or make use of lemmas
   545 fold_conv_foldr and fold_rev.
   546 
   547 * Congruence rules Option.map_cong and Option.bind_cong for recursion
   548 through option types.
   549 
   550 * "Transitive_Closure.ntrancl": bounded transitive closure on
   551 relations.
   552 
   553 * Constant "Set.not_member" now qualified.  INCOMPATIBILITY.
   554 
   555 * Theory Int: Discontinued many legacy theorems specific to type int.
   556 INCOMPATIBILITY, use the corresponding generic theorems instead.
   557 
   558   zminus_zminus ~> minus_minus
   559   zminus_0 ~> minus_zero
   560   zminus_zadd_distrib ~> minus_add_distrib
   561   zadd_commute ~> add_commute
   562   zadd_assoc ~> add_assoc
   563   zadd_left_commute ~> add_left_commute
   564   zadd_ac ~> add_ac
   565   zmult_ac ~> mult_ac
   566   zadd_0 ~> add_0_left
   567   zadd_0_right ~> add_0_right
   568   zadd_zminus_inverse2 ~> left_minus
   569   zmult_zminus ~> mult_minus_left
   570   zmult_commute ~> mult_commute
   571   zmult_assoc ~> mult_assoc
   572   zadd_zmult_distrib ~> left_distrib
   573   zadd_zmult_distrib2 ~> right_distrib
   574   zdiff_zmult_distrib ~> left_diff_distrib
   575   zdiff_zmult_distrib2 ~> right_diff_distrib
   576   zmult_1 ~> mult_1_left
   577   zmult_1_right ~> mult_1_right
   578   zle_refl ~> order_refl
   579   zle_trans ~> order_trans
   580   zle_antisym ~> order_antisym
   581   zle_linear ~> linorder_linear
   582   zless_linear ~> linorder_less_linear
   583   zadd_left_mono ~> add_left_mono
   584   zadd_strict_right_mono ~> add_strict_right_mono
   585   zadd_zless_mono ~> add_less_le_mono
   586   int_0_less_1 ~> zero_less_one
   587   int_0_neq_1 ~> zero_neq_one
   588   zless_le ~> less_le
   589   zpower_zadd_distrib ~> power_add
   590   zero_less_zpower_abs_iff ~> zero_less_power_abs_iff
   591   zero_le_zpower_abs ~> zero_le_power_abs
   592 
   593 * Theory Deriv: Renamed
   594 
   595   DERIV_nonneg_imp_nonincreasing ~> DERIV_nonneg_imp_nondecreasing
   596 
   597 * Theory Library/Multiset: Improved code generation of multisets.
   598 
   599 * Theory HOL/Library/Set_Algebras: Addition and multiplication on sets
   600 are expressed via type classes again. The special syntax
   601 \<oplus>/\<otimes> has been replaced by plain +/*. Removed constant
   602 setsum_set, which is now subsumed by Big_Operators.setsum.
   603 INCOMPATIBILITY.
   604 
   605 * Theory HOL/Library/Diagonalize has been removed. INCOMPATIBILITY,
   606 use theory HOL/Library/Nat_Bijection instead.
   607 
   608 * Theory HOL/Library/RBT_Impl: Backing implementation of red-black
   609 trees is now inside a type class context.  Names of affected
   610 operations and lemmas have been prefixed by rbt_.  INCOMPATIBILITY for
   611 theories working directly with raw red-black trees, adapt the names as
   612 follows:
   613 
   614   Operations:
   615   bulkload -> rbt_bulkload
   616   del_from_left -> rbt_del_from_left
   617   del_from_right -> rbt_del_from_right
   618   del -> rbt_del
   619   delete -> rbt_delete
   620   ins -> rbt_ins
   621   insert -> rbt_insert
   622   insertw -> rbt_insert_with
   623   insert_with_key -> rbt_insert_with_key
   624   map_entry -> rbt_map_entry
   625   lookup -> rbt_lookup
   626   sorted -> rbt_sorted
   627   tree_greater -> rbt_greater
   628   tree_less -> rbt_less
   629   tree_less_symbol -> rbt_less_symbol
   630   union -> rbt_union
   631   union_with -> rbt_union_with
   632   union_with_key -> rbt_union_with_key
   633 
   634   Lemmas:
   635   balance_left_sorted -> balance_left_rbt_sorted
   636   balance_left_tree_greater -> balance_left_rbt_greater
   637   balance_left_tree_less -> balance_left_rbt_less
   638   balance_right_sorted -> balance_right_rbt_sorted
   639   balance_right_tree_greater -> balance_right_rbt_greater
   640   balance_right_tree_less -> balance_right_rbt_less
   641   balance_sorted -> balance_rbt_sorted
   642   balance_tree_greater -> balance_rbt_greater
   643   balance_tree_less -> balance_rbt_less
   644   bulkload_is_rbt -> rbt_bulkload_is_rbt
   645   combine_sorted -> combine_rbt_sorted
   646   combine_tree_greater -> combine_rbt_greater
   647   combine_tree_less -> combine_rbt_less
   648   delete_in_tree -> rbt_delete_in_tree
   649   delete_is_rbt -> rbt_delete_is_rbt
   650   del_from_left_tree_greater -> rbt_del_from_left_rbt_greater
   651   del_from_left_tree_less -> rbt_del_from_left_rbt_less
   652   del_from_right_tree_greater -> rbt_del_from_right_rbt_greater
   653   del_from_right_tree_less -> rbt_del_from_right_rbt_less
   654   del_in_tree -> rbt_del_in_tree
   655   del_inv1_inv2 -> rbt_del_inv1_inv2
   656   del_sorted -> rbt_del_rbt_sorted
   657   del_tree_greater -> rbt_del_rbt_greater
   658   del_tree_less -> rbt_del_rbt_less
   659   dom_lookup_Branch -> dom_rbt_lookup_Branch
   660   entries_lookup -> entries_rbt_lookup
   661   finite_dom_lookup -> finite_dom_rbt_lookup
   662   insert_sorted -> rbt_insert_rbt_sorted
   663   insertw_is_rbt -> rbt_insertw_is_rbt
   664   insertwk_is_rbt -> rbt_insertwk_is_rbt
   665   insertwk_sorted -> rbt_insertwk_rbt_sorted
   666   insertw_sorted -> rbt_insertw_rbt_sorted
   667   ins_sorted -> ins_rbt_sorted
   668   ins_tree_greater -> ins_rbt_greater
   669   ins_tree_less -> ins_rbt_less
   670   is_rbt_sorted -> is_rbt_rbt_sorted
   671   lookup_balance -> rbt_lookup_balance
   672   lookup_bulkload -> rbt_lookup_rbt_bulkload
   673   lookup_delete -> rbt_lookup_rbt_delete
   674   lookup_Empty -> rbt_lookup_Empty
   675   lookup_from_in_tree -> rbt_lookup_from_in_tree
   676   lookup_in_tree -> rbt_lookup_in_tree
   677   lookup_ins -> rbt_lookup_ins
   678   lookup_insert -> rbt_lookup_rbt_insert
   679   lookup_insertw -> rbt_lookup_rbt_insertw
   680   lookup_insertwk -> rbt_lookup_rbt_insertwk
   681   lookup_keys -> rbt_lookup_keys
   682   lookup_map -> rbt_lookup_map
   683   lookup_map_entry -> rbt_lookup_rbt_map_entry
   684   lookup_tree_greater -> rbt_lookup_rbt_greater
   685   lookup_tree_less -> rbt_lookup_rbt_less
   686   lookup_union -> rbt_lookup_rbt_union
   687   map_entry_color_of -> rbt_map_entry_color_of
   688   map_entry_inv1 -> rbt_map_entry_inv1
   689   map_entry_inv2 -> rbt_map_entry_inv2
   690   map_entry_is_rbt -> rbt_map_entry_is_rbt
   691   map_entry_sorted -> rbt_map_entry_rbt_sorted
   692   map_entry_tree_greater -> rbt_map_entry_rbt_greater
   693   map_entry_tree_less -> rbt_map_entry_rbt_less
   694   map_tree_greater -> map_rbt_greater
   695   map_tree_less -> map_rbt_less
   696   map_sorted -> map_rbt_sorted
   697   paint_sorted -> paint_rbt_sorted
   698   paint_lookup -> paint_rbt_lookup
   699   paint_tree_greater -> paint_rbt_greater
   700   paint_tree_less -> paint_rbt_less
   701   sorted_entries -> rbt_sorted_entries
   702   tree_greater_eq_trans -> rbt_greater_eq_trans
   703   tree_greater_nit -> rbt_greater_nit
   704   tree_greater_prop -> rbt_greater_prop
   705   tree_greater_simps -> rbt_greater_simps
   706   tree_greater_trans -> rbt_greater_trans
   707   tree_less_eq_trans -> rbt_less_eq_trans
   708   tree_less_nit -> rbt_less_nit
   709   tree_less_prop -> rbt_less_prop
   710   tree_less_simps -> rbt_less_simps
   711   tree_less_trans -> rbt_less_trans
   712   tree_ord_props -> rbt_ord_props
   713   union_Branch -> rbt_union_Branch
   714   union_is_rbt -> rbt_union_is_rbt
   715   unionw_is_rbt -> rbt_unionw_is_rbt
   716   unionwk_is_rbt -> rbt_unionwk_is_rbt
   717   unionwk_sorted -> rbt_unionwk_rbt_sorted
   718 
   719 * Theory HOL/Library/Float: Floating point numbers are now defined as
   720 a subset of the real numbers.  All operations are defined using the
   721 lifing-framework and proofs use the transfer method.  INCOMPATIBILITY.
   722 
   723   Changed Operations:
   724   float_abs -> abs
   725   float_nprt -> nprt
   726   float_pprt -> pprt
   727   pow2 -> use powr
   728   round_down -> float_round_down
   729   round_up -> float_round_up
   730   scale -> exponent
   731 
   732   Removed Operations:
   733   ceiling_fl, lb_mult, lb_mod, ub_mult, ub_mod
   734 
   735   Renamed Lemmas:
   736   abs_float_def -> Float.compute_float_abs
   737   bitlen_ge0 -> bitlen_nonneg
   738   bitlen.simps -> Float.compute_bitlen
   739   float_components -> Float_mantissa_exponent
   740   float_divl.simps -> Float.compute_float_divl
   741   float_divr.simps -> Float.compute_float_divr
   742   float_eq_odd -> mult_powr_eq_mult_powr_iff
   743   float_power -> real_of_float_power
   744   lapprox_posrat_def -> Float.compute_lapprox_posrat
   745   lapprox_rat.simps -> Float.compute_lapprox_rat
   746   le_float_def' -> Float.compute_float_le
   747   le_float_def -> less_eq_float.rep_eq
   748   less_float_def' -> Float.compute_float_less
   749   less_float_def -> less_float.rep_eq
   750   normfloat_def -> Float.compute_normfloat
   751   normfloat_imp_odd_or_zero -> mantissa_not_dvd and mantissa_noteq_0
   752   normfloat -> normfloat_def
   753   normfloat_unique -> use normfloat_def
   754   number_of_float_Float -> Float.compute_float_numeral, Float.compute_float_neg_numeral
   755   one_float_def -> Float.compute_float_one
   756   plus_float_def -> Float.compute_float_plus
   757   rapprox_posrat_def -> Float.compute_rapprox_posrat
   758   rapprox_rat.simps -> Float.compute_rapprox_rat
   759   real_of_float_0 -> zero_float.rep_eq
   760   real_of_float_1 -> one_float.rep_eq
   761   real_of_float_abs -> abs_float.rep_eq
   762   real_of_float_add -> plus_float.rep_eq
   763   real_of_float_minus -> uminus_float.rep_eq
   764   real_of_float_mult -> times_float.rep_eq
   765   real_of_float_simp -> Float.rep_eq
   766   real_of_float_sub -> minus_float.rep_eq
   767   round_down.simps -> Float.compute_float_round_down
   768   round_up.simps -> Float.compute_float_round_up
   769   times_float_def -> Float.compute_float_times
   770   uminus_float_def -> Float.compute_float_uminus
   771   zero_float_def -> Float.compute_float_zero
   772 
   773   Lemmas not necessary anymore, use the transfer method:
   774   bitlen_B0, bitlen_B1, bitlen_ge1, bitlen_Min, bitlen_Pls, float_divl,
   775   float_divr, float_le_simp, float_less1_mantissa_bound,
   776   float_less_simp, float_less_zero, float_le_zero,
   777   float_pos_less1_e_neg, float_pos_m_pos, float_split, float_split2,
   778   floor_pos_exp, lapprox_posrat, lapprox_posrat_bottom, lapprox_rat,
   779   lapprox_rat_bottom, normalized_float, rapprox_posrat,
   780   rapprox_posrat_le1, rapprox_rat, real_of_float_ge0_exp,
   781   real_of_float_neg_exp, real_of_float_nge0_exp, round_down floor_fl,
   782   round_up, zero_le_float, zero_less_float
   783 
   784 * New theory HOL/Library/DAList provides an abstract type for
   785 association lists with distinct keys.
   786 
   787 * Session HOL/IMP: Added new theory of abstract interpretation of
   788 annotated commands.
   789 
   790 * Session HOL-Import: Re-implementation from scratch is faster,
   791 simpler, and more scalable.  Requires a proof bundle, which is
   792 available as an external component.  Discontinued old (and mostly
   793 dead) Importer for HOL4 and HOL Light.  INCOMPATIBILITY.
   794 
   795 * Session HOL-Word: Discontinued many redundant theorems specific to
   796 type 'a word. INCOMPATIBILITY, use the corresponding generic theorems
   797 instead.
   798 
   799   word_sub_alt ~> word_sub_wi
   800   word_add_alt ~> word_add_def
   801   word_mult_alt ~> word_mult_def
   802   word_minus_alt ~> word_minus_def
   803   word_0_alt ~> word_0_wi
   804   word_1_alt ~> word_1_wi
   805   word_add_0 ~> add_0_left
   806   word_add_0_right ~> add_0_right
   807   word_mult_1 ~> mult_1_left
   808   word_mult_1_right ~> mult_1_right
   809   word_add_commute ~> add_commute
   810   word_add_assoc ~> add_assoc
   811   word_add_left_commute ~> add_left_commute
   812   word_mult_commute ~> mult_commute
   813   word_mult_assoc ~> mult_assoc
   814   word_mult_left_commute ~> mult_left_commute
   815   word_left_distrib ~> left_distrib
   816   word_right_distrib ~> right_distrib
   817   word_left_minus ~> left_minus
   818   word_diff_0_right ~> diff_0_right
   819   word_diff_self ~> diff_self
   820   word_sub_def ~> diff_minus
   821   word_diff_minus ~> diff_minus
   822   word_add_ac ~> add_ac
   823   word_mult_ac ~> mult_ac
   824   word_plus_ac0 ~> add_0_left add_0_right add_ac
   825   word_times_ac1 ~> mult_1_left mult_1_right mult_ac
   826   word_order_trans ~> order_trans
   827   word_order_refl ~> order_refl
   828   word_order_antisym ~> order_antisym
   829   word_order_linear ~> linorder_linear
   830   lenw1_zero_neq_one ~> zero_neq_one
   831   word_number_of_eq ~> number_of_eq
   832   word_of_int_add_hom ~> wi_hom_add
   833   word_of_int_sub_hom ~> wi_hom_sub
   834   word_of_int_mult_hom ~> wi_hom_mult
   835   word_of_int_minus_hom ~> wi_hom_neg
   836   word_of_int_succ_hom ~> wi_hom_succ
   837   word_of_int_pred_hom ~> wi_hom_pred
   838   word_of_int_0_hom ~> word_0_wi
   839   word_of_int_1_hom ~> word_1_wi
   840 
   841 * Session HOL-Word: New proof method "word_bitwise" for splitting
   842 machine word equalities and inequalities into logical circuits,
   843 defined in HOL/Word/WordBitwise.thy.  Supports addition, subtraction,
   844 multiplication, shifting by constants, bitwise operators and numeric
   845 constants.  Requires fixed-length word types, not 'a word.  Solves
   846 many standard word identities outright and converts more into first
   847 order problems amenable to blast or similar.  See also examples in
   848 HOL/Word/Examples/WordExamples.thy.
   849 
   850 * Session HOL-Probability: Introduced the type "'a measure" to
   851 represent measures, this replaces the records 'a algebra and 'a
   852 measure_space.  The locales based on subset_class now have two
   853 locale-parameters the space \<Omega> and the set of measurable sets M.
   854 The product of probability spaces uses now the same constant as the
   855 finite product of sigma-finite measure spaces "PiM :: ('i => 'a)
   856 measure".  Most constants are defined now outside of locales and gain
   857 an additional parameter, like null_sets, almost_eventually or \<mu>'.
   858 Measure space constructions for distributions and densities now got
   859 their own constants distr and density.  Instead of using locales to
   860 describe measure spaces with a finite space, the measure count_space
   861 and point_measure is introduced.  INCOMPATIBILITY.
   862 
   863   Renamed constants:
   864   measure -> emeasure
   865   finite_measure.\<mu>' -> measure
   866   product_algebra_generator -> prod_algebra
   867   product_prob_space.emb -> prod_emb
   868   product_prob_space.infprod_algebra -> PiM
   869 
   870   Removed locales:
   871   completeable_measure_space
   872   finite_measure_space
   873   finite_prob_space
   874   finite_product_finite_prob_space
   875   finite_product_sigma_algebra
   876   finite_sigma_algebra
   877   measure_space
   878   pair_finite_prob_space
   879   pair_finite_sigma_algebra
   880   pair_finite_space
   881   pair_sigma_algebra
   882   product_sigma_algebra
   883 
   884   Removed constants:
   885   conditional_space
   886   distribution -> use distr measure, or distributed predicate
   887   image_space
   888   joint_distribution -> use distr measure, or distributed predicate
   889   pair_measure_generator
   890   product_prob_space.infprod_algebra -> use PiM
   891   subvimage
   892 
   893   Replacement theorems:
   894   finite_additivity_sufficient -> ring_of_sets.countably_additiveI_finite
   895   finite_measure.empty_measure -> measure_empty
   896   finite_measure.finite_continuity_from_above -> finite_measure.finite_Lim_measure_decseq
   897   finite_measure.finite_continuity_from_below -> finite_measure.finite_Lim_measure_incseq
   898   finite_measure.finite_measure_countably_subadditive -> finite_measure.finite_measure_subadditive_countably
   899   finite_measure.finite_measure_eq -> finite_measure.emeasure_eq_measure
   900   finite_measure.finite_measure -> finite_measure.emeasure_finite
   901   finite_measure.finite_measure_finite_singleton -> finite_measure.finite_measure_eq_setsum_singleton
   902   finite_measure.positive_measure' -> measure_nonneg
   903   finite_measure.real_measure -> finite_measure.emeasure_real
   904   finite_product_prob_space.finite_measure_times -> finite_product_prob_space.finite_measure_PiM_emb
   905   finite_product_sigma_algebra.in_P -> sets_PiM_I_finite
   906   finite_product_sigma_algebra.P_empty -> space_PiM_empty, sets_PiM_empty
   907   information_space.conditional_entropy_eq -> information_space.conditional_entropy_simple_distributed
   908   information_space.conditional_entropy_positive -> information_space.conditional_entropy_nonneg_simple
   909   information_space.conditional_mutual_information_eq_mutual_information -> information_space.conditional_mutual_information_eq_mutual_information_simple
   910   information_space.conditional_mutual_information_generic_positive -> information_space.conditional_mutual_information_nonneg_simple
   911   information_space.conditional_mutual_information_positive -> information_space.conditional_mutual_information_nonneg_simple
   912   information_space.entropy_commute -> information_space.entropy_commute_simple
   913   information_space.entropy_eq -> information_space.entropy_simple_distributed
   914   information_space.entropy_generic_eq -> information_space.entropy_simple_distributed
   915   information_space.entropy_positive -> information_space.entropy_nonneg_simple
   916   information_space.entropy_uniform_max -> information_space.entropy_uniform
   917   information_space.KL_eq_0_imp -> information_space.KL_eq_0_iff_eq
   918   information_space.KL_eq_0 -> information_space.KL_same_eq_0
   919   information_space.KL_ge_0 -> information_space.KL_nonneg
   920   information_space.mutual_information_eq -> information_space.mutual_information_simple_distributed
   921   information_space.mutual_information_positive -> information_space.mutual_information_nonneg_simple
   922   Int_stable_cuboids -> Int_stable_atLeastAtMost
   923   Int_stable_product_algebra_generator -> positive_integral
   924   measure_preserving -> equality "distr M N f = N" "f : measurable M N"
   925   measure_space.additive -> emeasure_additive
   926   measure_space.AE_iff_null_set -> AE_iff_null
   927   measure_space.almost_everywhere_def -> eventually_ae_filter
   928   measure_space.almost_everywhere_vimage -> AE_distrD
   929   measure_space.continuity_from_above -> INF_emeasure_decseq
   930   measure_space.continuity_from_above_Lim -> Lim_emeasure_decseq
   931   measure_space.continuity_from_below_Lim -> Lim_emeasure_incseq
   932   measure_space.continuity_from_below -> SUP_emeasure_incseq
   933   measure_space_density -> emeasure_density
   934   measure_space.density_is_absolutely_continuous -> absolutely_continuousI_density
   935   measure_space.integrable_vimage -> integrable_distr
   936   measure_space.integral_translated_density -> integral_density
   937   measure_space.integral_vimage -> integral_distr
   938   measure_space.measure_additive -> plus_emeasure
   939   measure_space.measure_compl -> emeasure_compl
   940   measure_space.measure_countable_increasing -> emeasure_countable_increasing
   941   measure_space.measure_countably_subadditive -> emeasure_subadditive_countably
   942   measure_space.measure_decseq -> decseq_emeasure
   943   measure_space.measure_Diff -> emeasure_Diff
   944   measure_space.measure_Diff_null_set -> emeasure_Diff_null_set
   945   measure_space.measure_eq_0 -> emeasure_eq_0
   946   measure_space.measure_finitely_subadditive -> emeasure_subadditive_finite
   947   measure_space.measure_finite_singleton -> emeasure_eq_setsum_singleton
   948   measure_space.measure_incseq -> incseq_emeasure
   949   measure_space.measure_insert -> emeasure_insert
   950   measure_space.measure_mono -> emeasure_mono
   951   measure_space.measure_not_negative -> emeasure_not_MInf
   952   measure_space.measure_preserving_Int_stable -> measure_eqI_generator_eq
   953   measure_space.measure_setsum -> setsum_emeasure
   954   measure_space.measure_setsum_split -> setsum_emeasure_cover
   955   measure_space.measure_space_vimage -> emeasure_distr
   956   measure_space.measure_subadditive_finite -> emeasure_subadditive_finite
   957   measure_space.measure_subadditive -> subadditive
   958   measure_space.measure_top -> emeasure_space
   959   measure_space.measure_UN_eq_0 -> emeasure_UN_eq_0
   960   measure_space.measure_Un_null_set -> emeasure_Un_null_set
   961   measure_space.positive_integral_translated_density -> positive_integral_density
   962   measure_space.positive_integral_vimage -> positive_integral_distr
   963   measure_space.real_continuity_from_above -> Lim_measure_decseq
   964   measure_space.real_continuity_from_below -> Lim_measure_incseq
   965   measure_space.real_measure_countably_subadditive -> measure_subadditive_countably
   966   measure_space.real_measure_Diff -> measure_Diff
   967   measure_space.real_measure_finite_Union -> measure_finite_Union
   968   measure_space.real_measure_setsum_singleton -> measure_eq_setsum_singleton
   969   measure_space.real_measure_subadditive -> measure_subadditive
   970   measure_space.real_measure_Union -> measure_Union
   971   measure_space.real_measure_UNION -> measure_UNION
   972   measure_space.simple_function_vimage -> simple_function_comp
   973   measure_space.simple_integral_vimage -> simple_integral_distr
   974   measure_space.simple_integral_vimage -> simple_integral_distr
   975   measure_unique_Int_stable -> measure_eqI_generator_eq
   976   measure_unique_Int_stable_vimage -> measure_eqI_generator_eq
   977   pair_sigma_algebra.measurable_cut_fst -> sets_Pair1
   978   pair_sigma_algebra.measurable_cut_snd -> sets_Pair2
   979   pair_sigma_algebra.measurable_pair_image_fst -> measurable_Pair1
   980   pair_sigma_algebra.measurable_pair_image_snd -> measurable_Pair2
   981   pair_sigma_algebra.measurable_product_swap -> measurable_pair_swap_iff
   982   pair_sigma_algebra.pair_sigma_algebra_measurable -> measurable_pair_swap
   983   pair_sigma_algebra.pair_sigma_algebra_swap_measurable -> measurable_pair_swap'
   984   pair_sigma_algebra.sets_swap -> sets_pair_swap
   985   pair_sigma_finite.measure_cut_measurable_fst -> pair_sigma_finite.measurable_emeasure_Pair1
   986   pair_sigma_finite.measure_cut_measurable_snd -> pair_sigma_finite.measurable_emeasure_Pair2
   987   pair_sigma_finite.measure_preserving_swap -> pair_sigma_finite.distr_pair_swap
   988   pair_sigma_finite.pair_measure_alt2 -> pair_sigma_finite.emeasure_pair_measure_alt2
   989   pair_sigma_finite.pair_measure_alt -> pair_sigma_finite.emeasure_pair_measure_alt
   990   pair_sigma_finite.pair_measure_times -> pair_sigma_finite.emeasure_pair_measure_Times
   991   prob_space.indep_distribution_eq_measure -> prob_space.indep_vars_iff_distr_eq_PiM
   992   prob_space.indep_var_distributionD -> prob_space.indep_var_distribution_eq
   993   prob_space.measure_space_1 -> prob_space.emeasure_space_1
   994   prob_space.prob_space_vimage -> prob_space_distr
   995   prob_space.random_variable_restrict -> measurable_restrict
   996   prob_space_unique_Int_stable -> measure_eqI_prob_space
   997   product_algebraE -> prod_algebraE_all
   998   product_algebra_generator_der -> prod_algebra_eq_finite
   999   product_algebra_generator_into_space -> prod_algebra_sets_into_space
  1000   product_algebraI -> sets_PiM_I_finite
  1001   product_measure_exists -> product_sigma_finite.sigma_finite
  1002   product_prob_space.finite_index_eq_finite_product -> product_prob_space.sets_PiM_generator
  1003   product_prob_space.finite_measure_infprod_emb_Pi -> product_prob_space.measure_PiM_emb
  1004   product_prob_space.infprod_spec -> product_prob_space.emeasure_PiM_emb_not_empty
  1005   product_prob_space.measurable_component -> measurable_component_singleton
  1006   product_prob_space.measurable_emb -> measurable_prod_emb
  1007   product_prob_space.measurable_into_infprod_algebra -> measurable_PiM_single
  1008   product_prob_space.measurable_singleton_infprod -> measurable_component_singleton
  1009   product_prob_space.measure_emb -> emeasure_prod_emb
  1010   product_prob_space.measure_preserving_restrict -> product_prob_space.distr_restrict
  1011   product_sigma_algebra.product_algebra_into_space -> space_closed
  1012   product_sigma_finite.measure_fold -> product_sigma_finite.distr_merge
  1013   product_sigma_finite.measure_preserving_component_singelton -> product_sigma_finite.distr_singleton
  1014   product_sigma_finite.measure_preserving_merge -> product_sigma_finite.distr_merge
  1015   sequence_space.measure_infprod -> sequence_space.measure_PiM_countable
  1016   sets_product_algebra -> sets_PiM
  1017   sigma_algebra.measurable_sigma -> measurable_measure_of
  1018   sigma_finite_measure.disjoint_sigma_finite -> sigma_finite_disjoint
  1019   sigma_finite_measure.RN_deriv_vimage -> sigma_finite_measure.RN_deriv_distr
  1020   sigma_product_algebra_sigma_eq -> sigma_prod_algebra_sigma_eq
  1021   space_product_algebra -> space_PiM
  1022 
  1023 * Session HOL-TPTP: support to parse and import TPTP problems (all
  1024 languages) into Isabelle/HOL.
  1025 
  1026 
  1027 *** FOL ***
  1028 
  1029 * New "case_product" attribute (see HOL).
  1030 
  1031 
  1032 *** ZF ***
  1033 
  1034 * Greater support for structured proofs involving induction or case
  1035 analysis.
  1036 
  1037 * Much greater use of mathematical symbols.
  1038 
  1039 * Removal of many ML theorem bindings.  INCOMPATIBILITY.
  1040 
  1041 
  1042 *** ML ***
  1043 
  1044 * Antiquotation @{keyword "name"} produces a parser for outer syntax
  1045 from a minor keyword introduced via theory header declaration.
  1046 
  1047 * Antiquotation @{command_spec "name"} produces the
  1048 Outer_Syntax.command_spec from a major keyword introduced via theory
  1049 header declaration; it can be passed to Outer_Syntax.command etc.
  1050 
  1051 * Local_Theory.define no longer hard-wires default theorem name
  1052 "foo_def", but retains the binding as given.  If that is Binding.empty
  1053 / Attrib.empty_binding, the result is not registered as user-level
  1054 fact.  The Local_Theory.define_internal variant allows to specify a
  1055 non-empty name (used for the foundation in the background theory),
  1056 while omitting the fact binding in the user-context.  Potential
  1057 INCOMPATIBILITY for derived definitional packages: need to specify
  1058 naming policy for primitive definitions more explicitly.
  1059 
  1060 * Renamed Thm.capply to Thm.apply, and Thm.cabs to Thm.lambda in
  1061 conformance with similar operations in structure Term and Logic.
  1062 
  1063 * Antiquotation @{attributes [...]} embeds attribute source
  1064 representation into the ML text, which is particularly useful with
  1065 declarations like Local_Theory.note.
  1066 
  1067 * Structure Proof_Context follows standard naming scheme.  Old
  1068 ProofContext has been discontinued.  INCOMPATIBILITY.
  1069 
  1070 * Refined Local_Theory.declaration {syntax, pervasive}, with subtle
  1071 change of semantics: update is applied to auxiliary local theory
  1072 context as well.
  1073 
  1074 * Modernized some old-style infix operations:
  1075 
  1076   addeqcongs    ~> Simplifier.add_eqcong
  1077   deleqcongs    ~> Simplifier.del_eqcong
  1078   addcongs      ~> Simplifier.add_cong
  1079   delcongs      ~> Simplifier.del_cong
  1080   setmksimps    ~> Simplifier.set_mksimps
  1081   setmkcong     ~> Simplifier.set_mkcong
  1082   setmksym      ~> Simplifier.set_mksym
  1083   setmkeqTrue   ~> Simplifier.set_mkeqTrue
  1084   settermless   ~> Simplifier.set_termless
  1085   setsubgoaler  ~> Simplifier.set_subgoaler
  1086   addsplits     ~> Splitter.add_split
  1087   delsplits     ~> Splitter.del_split
  1088 
  1089 
  1090 *** System ***
  1091 
  1092 * USER_HOME settings variable points to cross-platform user home
  1093 directory, which coincides with HOME on POSIX systems only.  Likewise,
  1094 the Isabelle path specification "~" now expands to $USER_HOME, instead
  1095 of former $HOME.  A different default for USER_HOME may be set
  1096 explicitly in shell environment, before Isabelle settings are
  1097 evaluated.  Minor INCOMPATIBILITY: need to adapt Isabelle path where
  1098 the generic user home was intended.
  1099 
  1100 * ISABELLE_HOME_WINDOWS refers to ISABELLE_HOME in windows file name
  1101 notation, which is useful for the jEdit file browser, for example.
  1102 
  1103 * ISABELLE_JDK_HOME settings variable points to JDK with javac and jar
  1104 (not just JRE).
  1105 
  1106 
  1107 
  1108 New in Isabelle2011-1 (October 2011)
  1109 ------------------------------------
  1110 
  1111 *** General ***
  1112 
  1113 * Improved Isabelle/jEdit Prover IDE (PIDE), which can be invoked as
  1114 "isabelle jedit" or "ISABELLE_HOME/Isabelle" on the command line.
  1115 
  1116   - Management of multiple theory files directly from the editor
  1117     buffer store -- bypassing the file-system (no requirement to save
  1118     files for checking).
  1119 
  1120   - Markup of formal entities within the text buffer, with semantic
  1121     highlighting, tooltips and hyperlinks to jump to defining source
  1122     positions.
  1123 
  1124   - Improved text rendering, with sub/superscripts in the source
  1125     buffer (including support for copy/paste wrt. output panel, HTML
  1126     theory output and other non-Isabelle text boxes).
  1127 
  1128   - Refined scheduling of proof checking and printing of results,
  1129     based on interactive editor view.  (Note: jEdit folding and
  1130     narrowing allows to restrict buffer perspectives explicitly.)
  1131 
  1132   - Reduced CPU performance requirements, usable on machines with few
  1133     cores.
  1134 
  1135   - Reduced memory requirements due to pruning of unused document
  1136     versions (garbage collection).
  1137 
  1138 See also ~~/src/Tools/jEdit/README.html for further information,
  1139 including some remaining limitations.
  1140 
  1141 * Theory loader: source files are exclusively located via the master
  1142 directory of each theory node (where the .thy file itself resides).
  1143 The global load path (such as src/HOL/Library) has been discontinued.
  1144 Note that the path element ~~ may be used to reference theories in the
  1145 Isabelle home folder -- for instance, "~~/src/HOL/Library/FuncSet".
  1146 INCOMPATIBILITY.
  1147 
  1148 * Theory loader: source files are identified by content via SHA1
  1149 digests.  Discontinued former path/modtime identification and optional
  1150 ISABELLE_FILE_IDENT plugin scripts.
  1151 
  1152 * Parallelization of nested Isar proofs is subject to
  1153 Goal.parallel_proofs_threshold (default 100).  See also isabelle
  1154 usedir option -Q.
  1155 
  1156 * Name space: former unsynchronized references are now proper
  1157 configuration options, with more conventional names:
  1158 
  1159   long_names   ~> names_long
  1160   short_names  ~> names_short
  1161   unique_names ~> names_unique
  1162 
  1163 Minor INCOMPATIBILITY, need to declare options in context like this:
  1164 
  1165   declare [[names_unique = false]]
  1166 
  1167 * Literal facts `prop` may contain dummy patterns, e.g. `_ = _`.  Note
  1168 that the result needs to be unique, which means fact specifications
  1169 may have to be refined after enriching a proof context.
  1170 
  1171 * Attribute "case_names" has been refined: the assumptions in each case
  1172 can be named now by following the case name with [name1 name2 ...].
  1173 
  1174 * Isabelle/Isar reference manual has been updated and extended:
  1175   - "Synopsis" provides a catalog of main Isar language concepts.
  1176   - Formal references in syntax diagrams, via @{rail} antiquotation.
  1177   - Updated material from classic "ref" manual, notably about
  1178     "Classical Reasoner".
  1179 
  1180 
  1181 *** HOL ***
  1182 
  1183 * Class bot and top require underlying partial order rather than
  1184 preorder: uniqueness of bot and top is guaranteed.  INCOMPATIBILITY.
  1185 
  1186 * Class complete_lattice: generalized a couple of lemmas from sets;
  1187 generalized theorems INF_cong and SUP_cong.  New type classes for
  1188 complete boolean algebras and complete linear orders.  Lemmas
  1189 Inf_less_iff, less_Sup_iff, INF_less_iff, less_SUP_iff now reside in
  1190 class complete_linorder.
  1191 
  1192 Changed proposition of lemmas Inf_bool_def, Sup_bool_def, Inf_fun_def,
  1193 Sup_fun_def, Inf_apply, Sup_apply.
  1194 
  1195 Removed redundant lemmas (the right hand side gives hints how to
  1196 replace them for (metis ...), or (simp only: ...) proofs):
  1197 
  1198   Inf_singleton ~> Inf_insert [where A="{}", unfolded Inf_empty inf_top_right]
  1199   Sup_singleton ~> Sup_insert [where A="{}", unfolded Sup_empty sup_bot_right]
  1200   Inf_binary ~> Inf_insert, Inf_empty, and inf_top_right
  1201   Sup_binary ~> Sup_insert, Sup_empty, and sup_bot_right
  1202   Int_eq_Inter ~> Inf_insert, Inf_empty, and inf_top_right
  1203   Un_eq_Union ~> Sup_insert, Sup_empty, and sup_bot_right
  1204   Inter_def ~> INF_def, image_def
  1205   Union_def ~> SUP_def, image_def
  1206   INT_eq ~> INF_def, and image_def
  1207   UN_eq ~> SUP_def, and image_def
  1208   INF_subset ~> INF_superset_mono [OF _ order_refl]
  1209 
  1210 More consistent and comprehensive names:
  1211 
  1212   INTER_eq_Inter_image ~> INF_def
  1213   UNION_eq_Union_image ~> SUP_def
  1214   INFI_def ~> INF_def
  1215   SUPR_def ~> SUP_def
  1216   INF_leI ~> INF_lower
  1217   INF_leI2 ~> INF_lower2
  1218   le_INFI ~> INF_greatest
  1219   le_SUPI ~> SUP_upper
  1220   le_SUPI2 ~> SUP_upper2
  1221   SUP_leI ~> SUP_least
  1222   INFI_bool_eq ~> INF_bool_eq
  1223   SUPR_bool_eq ~> SUP_bool_eq
  1224   INFI_apply ~> INF_apply
  1225   SUPR_apply ~> SUP_apply
  1226   INTER_def ~> INTER_eq
  1227   UNION_def ~> UNION_eq
  1228 
  1229 INCOMPATIBILITY.
  1230 
  1231 * Renamed theory Complete_Lattice to Complete_Lattices.
  1232 INCOMPATIBILITY.
  1233 
  1234 * Theory Complete_Lattices: lemmas Inf_eq_top_iff, INF_eq_top_iff,
  1235 INF_image, Inf_insert, INF_top, Inf_top_conv, INF_top_conv, SUP_bot,
  1236 Sup_bot_conv, SUP_bot_conv, Sup_eq_top_iff, SUP_eq_top_iff, SUP_image,
  1237 Sup_insert are now declared as [simp].  INCOMPATIBILITY.
  1238 
  1239 * Theory Lattice: lemmas compl_inf_bot, compl_le_comp_iff,
  1240 compl_sup_top, inf_idem, inf_left_idem, inf_sup_absorb, sup_idem,
  1241 sup_inf_absob, sup_left_idem are now declared as [simp].  Minor
  1242 INCOMPATIBILITY.
  1243 
  1244 * Added syntactic classes "inf" and "sup" for the respective
  1245 constants.  INCOMPATIBILITY: Changes in the argument order of the
  1246 (mostly internal) locale predicates for some derived classes.
  1247 
  1248 * Theorem collections ball_simps and bex_simps do not contain theorems
  1249 referring to UNION any longer; these have been moved to collection
  1250 UN_ball_bex_simps.  INCOMPATIBILITY.
  1251 
  1252 * Theory Archimedean_Field: floor now is defined as parameter of a
  1253 separate type class floor_ceiling.
  1254 
  1255 * Theory Finite_Set: more coherent development of fold_set locales:
  1256 
  1257     locale fun_left_comm ~> locale comp_fun_commute
  1258     locale fun_left_comm_idem ~> locale comp_fun_idem
  1259 
  1260 Both use point-free characterization; interpretation proofs may need
  1261 adjustment.  INCOMPATIBILITY.
  1262 
  1263 * Theory Limits: Type "'a net" has been renamed to "'a filter", in
  1264 accordance with standard mathematical terminology. INCOMPATIBILITY.
  1265 
  1266 * Theory Complex_Main: The locale interpretations for the
  1267 bounded_linear and bounded_bilinear locales have been removed, in
  1268 order to reduce the number of duplicate lemmas. Users must use the
  1269 original names for distributivity theorems, potential INCOMPATIBILITY.
  1270 
  1271   divide.add ~> add_divide_distrib
  1272   divide.diff ~> diff_divide_distrib
  1273   divide.setsum ~> setsum_divide_distrib
  1274   mult.add_right ~> right_distrib
  1275   mult.diff_right ~> right_diff_distrib
  1276   mult_right.setsum ~> setsum_right_distrib
  1277   mult_left.diff ~> left_diff_distrib
  1278 
  1279 * Theory Complex_Main: Several redundant theorems have been removed or
  1280 replaced by more general versions. INCOMPATIBILITY.
  1281 
  1282   real_diff_def ~> minus_real_def
  1283   real_divide_def ~> divide_real_def
  1284   real_less_def ~> less_le
  1285   real_abs_def ~> abs_real_def
  1286   real_sgn_def ~> sgn_real_def
  1287   real_mult_commute ~> mult_commute
  1288   real_mult_assoc ~> mult_assoc
  1289   real_mult_1 ~> mult_1_left
  1290   real_add_mult_distrib ~> left_distrib
  1291   real_zero_not_eq_one ~> zero_neq_one
  1292   real_mult_inverse_left ~> left_inverse
  1293   INVERSE_ZERO ~> inverse_zero
  1294   real_le_refl ~> order_refl
  1295   real_le_antisym ~> order_antisym
  1296   real_le_trans ~> order_trans
  1297   real_le_linear ~> linear
  1298   real_le_eq_diff ~> le_iff_diff_le_0
  1299   real_add_left_mono ~> add_left_mono
  1300   real_mult_order ~> mult_pos_pos
  1301   real_mult_less_mono2 ~> mult_strict_left_mono
  1302   real_of_int_real_of_nat ~> real_of_int_of_nat_eq
  1303   real_0_le_divide_iff ~> zero_le_divide_iff
  1304   realpow_two_disj ~> power2_eq_iff
  1305   real_squared_diff_one_factored ~> square_diff_one_factored
  1306   realpow_two_diff ~> square_diff_square_factored
  1307   reals_complete2 ~> complete_real
  1308   real_sum_squared_expand ~> power2_sum
  1309   exp_ln_eq ~> ln_unique
  1310   expi_add ~> exp_add
  1311   expi_zero ~> exp_zero
  1312   lemma_DERIV_subst ~> DERIV_cong
  1313   LIMSEQ_Zfun_iff ~> tendsto_Zfun_iff
  1314   LIMSEQ_const ~> tendsto_const
  1315   LIMSEQ_norm ~> tendsto_norm
  1316   LIMSEQ_add ~> tendsto_add
  1317   LIMSEQ_minus ~> tendsto_minus
  1318   LIMSEQ_minus_cancel ~> tendsto_minus_cancel
  1319   LIMSEQ_diff ~> tendsto_diff
  1320   bounded_linear.LIMSEQ ~> bounded_linear.tendsto
  1321   bounded_bilinear.LIMSEQ ~> bounded_bilinear.tendsto
  1322   LIMSEQ_mult ~> tendsto_mult
  1323   LIMSEQ_inverse ~> tendsto_inverse
  1324   LIMSEQ_divide ~> tendsto_divide
  1325   LIMSEQ_pow ~> tendsto_power
  1326   LIMSEQ_setsum ~> tendsto_setsum
  1327   LIMSEQ_setprod ~> tendsto_setprod
  1328   LIMSEQ_norm_zero ~> tendsto_norm_zero_iff
  1329   LIMSEQ_rabs_zero ~> tendsto_rabs_zero_iff
  1330   LIMSEQ_imp_rabs ~> tendsto_rabs
  1331   LIMSEQ_add_minus ~> tendsto_add [OF _ tendsto_minus]
  1332   LIMSEQ_add_const ~> tendsto_add [OF _ tendsto_const]
  1333   LIMSEQ_diff_const ~> tendsto_diff [OF _ tendsto_const]
  1334   LIMSEQ_Complex ~> tendsto_Complex
  1335   LIM_ident ~> tendsto_ident_at
  1336   LIM_const ~> tendsto_const
  1337   LIM_add ~> tendsto_add
  1338   LIM_add_zero ~> tendsto_add_zero
  1339   LIM_minus ~> tendsto_minus
  1340   LIM_diff ~> tendsto_diff
  1341   LIM_norm ~> tendsto_norm
  1342   LIM_norm_zero ~> tendsto_norm_zero
  1343   LIM_norm_zero_cancel ~> tendsto_norm_zero_cancel
  1344   LIM_norm_zero_iff ~> tendsto_norm_zero_iff
  1345   LIM_rabs ~> tendsto_rabs
  1346   LIM_rabs_zero ~> tendsto_rabs_zero
  1347   LIM_rabs_zero_cancel ~> tendsto_rabs_zero_cancel
  1348   LIM_rabs_zero_iff ~> tendsto_rabs_zero_iff
  1349   LIM_compose ~> tendsto_compose
  1350   LIM_mult ~> tendsto_mult
  1351   LIM_scaleR ~> tendsto_scaleR
  1352   LIM_of_real ~> tendsto_of_real
  1353   LIM_power ~> tendsto_power
  1354   LIM_inverse ~> tendsto_inverse
  1355   LIM_sgn ~> tendsto_sgn
  1356   isCont_LIM_compose ~> isCont_tendsto_compose
  1357   bounded_linear.LIM ~> bounded_linear.tendsto
  1358   bounded_linear.LIM_zero ~> bounded_linear.tendsto_zero
  1359   bounded_bilinear.LIM ~> bounded_bilinear.tendsto
  1360   bounded_bilinear.LIM_prod_zero ~> bounded_bilinear.tendsto_zero
  1361   bounded_bilinear.LIM_left_zero ~> bounded_bilinear.tendsto_left_zero
  1362   bounded_bilinear.LIM_right_zero ~> bounded_bilinear.tendsto_right_zero
  1363   LIM_inverse_fun ~> tendsto_inverse [OF tendsto_ident_at]
  1364 
  1365 * Theory Complex_Main: The definition of infinite series was
  1366 generalized.  Now it is defined on the type class {topological_space,
  1367 comm_monoid_add}.  Hence it is useable also for extended real numbers.
  1368 
  1369 * Theory Complex_Main: The complex exponential function "expi" is now
  1370 a type-constrained abbreviation for "exp :: complex => complex"; thus
  1371 several polymorphic lemmas about "exp" are now applicable to "expi".
  1372 
  1373 * Code generation:
  1374 
  1375   - Theory Library/Code_Char_ord provides native ordering of
  1376     characters in the target language.
  1377 
  1378   - Commands code_module and code_library are legacy, use export_code
  1379     instead.
  1380 
  1381   - Method "evaluation" is legacy, use method "eval" instead.
  1382 
  1383   - Legacy evaluator "SML" is deactivated by default.  May be
  1384     reactivated by the following theory command:
  1385 
  1386       setup {* Value.add_evaluator ("SML", Codegen.eval_term) *}
  1387 
  1388 * Declare ext [intro] by default.  Rare INCOMPATIBILITY.
  1389 
  1390 * New proof method "induction" that gives induction hypotheses the
  1391 name "IH", thus distinguishing them from further hypotheses that come
  1392 from rule induction.  The latter are still called "hyps".  Method
  1393 "induction" is a thin wrapper around "induct" and follows the same
  1394 syntax.
  1395 
  1396 * Method "fastsimp" has been renamed to "fastforce", but "fastsimp" is
  1397 still available as a legacy feature for some time.
  1398 
  1399 * Nitpick:
  1400   - Added "need" and "total_consts" options.
  1401   - Reintroduced "show_skolems" option by popular demand.
  1402   - Renamed attribute: nitpick_def ~> nitpick_unfold.
  1403     INCOMPATIBILITY.
  1404 
  1405 * Sledgehammer:
  1406   - Use quasi-sound (and efficient) translations by default.
  1407   - Added support for the following provers: E-ToFoF, LEO-II,
  1408     Satallax, SNARK, Waldmeister, and Z3 with TPTP syntax.
  1409   - Automatically preplay and minimize proofs before showing them if
  1410     this can be done within reasonable time.
  1411   - sledgehammer available_provers ~> sledgehammer supported_provers.
  1412     INCOMPATIBILITY.
  1413   - Added "preplay_timeout", "slicing", "type_enc", "sound",
  1414     "max_mono_iters", and "max_new_mono_instances" options.
  1415   - Removed "explicit_apply" and "full_types" options as well as "Full
  1416     Types" Proof General menu item. INCOMPATIBILITY.
  1417 
  1418 * Metis:
  1419   - Removed "metisF" -- use "metis" instead. INCOMPATIBILITY.
  1420   - Obsoleted "metisFT" -- use "metis (full_types)" instead.
  1421     INCOMPATIBILITY.
  1422 
  1423 * Command 'try':
  1424   - Renamed 'try_methods' and added "simp:", "intro:", "dest:", and
  1425     "elim:" options. INCOMPATIBILITY.
  1426   - Introduced 'try' that not only runs 'try_methods' but also
  1427     'solve_direct', 'sledgehammer', 'quickcheck', and 'nitpick'.
  1428 
  1429 * Quickcheck:
  1430   - Added "eval" option to evaluate terms for the found counterexample
  1431     (currently only supported by the default (exhaustive) tester).
  1432   - Added post-processing of terms to obtain readable counterexamples
  1433     (currently only supported by the default (exhaustive) tester).
  1434   - New counterexample generator quickcheck[narrowing] enables
  1435     narrowing-based testing.  Requires the Glasgow Haskell compiler
  1436     with its installation location defined in the Isabelle settings
  1437     environment as ISABELLE_GHC.
  1438   - Removed quickcheck tester "SML" based on the SML code generator
  1439     (formly in HOL/Library).
  1440 
  1441 * Function package: discontinued option "tailrec".  INCOMPATIBILITY,
  1442 use 'partial_function' instead.
  1443 
  1444 * Theory Library/Extended_Reals replaces now the positive extended
  1445 reals found in probability theory. This file is extended by
  1446 Multivariate_Analysis/Extended_Real_Limits.
  1447 
  1448 * Theory Library/Old_Recdef: old 'recdef' package has been moved here,
  1449 from where it must be imported explicitly if it is really required.
  1450 INCOMPATIBILITY.
  1451 
  1452 * Theory Library/Wfrec: well-founded recursion combinator "wfrec" has
  1453 been moved here.  INCOMPATIBILITY.
  1454 
  1455 * Theory Library/Saturated provides type of numbers with saturated
  1456 arithmetic.
  1457 
  1458 * Theory Library/Product_Lattice defines a pointwise ordering for the
  1459 product type 'a * 'b, and provides instance proofs for various order
  1460 and lattice type classes.
  1461 
  1462 * Theory Library/Countable now provides the "countable_datatype" proof
  1463 method for proving "countable" class instances for datatypes.
  1464 
  1465 * Theory Library/Cset_Monad allows do notation for computable sets
  1466 (cset) via the generic monad ad-hoc overloading facility.
  1467 
  1468 * Library: Theories of common data structures are split into theories
  1469 for implementation, an invariant-ensuring type, and connection to an
  1470 abstract type. INCOMPATIBILITY.
  1471 
  1472   - RBT is split into RBT and RBT_Mapping.
  1473   - AssocList is split and renamed into AList and AList_Mapping.
  1474   - DList is split into DList_Impl, DList, and DList_Cset.
  1475   - Cset is split into Cset and List_Cset.
  1476 
  1477 * Theory Library/Nat_Infinity has been renamed to
  1478 Library/Extended_Nat, with name changes of the following types and
  1479 constants:
  1480 
  1481   type inat   ~> type enat
  1482   Fin         ~> enat
  1483   Infty       ~> infinity (overloaded)
  1484   iSuc        ~> eSuc
  1485   the_Fin     ~> the_enat
  1486 
  1487 Every theorem name containing "inat", "Fin", "Infty", or "iSuc" has
  1488 been renamed accordingly. INCOMPATIBILITY.
  1489 
  1490 * Session Multivariate_Analysis: The euclidean_space type class now
  1491 fixes a constant "Basis :: 'a set" consisting of the standard
  1492 orthonormal basis for the type. Users now have the option of
  1493 quantifying over this set instead of using the "basis" function, e.g.
  1494 "ALL x:Basis. P x" vs "ALL i<DIM('a). P (basis i)".
  1495 
  1496 * Session Multivariate_Analysis: Type "('a, 'b) cart" has been renamed
  1497 to "('a, 'b) vec" (the syntax "'a ^ 'b" remains unaffected). Constants
  1498 "Cart_nth" and "Cart_lambda" have been respectively renamed to
  1499 "vec_nth" and "vec_lambda"; theorems mentioning those names have
  1500 changed to match. Definition theorems for overloaded constants now use
  1501 the standard "foo_vec_def" naming scheme. A few other theorems have
  1502 been renamed as follows (INCOMPATIBILITY):
  1503 
  1504   Cart_eq          ~> vec_eq_iff
  1505   dist_nth_le_cart ~> dist_vec_nth_le
  1506   tendsto_vector   ~> vec_tendstoI
  1507   Cauchy_vector    ~> vec_CauchyI
  1508 
  1509 * Session Multivariate_Analysis: Several duplicate theorems have been
  1510 removed, and other theorems have been renamed or replaced with more
  1511 general versions. INCOMPATIBILITY.
  1512 
  1513   finite_choice ~> finite_set_choice
  1514   eventually_conjI ~> eventually_conj
  1515   eventually_and ~> eventually_conj_iff
  1516   eventually_false ~> eventually_False
  1517   setsum_norm ~> norm_setsum
  1518   Lim_sequentially ~> LIMSEQ_def
  1519   Lim_ident_at ~> LIM_ident
  1520   Lim_const ~> tendsto_const
  1521   Lim_cmul ~> tendsto_scaleR [OF tendsto_const]
  1522   Lim_neg ~> tendsto_minus
  1523   Lim_add ~> tendsto_add
  1524   Lim_sub ~> tendsto_diff
  1525   Lim_mul ~> tendsto_scaleR
  1526   Lim_vmul ~> tendsto_scaleR [OF _ tendsto_const]
  1527   Lim_null_norm ~> tendsto_norm_zero_iff [symmetric]
  1528   Lim_linear ~> bounded_linear.tendsto
  1529   Lim_component ~> tendsto_euclidean_component
  1530   Lim_component_cart ~> tendsto_vec_nth
  1531   Lim_inner ~> tendsto_inner [OF tendsto_const]
  1532   dot_lsum ~> inner_setsum_left
  1533   dot_rsum ~> inner_setsum_right
  1534   continuous_cmul ~> continuous_scaleR [OF continuous_const]
  1535   continuous_neg ~> continuous_minus
  1536   continuous_sub ~> continuous_diff
  1537   continuous_vmul ~> continuous_scaleR [OF _ continuous_const]
  1538   continuous_mul ~> continuous_scaleR
  1539   continuous_inv ~> continuous_inverse
  1540   continuous_at_within_inv ~> continuous_at_within_inverse
  1541   continuous_at_inv ~> continuous_at_inverse
  1542   continuous_at_norm ~> continuous_norm [OF continuous_at_id]
  1543   continuous_at_infnorm ~> continuous_infnorm [OF continuous_at_id]
  1544   continuous_at_component ~> continuous_component [OF continuous_at_id]
  1545   continuous_on_neg ~> continuous_on_minus
  1546   continuous_on_sub ~> continuous_on_diff
  1547   continuous_on_cmul ~> continuous_on_scaleR [OF continuous_on_const]
  1548   continuous_on_vmul ~> continuous_on_scaleR [OF _ continuous_on_const]
  1549   continuous_on_mul ~> continuous_on_scaleR
  1550   continuous_on_mul_real ~> continuous_on_mult
  1551   continuous_on_inner ~> continuous_on_inner [OF continuous_on_const]
  1552   continuous_on_norm ~> continuous_on_norm [OF continuous_on_id]
  1553   continuous_on_inverse ~> continuous_on_inv
  1554   uniformly_continuous_on_neg ~> uniformly_continuous_on_minus
  1555   uniformly_continuous_on_sub ~> uniformly_continuous_on_diff
  1556   subset_interior ~> interior_mono
  1557   subset_closure ~> closure_mono
  1558   closure_univ ~> closure_UNIV
  1559   real_arch_lt ~> reals_Archimedean2
  1560   real_arch ~> reals_Archimedean3
  1561   real_abs_norm ~> abs_norm_cancel
  1562   real_abs_sub_norm ~> norm_triangle_ineq3
  1563   norm_cauchy_schwarz_abs ~> Cauchy_Schwarz_ineq2
  1564 
  1565 * Session HOL-Probability:
  1566   - Caratheodory's extension lemma is now proved for ring_of_sets.
  1567   - Infinite products of probability measures are now available.
  1568   - Sigma closure is independent, if the generator is independent
  1569   - Use extended reals instead of positive extended
  1570     reals. INCOMPATIBILITY.
  1571 
  1572 * Session HOLCF: Discontinued legacy theorem names, INCOMPATIBILITY.
  1573 
  1574   expand_fun_below ~> fun_below_iff
  1575   below_fun_ext ~> fun_belowI
  1576   expand_cfun_eq ~> cfun_eq_iff
  1577   ext_cfun ~> cfun_eqI
  1578   expand_cfun_below ~> cfun_below_iff
  1579   below_cfun_ext ~> cfun_belowI
  1580   monofun_fun_fun ~> fun_belowD
  1581   monofun_fun_arg ~> monofunE
  1582   monofun_lub_fun ~> adm_monofun [THEN admD]
  1583   cont_lub_fun ~> adm_cont [THEN admD]
  1584   cont2cont_Rep_CFun ~> cont2cont_APP
  1585   cont_Rep_CFun_app ~> cont_APP_app
  1586   cont_Rep_CFun_app_app ~> cont_APP_app_app
  1587   cont_cfun_fun ~> cont_Rep_cfun1 [THEN contE]
  1588   cont_cfun_arg ~> cont_Rep_cfun2 [THEN contE]
  1589   contlub_cfun ~> lub_APP [symmetric]
  1590   contlub_LAM ~> lub_LAM [symmetric]
  1591   thelubI ~> lub_eqI
  1592   UU_I ~> bottomI
  1593   lift_distinct1 ~> lift.distinct(1)
  1594   lift_distinct2 ~> lift.distinct(2)
  1595   Def_not_UU ~> lift.distinct(2)
  1596   Def_inject ~> lift.inject
  1597   below_UU_iff ~> below_bottom_iff
  1598   eq_UU_iff ~> eq_bottom_iff
  1599 
  1600 
  1601 *** Document preparation ***
  1602 
  1603 * Antiquotation @{rail} layouts railroad syntax diagrams, see also
  1604 isar-ref manual, both for description and actual application of the
  1605 same.
  1606 
  1607 * Antiquotation @{value} evaluates the given term and presents its
  1608 result.
  1609 
  1610 * Antiquotations: term style "isub" provides ad-hoc conversion of
  1611 variables x1, y23 into subscripted form x\<^isub>1,
  1612 y\<^isub>2\<^isub>3.
  1613 
  1614 * Predefined LaTeX macros for Isabelle symbols \<bind> and \<then>
  1615 (e.g. see ~~/src/HOL/Library/Monad_Syntax.thy).
  1616 
  1617 * Localized \isabellestyle switch can be used within blocks or groups
  1618 like this:
  1619 
  1620   \isabellestyle{it}  %preferred default
  1621   {\isabellestylett @{text "typewriter stuff"}}
  1622 
  1623 * Discontinued special treatment of hard tabulators.  Implicit
  1624 tab-width is now defined as 1.  Potential INCOMPATIBILITY for visual
  1625 layouts.
  1626 
  1627 
  1628 *** ML ***
  1629 
  1630 * The inner syntax of sort/type/term/prop supports inlined YXML
  1631 representations within quoted string tokens.  By encoding logical
  1632 entities via Term_XML (in ML or Scala) concrete syntax can be
  1633 bypassed, which is particularly useful for producing bits of text
  1634 under external program control.
  1635 
  1636 * Antiquotations for ML and document preparation are managed as theory
  1637 data, which requires explicit setup.
  1638 
  1639 * Isabelle_Process.is_active allows tools to check if the official
  1640 process wrapper is running (Isabelle/Scala/jEdit) or the old TTY loop
  1641 (better known as Proof General).
  1642 
  1643 * Structure Proof_Context follows standard naming scheme.  Old
  1644 ProofContext is still available for some time as legacy alias.
  1645 
  1646 * Structure Timing provides various operations for timing; supersedes
  1647 former start_timing/end_timing etc.
  1648 
  1649 * Path.print is the official way to show file-system paths to users
  1650 (including quotes etc.).
  1651 
  1652 * Inner syntax: identifiers in parse trees of generic categories
  1653 "logic", "aprop", "idt" etc. carry position information (disguised as
  1654 type constraints).  Occasional INCOMPATIBILITY with non-compliant
  1655 translations that choke on unexpected type constraints.  Positions can
  1656 be stripped in ML translations via Syntax.strip_positions /
  1657 Syntax.strip_positions_ast, or via the syntax constant
  1658 "_strip_positions" within parse trees.  As last resort, positions can
  1659 be disabled via the configuration option Syntax.positions, which is
  1660 called "syntax_positions" in Isar attribute syntax.
  1661 
  1662 * Discontinued special status of various ML structures that contribute
  1663 to structure Syntax (Ast, Lexicon, Mixfix, Parser, Printer etc.): less
  1664 pervasive content, no inclusion in structure Syntax.  INCOMPATIBILITY,
  1665 refer directly to Ast.Constant, Lexicon.is_identifier,
  1666 Syntax_Trans.mk_binder_tr etc.
  1667 
  1668 * Typed print translation: discontinued show_sorts argument, which is
  1669 already available via context of "advanced" translation.
  1670 
  1671 * Refined PARALLEL_GOALS tactical: degrades gracefully for schematic
  1672 goal states; body tactic needs to address all subgoals uniformly.
  1673 
  1674 * Slightly more special eq_list/eq_set, with shortcut involving
  1675 pointer equality (assumes that eq relation is reflexive).
  1676 
  1677 * Classical tactics use proper Proof.context instead of historic types
  1678 claset/clasimpset.  Old-style declarations like addIs, addEs, addDs
  1679 operate directly on Proof.context.  Raw type claset retains its use as
  1680 snapshot of the classical context, which can be recovered via
  1681 (put_claset HOL_cs) etc.  Type clasimpset has been discontinued.
  1682 INCOMPATIBILITY, classical tactics and derived proof methods require
  1683 proper Proof.context.
  1684 
  1685 
  1686 *** System ***
  1687 
  1688 * Discontinued support for Poly/ML 5.2, which was the last version
  1689 without proper multithreading and TimeLimit implementation.
  1690 
  1691 * Discontinued old lib/scripts/polyml-platform, which has been
  1692 obsolete since Isabelle2009-2.
  1693 
  1694 * Various optional external tools are referenced more robustly and
  1695 uniformly by explicit Isabelle settings as follows:
  1696 
  1697   ISABELLE_CSDP   (formerly CSDP_EXE)
  1698   ISABELLE_GHC    (formerly EXEC_GHC or GHC_PATH)
  1699   ISABELLE_OCAML  (formerly EXEC_OCAML)
  1700   ISABELLE_SWIPL  (formerly EXEC_SWIPL)
  1701   ISABELLE_YAP    (formerly EXEC_YAP)
  1702 
  1703 Note that automated detection from the file-system or search path has
  1704 been discontinued.  INCOMPATIBILITY.
  1705 
  1706 * Scala layer provides JVM method invocation service for static
  1707 methods of type (String)String, see Invoke_Scala.method in ML.  For
  1708 example:
  1709 
  1710   Invoke_Scala.method "java.lang.System.getProperty" "java.home"
  1711 
  1712 Together with YXML.string_of_body/parse_body and XML.Encode/Decode
  1713 this allows to pass structured values between ML and Scala.
  1714 
  1715 * The IsabelleText fonts includes some further glyphs to support the
  1716 Prover IDE.  Potential INCOMPATIBILITY: users who happen to have
  1717 installed a local copy (which is normally *not* required) need to
  1718 delete or update it from ~~/lib/fonts/.
  1719 
  1720 
  1721 
  1722 New in Isabelle2011 (January 2011)
  1723 ----------------------------------
  1724 
  1725 *** General ***
  1726 
  1727 * Experimental Prover IDE based on Isabelle/Scala and jEdit (see
  1728 src/Tools/jEdit).  This also serves as IDE for Isabelle/ML, with
  1729 useful tooltips and hyperlinks produced from its static analysis.  The
  1730 bundled component provides an executable Isabelle tool that can be run
  1731 like this:
  1732 
  1733   Isabelle2011/bin/isabelle jedit
  1734 
  1735 * Significantly improved Isabelle/Isar implementation manual.
  1736 
  1737 * System settings: ISABELLE_HOME_USER now includes ISABELLE_IDENTIFIER
  1738 (and thus refers to something like $HOME/.isabelle/Isabelle2011),
  1739 while the default heap location within that directory lacks that extra
  1740 suffix.  This isolates multiple Isabelle installations from each
  1741 other, avoiding problems with old settings in new versions.
  1742 INCOMPATIBILITY, need to copy/upgrade old user settings manually.
  1743 
  1744 * Source files are always encoded as UTF-8, instead of old-fashioned
  1745 ISO-Latin-1.  INCOMPATIBILITY.  Isabelle LaTeX documents might require
  1746 the following package declarations:
  1747 
  1748   \usepackage[utf8]{inputenc}
  1749   \usepackage{textcomp}
  1750 
  1751 * Explicit treatment of UTF-8 sequences as Isabelle symbols, such that
  1752 a Unicode character is treated as a single symbol, not a sequence of
  1753 non-ASCII bytes as before.  Since Isabelle/ML string literals may
  1754 contain symbols without further backslash escapes, Unicode can now be
  1755 used here as well.  Recall that Symbol.explode in ML provides a
  1756 consistent view on symbols, while raw explode (or String.explode)
  1757 merely give a byte-oriented representation.
  1758 
  1759 * Theory loader: source files are primarily located via the master
  1760 directory of each theory node (where the .thy file itself resides).
  1761 The global load path is still partially available as legacy feature.
  1762 Minor INCOMPATIBILITY due to subtle change in file lookup: use
  1763 explicit paths, relatively to the theory.
  1764 
  1765 * Special treatment of ML file names has been discontinued.
  1766 Historically, optional extensions .ML or .sml were added on demand --
  1767 at the cost of clarity of file dependencies.  Recall that Isabelle/ML
  1768 files exclusively use the .ML extension.  Minor INCOMPATIBILTY.
  1769 
  1770 * Various options that affect pretty printing etc. are now properly
  1771 handled within the context via configuration options, instead of
  1772 unsynchronized references or print modes.  There are both ML Config.T
  1773 entities and Isar declaration attributes to access these.
  1774 
  1775   ML (Config.T)                 Isar (attribute)
  1776 
  1777   eta_contract                  eta_contract
  1778   show_brackets                 show_brackets
  1779   show_sorts                    show_sorts
  1780   show_types                    show_types
  1781   show_question_marks           show_question_marks
  1782   show_consts                   show_consts
  1783   show_abbrevs                  show_abbrevs
  1784 
  1785   Syntax.ast_trace              syntax_ast_trace
  1786   Syntax.ast_stat               syntax_ast_stat
  1787   Syntax.ambiguity_level        syntax_ambiguity_level
  1788 
  1789   Goal_Display.goals_limit      goals_limit
  1790   Goal_Display.show_main_goal   show_main_goal
  1791 
  1792   Method.rule_trace             rule_trace
  1793 
  1794   Thy_Output.display            thy_output_display
  1795   Thy_Output.quotes             thy_output_quotes
  1796   Thy_Output.indent             thy_output_indent
  1797   Thy_Output.source             thy_output_source
  1798   Thy_Output.break              thy_output_break
  1799 
  1800 Note that corresponding "..._default" references in ML may only be
  1801 changed globally at the ROOT session setup, but *not* within a theory.
  1802 The option "show_abbrevs" supersedes the former print mode
  1803 "no_abbrevs" with inverted meaning.
  1804 
  1805 * More systematic naming of some configuration options.
  1806 INCOMPATIBILITY.
  1807 
  1808   trace_simp  ~>  simp_trace
  1809   debug_simp  ~>  simp_debug
  1810 
  1811 * Support for real valued configuration options, using simplistic
  1812 floating-point notation that coincides with the inner syntax for
  1813 float_token.
  1814 
  1815 * Support for real valued preferences (with approximative PGIP type):
  1816 front-ends need to accept "pgint" values in float notation.
  1817 INCOMPATIBILITY.
  1818 
  1819 * The IsabelleText font now includes Cyrillic, Hebrew, Arabic from
  1820 DejaVu Sans.
  1821 
  1822 * Discontinued support for Poly/ML 5.0 and 5.1 versions.
  1823 
  1824 
  1825 *** Pure ***
  1826 
  1827 * Command 'type_synonym' (with single argument) replaces somewhat
  1828 outdated 'types', which is still available as legacy feature for some
  1829 time.
  1830 
  1831 * Command 'nonterminal' (with 'and' separated list of arguments)
  1832 replaces somewhat outdated 'nonterminals'.  INCOMPATIBILITY.
  1833 
  1834 * Command 'notepad' replaces former 'example_proof' for
  1835 experimentation in Isar without any result.  INCOMPATIBILITY.
  1836 
  1837 * Locale interpretation commands 'interpret' and 'sublocale' accept
  1838 lists of equations to map definitions in a locale to appropriate
  1839 entities in the context of the interpretation.  The 'interpretation'
  1840 command already provided this functionality.
  1841 
  1842 * Diagnostic command 'print_dependencies' prints the locale instances
  1843 that would be activated if the specified expression was interpreted in
  1844 the current context.  Variant "print_dependencies!" assumes a context
  1845 without interpretations.
  1846 
  1847 * Diagnostic command 'print_interps' prints interpretations in proofs
  1848 in addition to interpretations in theories.
  1849 
  1850 * Discontinued obsolete 'global' and 'local' commands to manipulate
  1851 the theory name space.  Rare INCOMPATIBILITY.  The ML functions
  1852 Sign.root_path and Sign.local_path may be applied directly where this
  1853 feature is still required for historical reasons.
  1854 
  1855 * Discontinued obsolete 'constdefs' command.  INCOMPATIBILITY, use
  1856 'definition' instead.
  1857 
  1858 * The "prems" fact, which refers to the accidental collection of
  1859 foundational premises in the context, is now explicitly marked as
  1860 legacy feature and will be discontinued soon.  Consider using "assms"
  1861 of the head statement or reference facts by explicit names.
  1862 
  1863 * Document antiquotations @{class} and @{type} print classes and type
  1864 constructors.
  1865 
  1866 * Document antiquotation @{file} checks file/directory entries within
  1867 the local file system.
  1868 
  1869 
  1870 *** HOL ***
  1871 
  1872 * Coercive subtyping: functions can be declared as coercions and type
  1873 inference will add them as necessary upon input of a term.  Theory
  1874 Complex_Main declares real :: nat => real and real :: int => real as
  1875 coercions. A coercion function f is declared like this:
  1876 
  1877   declare [[coercion f]]
  1878 
  1879 To lift coercions through type constructors (e.g. from nat => real to
  1880 nat list => real list), map functions can be declared, e.g.
  1881 
  1882   declare [[coercion_map map]]
  1883 
  1884 Currently coercion inference is activated only in theories including
  1885 real numbers, i.e. descendants of Complex_Main.  This is controlled by
  1886 the configuration option "coercion_enabled", e.g. it can be enabled in
  1887 other theories like this:
  1888 
  1889   declare [[coercion_enabled]]
  1890 
  1891 * Command 'partial_function' provides basic support for recursive
  1892 function definitions over complete partial orders.  Concrete instances
  1893 are provided for i) the option type, ii) tail recursion on arbitrary
  1894 types, and iii) the heap monad of Imperative_HOL.  See
  1895 src/HOL/ex/Fundefs.thy and src/HOL/Imperative_HOL/ex/Linked_Lists.thy
  1896 for examples.
  1897 
  1898 * Function package: f.psimps rules are no longer implicitly declared
  1899 as [simp].  INCOMPATIBILITY.
  1900 
  1901 * Datatype package: theorems generated for executable equality (class
  1902 "eq") carry proper names and are treated as default code equations.
  1903 
  1904 * Inductive package: now offers command 'inductive_simps' to
  1905 automatically derive instantiated and simplified equations for
  1906 inductive predicates, similar to 'inductive_cases'.
  1907 
  1908 * Command 'enriched_type' allows to register properties of the
  1909 functorial structure of types.
  1910 
  1911 * Improved infrastructure for term evaluation using code generator
  1912 techniques, in particular static evaluation conversions.
  1913 
  1914 * Code generator: Scala (2.8 or higher) has been added to the target
  1915 languages.
  1916 
  1917 * Code generator: globbing constant expressions "*" and "Theory.*"
  1918 have been replaced by the more idiomatic "_" and "Theory._".
  1919 INCOMPATIBILITY.
  1920 
  1921 * Code generator: export_code without explicit file declaration prints
  1922 to standard output.  INCOMPATIBILITY.
  1923 
  1924 * Code generator: do not print function definitions for case
  1925 combinators any longer.
  1926 
  1927 * Code generator: simplification with rules determined with
  1928 src/Tools/Code/code_simp.ML and method "code_simp".
  1929 
  1930 * Code generator for records: more idiomatic representation of record
  1931 types.  Warning: records are not covered by ancient SML code
  1932 generation any longer.  INCOMPATIBILITY.  In cases of need, a suitable
  1933 rep_datatype declaration helps to succeed then:
  1934 
  1935   record 'a foo = ...
  1936   ...
  1937   rep_datatype foo_ext ...
  1938 
  1939 * Records: logical foundation type for records does not carry a
  1940 '_type' suffix any longer (obsolete due to authentic syntax).
  1941 INCOMPATIBILITY.
  1942 
  1943 * Quickcheck now by default uses exhaustive testing instead of random
  1944 testing.  Random testing can be invoked by "quickcheck [random]",
  1945 exhaustive testing by "quickcheck [exhaustive]".
  1946 
  1947 * Quickcheck instantiates polymorphic types with small finite
  1948 datatypes by default. This enables a simple execution mechanism to
  1949 handle quantifiers and function equality over the finite datatypes.
  1950 
  1951 * Quickcheck random generator has been renamed from "code" to
  1952 "random".  INCOMPATIBILITY.
  1953 
  1954 * Quickcheck now has a configurable time limit which is set to 30
  1955 seconds by default. This can be changed by adding [timeout = n] to the
  1956 quickcheck command. The time limit for Auto Quickcheck is still set
  1957 independently.
  1958 
  1959 * Quickcheck in locales considers interpretations of that locale for
  1960 counter example search.
  1961 
  1962 * Sledgehammer:
  1963   - Added "smt" and "remote_smt" provers based on the "smt" proof
  1964     method. See the Sledgehammer manual for details ("isabelle doc
  1965     sledgehammer").
  1966   - Renamed commands:
  1967     sledgehammer atp_info ~> sledgehammer running_provers
  1968     sledgehammer atp_kill ~> sledgehammer kill_provers
  1969     sledgehammer available_atps ~> sledgehammer available_provers
  1970     INCOMPATIBILITY.
  1971   - Renamed options:
  1972     sledgehammer [atps = ...] ~> sledgehammer [provers = ...]
  1973     sledgehammer [atp = ...] ~> sledgehammer [prover = ...]
  1974     sledgehammer [timeout = 77 s] ~> sledgehammer [timeout = 77]
  1975     (and "ms" and "min" are no longer supported)
  1976     INCOMPATIBILITY.
  1977 
  1978 * Nitpick:
  1979   - Renamed options:
  1980     nitpick [timeout = 77 s] ~> nitpick [timeout = 77]
  1981     nitpick [tac_timeout = 777 ms] ~> nitpick [tac_timeout = 0.777]
  1982     INCOMPATIBILITY.
  1983   - Added support for partial quotient types.
  1984   - Added local versions of the "Nitpick.register_xxx" functions.
  1985   - Added "whack" option.
  1986   - Allow registration of quotient types as codatatypes.
  1987   - Improved "merge_type_vars" option to merge more types.
  1988   - Removed unsound "fast_descrs" option.
  1989   - Added custom symmetry breaking for datatypes, making it possible to reach
  1990     higher cardinalities.
  1991   - Prevent the expansion of too large definitions.
  1992 
  1993 * Proof methods "metis" and "meson" now have configuration options
  1994 "meson_trace", "metis_trace", and "metis_verbose" that can be enabled
  1995 to diagnose these tools. E.g.
  1996 
  1997     using [[metis_trace = true]]
  1998 
  1999 * Auto Solve: Renamed "Auto Solve Direct".  The tool is now available
  2000 manually as command 'solve_direct'.
  2001 
  2002 * The default SMT solver Z3 must be enabled explicitly (due to
  2003 licensing issues) by setting the environment variable
  2004 Z3_NON_COMMERCIAL in etc/settings of the component, for example.  For
  2005 commercial applications, the SMT solver CVC3 is provided as fall-back;
  2006 changing the SMT solver is done via the configuration option
  2007 "smt_solver".
  2008 
  2009 * Remote SMT solvers need to be referred to by the "remote_" prefix,
  2010 i.e. "remote_cvc3" and "remote_z3".
  2011 
  2012 * Added basic SMT support for datatypes, records, and typedefs using
  2013 the oracle mode (no proofs).  Direct support of pairs has been dropped
  2014 in exchange (pass theorems fst_conv snd_conv pair_collapse to the SMT
  2015 support for a similar behavior).  Minor INCOMPATIBILITY.
  2016 
  2017 * Changed SMT configuration options:
  2018   - Renamed:
  2019     z3_proofs ~> smt_oracle (with inverted meaning)
  2020     z3_trace_assms ~> smt_trace_used_facts
  2021     INCOMPATIBILITY.
  2022   - Added:
  2023     smt_verbose
  2024     smt_random_seed
  2025     smt_datatypes
  2026     smt_infer_triggers
  2027     smt_monomorph_limit
  2028     cvc3_options
  2029     remote_cvc3_options
  2030     remote_z3_options
  2031     yices_options
  2032 
  2033 * Boogie output files (.b2i files) need to be declared in the theory
  2034 header.
  2035 
  2036 * Simplification procedure "list_to_set_comprehension" rewrites list
  2037 comprehensions applied to List.set to set comprehensions.  Occasional
  2038 INCOMPATIBILITY, may be deactivated like this:
  2039 
  2040   declare [[simproc del: list_to_set_comprehension]]
  2041 
  2042 * Removed old version of primrec package.  INCOMPATIBILITY.
  2043 
  2044 * Removed simplifier congruence rule of "prod_case", as has for long
  2045 been the case with "split".  INCOMPATIBILITY.
  2046 
  2047 * String.literal is a type, but not a datatype.  INCOMPATIBILITY.
  2048 
  2049 * Removed [split_format ... and ... and ...] version of
  2050 [split_format].  Potential INCOMPATIBILITY.
  2051 
  2052 * Predicate "sorted" now defined inductively, with nice induction
  2053 rules.  INCOMPATIBILITY: former sorted.simps now named sorted_simps.
  2054 
  2055 * Constant "contents" renamed to "the_elem", to free the generic name
  2056 contents for other uses.  INCOMPATIBILITY.
  2057 
  2058 * Renamed class eq and constant eq (for code generation) to class
  2059 equal and constant equal, plus renaming of related facts and various
  2060 tuning.  INCOMPATIBILITY.
  2061 
  2062 * Dropped type classes mult_mono and mult_mono1.  INCOMPATIBILITY.
  2063 
  2064 * Removed output syntax "'a ~=> 'b" for "'a => 'b option".
  2065 INCOMPATIBILITY.
  2066 
  2067 * Renamed theory Fset to Cset, type Fset.fset to Cset.set, in order to
  2068 avoid confusion with finite sets.  INCOMPATIBILITY.
  2069 
  2070 * Abandoned locales equiv, congruent and congruent2 for equivalence
  2071 relations.  INCOMPATIBILITY: use equivI rather than equiv_intro (same
  2072 for congruent(2)).
  2073 
  2074 * Some previously unqualified names have been qualified:
  2075 
  2076   types
  2077     bool ~> HOL.bool
  2078     nat ~> Nat.nat
  2079 
  2080   constants
  2081     Trueprop ~> HOL.Trueprop
  2082     True ~> HOL.True
  2083     False ~> HOL.False
  2084     op & ~> HOL.conj
  2085     op | ~> HOL.disj
  2086     op --> ~> HOL.implies
  2087     op = ~> HOL.eq
  2088     Not ~> HOL.Not
  2089     The ~> HOL.The
  2090     All ~> HOL.All
  2091     Ex ~> HOL.Ex
  2092     Ex1 ~> HOL.Ex1
  2093     Let ~> HOL.Let
  2094     If ~> HOL.If
  2095     Ball ~> Set.Ball
  2096     Bex ~> Set.Bex
  2097     Suc ~> Nat.Suc
  2098     Pair ~> Product_Type.Pair
  2099     fst ~> Product_Type.fst
  2100     snd ~> Product_Type.snd
  2101     curry ~> Product_Type.curry
  2102     op : ~> Set.member
  2103     Collect ~> Set.Collect
  2104 
  2105 INCOMPATIBILITY.
  2106 
  2107 * More canonical naming convention for some fundamental definitions:
  2108 
  2109     bot_bool_eq ~> bot_bool_def
  2110     top_bool_eq ~> top_bool_def
  2111     inf_bool_eq ~> inf_bool_def
  2112     sup_bool_eq ~> sup_bool_def
  2113     bot_fun_eq  ~> bot_fun_def
  2114     top_fun_eq  ~> top_fun_def
  2115     inf_fun_eq  ~> inf_fun_def
  2116     sup_fun_eq  ~> sup_fun_def
  2117 
  2118 INCOMPATIBILITY.
  2119 
  2120 * More stylized fact names:
  2121 
  2122   expand_fun_eq ~> fun_eq_iff
  2123   expand_set_eq ~> set_eq_iff
  2124   set_ext       ~> set_eqI
  2125   nat_number    ~> eval_nat_numeral
  2126 
  2127 INCOMPATIBILITY.
  2128 
  2129 * Refactoring of code-generation specific operations in theory List:
  2130 
  2131   constants
  2132     null ~> List.null
  2133 
  2134   facts
  2135     mem_iff ~> member_def
  2136     null_empty ~> null_def
  2137 
  2138 INCOMPATIBILITY.  Note that these were not supposed to be used
  2139 regularly unless for striking reasons; their main purpose was code
  2140 generation.
  2141 
  2142 Various operations from the Haskell prelude are used for generating
  2143 Haskell code.
  2144 
  2145 * Term "bij f" is now an abbreviation of "bij_betw f UNIV UNIV".  Term
  2146 "surj f" is now an abbreviation of "range f = UNIV".  The theorems
  2147 bij_def and surj_def are unchanged.  INCOMPATIBILITY.
  2148 
  2149 * Abolished some non-alphabetic type names: "prod" and "sum" replace
  2150 "*" and "+" respectively.  INCOMPATIBILITY.
  2151 
  2152 * Name "Plus" of disjoint sum operator "<+>" is now hidden.  Write
  2153 "Sum_Type.Plus" instead.
  2154 
  2155 * Constant "split" has been merged with constant "prod_case"; names of
  2156 ML functions, facts etc. involving split have been retained so far,
  2157 though.  INCOMPATIBILITY.
  2158 
  2159 * Dropped old infix syntax "_ mem _" for List.member; use "_ : set _"
  2160 instead.  INCOMPATIBILITY.
  2161 
  2162 * Removed lemma "Option.is_none_none" which duplicates "is_none_def".
  2163 INCOMPATIBILITY.
  2164 
  2165 * Former theory Library/Enum is now part of the HOL-Main image.
  2166 INCOMPATIBILITY: all constants of the Enum theory now have to be
  2167 referred to by its qualified name.
  2168 
  2169   enum    ~>  Enum.enum
  2170   nlists  ~>  Enum.nlists
  2171   product ~>  Enum.product
  2172 
  2173 * Theory Library/Monad_Syntax provides do-syntax for monad types.
  2174 Syntax in Library/State_Monad has been changed to avoid ambiguities.
  2175 INCOMPATIBILITY.
  2176 
  2177 * Theory Library/SetsAndFunctions has been split into
  2178 Library/Function_Algebras and Library/Set_Algebras; canonical names
  2179 for instance definitions for functions; various improvements.
  2180 INCOMPATIBILITY.
  2181 
  2182 * Theory Library/Multiset provides stable quicksort implementation of
  2183 sort_key.
  2184 
  2185 * Theory Library/Multiset: renamed empty_idemp ~> empty_neutral.
  2186 INCOMPATIBILITY.
  2187 
  2188 * Session Multivariate_Analysis: introduced a type class for euclidean
  2189 space.  Most theorems are now stated in terms of euclidean spaces
  2190 instead of finite cartesian products.
  2191 
  2192   types
  2193     real ^ 'n ~>  'a::real_vector
  2194               ~>  'a::euclidean_space
  2195               ~>  'a::ordered_euclidean_space
  2196         (depends on your needs)
  2197 
  2198   constants
  2199      _ $ _        ~> _ $$ _
  2200      \<chi> x. _  ~> \<chi>\<chi> x. _
  2201      CARD('n)     ~> DIM('a)
  2202 
  2203 Also note that the indices are now natural numbers and not from some
  2204 finite type. Finite cartesian products of euclidean spaces, products
  2205 of euclidean spaces the real and complex numbers are instantiated to
  2206 be euclidean_spaces.  INCOMPATIBILITY.
  2207 
  2208 * Session Probability: introduced pextreal as positive extended real
  2209 numbers.  Use pextreal as value for measures.  Introduce the
  2210 Radon-Nikodym derivative, product spaces and Fubini's theorem for
  2211 arbitrary sigma finite measures.  Introduces Lebesgue measure based on
  2212 the integral in Multivariate Analysis.  INCOMPATIBILITY.
  2213 
  2214 * Session Imperative_HOL: revamped, corrected dozens of inadequacies.
  2215 INCOMPATIBILITY.
  2216 
  2217 * Session SPARK (with image HOL-SPARK) provides commands to load and
  2218 prove verification conditions generated by the SPARK Ada program
  2219 verifier.  See also src/HOL/SPARK and src/HOL/SPARK/Examples.
  2220 
  2221 
  2222 *** HOL-Algebra ***
  2223 
  2224 * Theorems for additive ring operations (locale abelian_monoid and
  2225 descendants) are generated by interpretation from their multiplicative
  2226 counterparts.  Names (in particular theorem names) have the mandatory
  2227 qualifier 'add'.  Previous theorem names are redeclared for
  2228 compatibility.
  2229 
  2230 * Structure "int_ring" is now an abbreviation (previously a
  2231 definition).  This fits more natural with advanced interpretations.
  2232 
  2233 
  2234 *** HOLCF ***
  2235 
  2236 * The domain package now runs in definitional mode by default: The
  2237 former command 'new_domain' is now called 'domain'.  To use the domain
  2238 package in its original axiomatic mode, use 'domain (unsafe)'.
  2239 INCOMPATIBILITY.
  2240 
  2241 * The new class "domain" is now the default sort.  Class "predomain"
  2242 is an unpointed version of "domain". Theories can be updated by
  2243 replacing sort annotations as shown below.  INCOMPATIBILITY.
  2244 
  2245   'a::type ~> 'a::countable
  2246   'a::cpo  ~> 'a::predomain
  2247   'a::pcpo ~> 'a::domain
  2248 
  2249 * The old type class "rep" has been superseded by class "domain".
  2250 Accordingly, users of the definitional package must remove any
  2251 "default_sort rep" declarations.  INCOMPATIBILITY.
  2252 
  2253 * The domain package (definitional mode) now supports unpointed
  2254 predomain argument types, as long as they are marked 'lazy'. (Strict
  2255 arguments must be in class "domain".) For example, the following
  2256 domain definition now works:
  2257 
  2258   domain natlist = nil | cons (lazy "nat discr") (lazy "natlist")
  2259 
  2260 * Theory HOLCF/Library/HOL_Cpo provides cpo and predomain class
  2261 instances for types from main HOL: bool, nat, int, char, 'a + 'b,
  2262 'a option, and 'a list.  Additionally, it configures fixrec and the
  2263 domain package to work with these types.  For example:
  2264 
  2265   fixrec isInl :: "('a + 'b) u -> tr"
  2266     where "isInl$(up$(Inl x)) = TT" | "isInl$(up$(Inr y)) = FF"
  2267 
  2268   domain V = VFun (lazy "V -> V") | VCon (lazy "nat") (lazy "V list")
  2269 
  2270 * The "(permissive)" option of fixrec has been replaced with a
  2271 per-equation "(unchecked)" option. See
  2272 src/HOL/HOLCF/Tutorial/Fixrec_ex.thy for examples. INCOMPATIBILITY.
  2273 
  2274 * The "bifinite" class no longer fixes a constant "approx"; the class
  2275 now just asserts that such a function exists.  INCOMPATIBILITY.
  2276 
  2277 * Former type "alg_defl" has been renamed to "defl".  HOLCF no longer
  2278 defines an embedding of type 'a defl into udom by default; instances
  2279 of "bifinite" and "domain" classes are available in
  2280 src/HOL/HOLCF/Library/Defl_Bifinite.thy.
  2281 
  2282 * The syntax "REP('a)" has been replaced with "DEFL('a)".
  2283 
  2284 * The predicate "directed" has been removed.  INCOMPATIBILITY.
  2285 
  2286 * The type class "finite_po" has been removed.  INCOMPATIBILITY.
  2287 
  2288 * The function "cprod_map" has been renamed to "prod_map".
  2289 INCOMPATIBILITY.
  2290 
  2291 * The monadic bind operator on each powerdomain has new binder syntax
  2292 similar to sets, e.g. "\<Union>\<sharp>x\<in>xs. t" represents
  2293 "upper_bind\<cdot>xs\<cdot>(\<Lambda> x. t)".
  2294 
  2295 * The infix syntax for binary union on each powerdomain has changed
  2296 from e.g. "+\<sharp>" to "\<union>\<sharp>", for consistency with set
  2297 syntax.  INCOMPATIBILITY.
  2298 
  2299 * The constant "UU" has been renamed to "bottom".  The syntax "UU" is
  2300 still supported as an input translation.
  2301 
  2302 * Renamed some theorems (the original names are also still available).
  2303 
  2304   expand_fun_below   ~> fun_below_iff
  2305   below_fun_ext      ~> fun_belowI
  2306   expand_cfun_eq     ~> cfun_eq_iff
  2307   ext_cfun           ~> cfun_eqI
  2308   expand_cfun_below  ~> cfun_below_iff
  2309   below_cfun_ext     ~> cfun_belowI
  2310   cont2cont_Rep_CFun ~> cont2cont_APP
  2311 
  2312 * The Abs and Rep functions for various types have changed names.
  2313 Related theorem names have also changed to match. INCOMPATIBILITY.
  2314 
  2315   Rep_CFun  ~> Rep_cfun
  2316   Abs_CFun  ~> Abs_cfun
  2317   Rep_Sprod ~> Rep_sprod
  2318   Abs_Sprod ~> Abs_sprod
  2319   Rep_Ssum  ~> Rep_ssum
  2320   Abs_Ssum  ~> Abs_ssum
  2321 
  2322 * Lemmas with names of the form *_defined_iff or *_strict_iff have
  2323 been renamed to *_bottom_iff.  INCOMPATIBILITY.
  2324 
  2325 * Various changes to bisimulation/coinduction with domain package:
  2326 
  2327   - Definitions of "bisim" constants no longer mention definedness.
  2328   - With mutual recursion, "bisim" predicate is now curried.
  2329   - With mutual recursion, each type gets a separate coind theorem.
  2330   - Variable names in bisim_def and coinduct rules have changed.
  2331 
  2332 INCOMPATIBILITY.
  2333 
  2334 * Case combinators generated by the domain package for type "foo" are
  2335 now named "foo_case" instead of "foo_when".  INCOMPATIBILITY.
  2336 
  2337 * Several theorems have been renamed to more accurately reflect the
  2338 names of constants and types involved.  INCOMPATIBILITY.
  2339 
  2340   thelub_const    ~> lub_const
  2341   lub_const       ~> is_lub_const
  2342   thelubI         ~> lub_eqI
  2343   is_lub_lub      ~> is_lubD2
  2344   lubI            ~> is_lub_lub
  2345   unique_lub      ~> is_lub_unique
  2346   is_ub_lub       ~> is_lub_rangeD1
  2347   lub_bin_chain   ~> is_lub_bin_chain
  2348   lub_fun         ~> is_lub_fun
  2349   thelub_fun      ~> lub_fun
  2350   thelub_cfun     ~> lub_cfun
  2351   thelub_Pair     ~> lub_Pair
  2352   lub_cprod       ~> is_lub_prod
  2353   thelub_cprod    ~> lub_prod
  2354   minimal_cprod   ~> minimal_prod
  2355   inst_cprod_pcpo ~> inst_prod_pcpo
  2356   UU_I            ~> bottomI
  2357   compact_UU      ~> compact_bottom
  2358   deflation_UU    ~> deflation_bottom
  2359   finite_deflation_UU ~> finite_deflation_bottom
  2360 
  2361 * Many legacy theorem names have been discontinued.  INCOMPATIBILITY.
  2362 
  2363   sq_ord_less_eq_trans ~> below_eq_trans
  2364   sq_ord_eq_less_trans ~> eq_below_trans
  2365   refl_less            ~> below_refl
  2366   trans_less           ~> below_trans
  2367   antisym_less         ~> below_antisym
  2368   antisym_less_inverse ~> po_eq_conv [THEN iffD1]
  2369   box_less             ~> box_below
  2370   rev_trans_less       ~> rev_below_trans
  2371   not_less2not_eq      ~> not_below2not_eq
  2372   less_UU_iff          ~> below_UU_iff
  2373   flat_less_iff        ~> flat_below_iff
  2374   adm_less             ~> adm_below
  2375   adm_not_less         ~> adm_not_below
  2376   adm_compact_not_less ~> adm_compact_not_below
  2377   less_fun_def         ~> below_fun_def
  2378   expand_fun_less      ~> fun_below_iff
  2379   less_fun_ext         ~> fun_belowI
  2380   less_discr_def       ~> below_discr_def
  2381   discr_less_eq        ~> discr_below_eq
  2382   less_unit_def        ~> below_unit_def
  2383   less_cprod_def       ~> below_prod_def
  2384   prod_lessI           ~> prod_belowI
  2385   Pair_less_iff        ~> Pair_below_iff
  2386   fst_less_iff         ~> fst_below_iff
  2387   snd_less_iff         ~> snd_below_iff
  2388   expand_cfun_less     ~> cfun_below_iff
  2389   less_cfun_ext        ~> cfun_belowI
  2390   injection_less       ~> injection_below
  2391   less_up_def          ~> below_up_def
  2392   not_Iup_less         ~> not_Iup_below
  2393   Iup_less             ~> Iup_below
  2394   up_less              ~> up_below
  2395   Def_inject_less_eq   ~> Def_below_Def
  2396   Def_less_is_eq       ~> Def_below_iff
  2397   spair_less_iff       ~> spair_below_iff
  2398   less_sprod           ~> below_sprod
  2399   spair_less           ~> spair_below
  2400   sfst_less_iff        ~> sfst_below_iff
  2401   ssnd_less_iff        ~> ssnd_below_iff
  2402   fix_least_less       ~> fix_least_below
  2403   dist_less_one        ~> dist_below_one
  2404   less_ONE             ~> below_ONE
  2405   ONE_less_iff         ~> ONE_below_iff
  2406   less_sinlD           ~> below_sinlD
  2407   less_sinrD           ~> below_sinrD
  2408 
  2409 
  2410 *** FOL and ZF ***
  2411 
  2412 * All constant names are now qualified internally and use proper
  2413 identifiers, e.g. "IFOL.eq" instead of "op =".  INCOMPATIBILITY.
  2414 
  2415 
  2416 *** ML ***
  2417 
  2418 * Antiquotation @{assert} inlines a function bool -> unit that raises
  2419 Fail if the argument is false.  Due to inlining the source position of
  2420 failed assertions is included in the error output.
  2421 
  2422 * Discontinued antiquotation @{theory_ref}, which is obsolete since ML
  2423 text is in practice always evaluated with a stable theory checkpoint.
  2424 Minor INCOMPATIBILITY, use (Theory.check_thy @{theory}) instead.
  2425 
  2426 * Antiquotation @{theory A} refers to theory A from the ancestry of
  2427 the current context, not any accidental theory loader state as before.
  2428 Potential INCOMPATIBILITY, subtle change in semantics.
  2429 
  2430 * Syntax.pretty_priority (default 0) configures the required priority
  2431 of pretty-printed output and thus affects insertion of parentheses.
  2432 
  2433 * Syntax.default_root (default "any") configures the inner syntax
  2434 category (nonterminal symbol) for parsing of terms.
  2435 
  2436 * Former exception Library.UnequalLengths now coincides with
  2437 ListPair.UnequalLengths.
  2438 
  2439 * Renamed structure MetaSimplifier to Raw_Simplifier.  Note that the
  2440 main functionality is provided by structure Simplifier.
  2441 
  2442 * Renamed raw "explode" function to "raw_explode" to emphasize its
  2443 meaning.  Note that internally to Isabelle, Symbol.explode is used in
  2444 almost all situations.
  2445 
  2446 * Discontinued obsolete function sys_error and exception SYS_ERROR.
  2447 See implementation manual for further details on exceptions in
  2448 Isabelle/ML.
  2449 
  2450 * Renamed setmp_noncritical to Unsynchronized.setmp to emphasize its
  2451 meaning.
  2452 
  2453 * Renamed structure PureThy to Pure_Thy and moved most of its
  2454 operations to structure Global_Theory, to emphasize that this is
  2455 rarely-used global-only stuff.
  2456 
  2457 * Discontinued Output.debug.  Minor INCOMPATIBILITY, use plain writeln
  2458 instead (or tracing for high-volume output).
  2459 
  2460 * Configuration option show_question_marks only affects regular pretty
  2461 printing of types and terms, not raw Term.string_of_vname.
  2462 
  2463 * ML_Context.thm and ML_Context.thms are no longer pervasive.  Rare
  2464 INCOMPATIBILITY, superseded by static antiquotations @{thm} and
  2465 @{thms} for most purposes.
  2466 
  2467 * ML structure Unsynchronized is never opened, not even in Isar
  2468 interaction mode as before.  Old Unsynchronized.set etc. have been
  2469 discontinued -- use plain := instead.  This should be *rare* anyway,
  2470 since modern tools always work via official context data, notably
  2471 configuration options.
  2472 
  2473 * Parallel and asynchronous execution requires special care concerning
  2474 interrupts.  Structure Exn provides some convenience functions that
  2475 avoid working directly with raw Interrupt.  User code must not absorb
  2476 interrupts -- intermediate handling (for cleanup etc.) needs to be
  2477 followed by re-raising of the original exception.  Another common
  2478 source of mistakes are "handle _" patterns, which make the meaning of
  2479 the program subject to physical effects of the environment.
  2480 
  2481 
  2482 
  2483 New in Isabelle2009-2 (June 2010)
  2484 ---------------------------------
  2485 
  2486 *** General ***
  2487 
  2488 * Authentic syntax for *all* logical entities (type classes, type
  2489 constructors, term constants): provides simple and robust
  2490 correspondence between formal entities and concrete syntax.  Within
  2491 the parse tree / AST representations, "constants" are decorated by
  2492 their category (class, type, const) and spelled out explicitly with
  2493 their full internal name.
  2494 
  2495 Substantial INCOMPATIBILITY concerning low-level syntax declarations
  2496 and translations (translation rules and translation functions in ML).
  2497 Some hints on upgrading:
  2498 
  2499   - Many existing uses of 'syntax' and 'translations' can be replaced
  2500     by more modern 'type_notation', 'notation' and 'abbreviation',
  2501     which are independent of this issue.
  2502 
  2503   - 'translations' require markup within the AST; the term syntax
  2504     provides the following special forms:
  2505 
  2506       CONST c   -- produces syntax version of constant c from context
  2507       XCONST c  -- literally c, checked as constant from context
  2508       c         -- literally c, if declared by 'syntax'
  2509 
  2510     Plain identifiers are treated as AST variables -- occasionally the
  2511     system indicates accidental variables via the error "rhs contains
  2512     extra variables".
  2513 
  2514     Type classes and type constructors are marked according to their
  2515     concrete syntax.  Some old translations rules need to be written
  2516     for the "type" category, using type constructor application
  2517     instead of pseudo-term application of the default category
  2518     "logic".
  2519 
  2520   - 'parse_translation' etc. in ML may use the following
  2521     antiquotations:
  2522 
  2523       @{class_syntax c}   -- type class c within parse tree / AST
  2524       @{term_syntax c}    -- type constructor c within parse tree / AST
  2525       @{const_syntax c}   -- ML version of "CONST c" above
  2526       @{syntax_const c}   -- literally c (checked wrt. 'syntax' declarations)
  2527 
  2528   - Literal types within 'typed_print_translations', i.e. those *not*
  2529     represented as pseudo-terms are represented verbatim.  Use @{class
  2530     c} or @{type_name c} here instead of the above syntax
  2531     antiquotations.
  2532 
  2533 Note that old non-authentic syntax was based on unqualified base
  2534 names, so all of the above "constant" names would coincide.  Recall
  2535 that 'print_syntax' and ML_command "set Syntax.trace_ast" help to
  2536 diagnose syntax problems.
  2537 
  2538 * Type constructors admit general mixfix syntax, not just infix.
  2539 
  2540 * Concrete syntax may be attached to local entities without a proof
  2541 body, too.  This works via regular mixfix annotations for 'fix',
  2542 'def', 'obtain' etc. or via the explicit 'write' command, which is
  2543 similar to the 'notation' command in theory specifications.
  2544 
  2545 * Discontinued unnamed infix syntax (legacy feature for many years) --
  2546 need to specify constant name and syntax separately.  Internal ML
  2547 datatype constructors have been renamed from InfixName to Infix etc.
  2548 Minor INCOMPATIBILITY.
  2549 
  2550 * Schematic theorem statements need to be explicitly markup as such,
  2551 via commands 'schematic_lemma', 'schematic_theorem',
  2552 'schematic_corollary'.  Thus the relevance of the proof is made
  2553 syntactically clear, which impacts performance in a parallel or
  2554 asynchronous interactive environment.  Minor INCOMPATIBILITY.
  2555 
  2556 * Use of cumulative prems via "!" in some proof methods has been
  2557 discontinued (old legacy feature).
  2558 
  2559 * References 'trace_simp' and 'debug_simp' have been replaced by
  2560 configuration options stored in the context. Enabling tracing (the
  2561 case of debugging is similar) in proofs works via
  2562 
  2563   using [[trace_simp = true]]
  2564 
  2565 Tracing is then active for all invocations of the simplifier in
  2566 subsequent goal refinement steps. Tracing may also still be enabled or
  2567 disabled via the ProofGeneral settings menu.
  2568 
  2569 * Separate commands 'hide_class', 'hide_type', 'hide_const',
  2570 'hide_fact' replace the former 'hide' KIND command.  Minor
  2571 INCOMPATIBILITY.
  2572 
  2573 * Improved parallelism of proof term normalization: usedir -p2 -q0 is
  2574 more efficient than combinations with -q1 or -q2.
  2575 
  2576 
  2577 *** Pure ***
  2578 
  2579 * Proofterms record type-class reasoning explicitly, using the
  2580 "unconstrain" operation internally.  This eliminates all sort
  2581 constraints from a theorem and proof, introducing explicit
  2582 OFCLASS-premises.  On the proof term level, this operation is
  2583 automatically applied at theorem boundaries, such that closed proofs
  2584 are always free of sort constraints.  INCOMPATIBILITY for tools that
  2585 inspect proof terms.
  2586 
  2587 * Local theory specifications may depend on extra type variables that
  2588 are not present in the result type -- arguments TYPE('a) :: 'a itself
  2589 are added internally.  For example:
  2590 
  2591   definition unitary :: bool where "unitary = (ALL (x::'a) y. x = y)"
  2592 
  2593 * Predicates of locales introduced by classes carry a mandatory
  2594 "class" prefix.  INCOMPATIBILITY.
  2595 
  2596 * Vacuous class specifications observe default sort.  INCOMPATIBILITY.
  2597 
  2598 * Old 'axclass' command has been discontinued.  INCOMPATIBILITY, use
  2599 'class' instead.
  2600 
  2601 * Command 'code_reflect' allows to incorporate generated ML code into
  2602 runtime environment; replaces immature code_datatype antiquotation.
  2603 INCOMPATIBILITY.
  2604 
  2605 * Code generator: simple concept for abstract datatypes obeying
  2606 invariants.
  2607 
  2608 * Code generator: details of internal data cache have no impact on the
  2609 user space functionality any longer.
  2610 
  2611 * Methods "unfold_locales" and "intro_locales" ignore non-locale
  2612 subgoals.  This is more appropriate for interpretations with 'where'.
  2613 INCOMPATIBILITY.
  2614 
  2615 * Command 'example_proof' opens an empty proof body.  This allows to
  2616 experiment with Isar, without producing any persistent result.
  2617 
  2618 * Commands 'type_notation' and 'no_type_notation' declare type syntax
  2619 within a local theory context, with explicit checking of the
  2620 constructors involved (in contrast to the raw 'syntax' versions).
  2621 
  2622 * Commands 'types' and 'typedecl' now work within a local theory
  2623 context -- without introducing dependencies on parameters or
  2624 assumptions, which is not possible in Isabelle/Pure.
  2625 
  2626 * Command 'defaultsort' has been renamed to 'default_sort', it works
  2627 within a local theory context.  Minor INCOMPATIBILITY.
  2628 
  2629 
  2630 *** HOL ***
  2631 
  2632 * Command 'typedef' now works within a local theory context -- without
  2633 introducing dependencies on parameters or assumptions, which is not
  2634 possible in Isabelle/Pure/HOL.  Note that the logical environment may
  2635 contain multiple interpretations of local typedefs (with different
  2636 non-emptiness proofs), even in a global theory context.
  2637 
  2638 * New package for quotient types.  Commands 'quotient_type' and
  2639 'quotient_definition' may be used for defining types and constants by
  2640 quotient constructions.  An example is the type of integers created by
  2641 quotienting pairs of natural numbers:
  2642 
  2643   fun
  2644     intrel :: "(nat * nat) => (nat * nat) => bool"
  2645   where
  2646     "intrel (x, y) (u, v) = (x + v = u + y)"
  2647 
  2648   quotient_type int = "nat * nat" / intrel
  2649     by (auto simp add: equivp_def expand_fun_eq)
  2650 
  2651   quotient_definition
  2652     "0::int" is "(0::nat, 0::nat)"
  2653 
  2654 The method "lifting" can be used to lift of theorems from the
  2655 underlying "raw" type to the quotient type.  The example
  2656 src/HOL/Quotient_Examples/FSet.thy includes such a quotient
  2657 construction and provides a reasoning infrastructure for finite sets.
  2658 
  2659 * Renamed Library/Quotient.thy to Library/Quotient_Type.thy to avoid
  2660 clash with new theory Quotient in Main HOL.
  2661 
  2662 * Moved the SMT binding into the main HOL session, eliminating
  2663 separate HOL-SMT session.
  2664 
  2665 * List membership infix mem operation is only an input abbreviation.
  2666 INCOMPATIBILITY.
  2667 
  2668 * Theory Library/Word.thy has been removed.  Use library Word/Word.thy
  2669 for future developements; former Library/Word.thy is still present in
  2670 the AFP entry RSAPPS.
  2671 
  2672 * Theorem Int.int_induct renamed to Int.int_of_nat_induct and is no
  2673 longer shadowed.  INCOMPATIBILITY.
  2674 
  2675 * Dropped theorem duplicate comp_arith; use semiring_norm instead.
  2676 INCOMPATIBILITY.
  2677 
  2678 * Dropped theorem RealPow.real_sq_order; use power2_le_imp_le instead.
  2679 INCOMPATIBILITY.
  2680 
  2681 * Dropped normalizing_semiring etc; use the facts in semiring classes
  2682 instead.  INCOMPATIBILITY.
  2683 
  2684 * Dropped several real-specific versions of lemmas about floor and
  2685 ceiling; use the generic lemmas from theory "Archimedean_Field"
  2686 instead.  INCOMPATIBILITY.
  2687 
  2688   floor_number_of_eq         ~> floor_number_of
  2689   le_floor_eq_number_of      ~> number_of_le_floor
  2690   le_floor_eq_zero           ~> zero_le_floor
  2691   le_floor_eq_one            ~> one_le_floor
  2692   floor_less_eq_number_of    ~> floor_less_number_of
  2693   floor_less_eq_zero         ~> floor_less_zero
  2694   floor_less_eq_one          ~> floor_less_one
  2695   less_floor_eq_number_of    ~> number_of_less_floor
  2696   less_floor_eq_zero         ~> zero_less_floor
  2697   less_floor_eq_one          ~> one_less_floor
  2698   floor_le_eq_number_of      ~> floor_le_number_of
  2699   floor_le_eq_zero           ~> floor_le_zero
  2700   floor_le_eq_one            ~> floor_le_one
  2701   floor_subtract_number_of   ~> floor_diff_number_of
  2702   floor_subtract_one         ~> floor_diff_one
  2703   ceiling_number_of_eq       ~> ceiling_number_of
  2704   ceiling_le_eq_number_of    ~> ceiling_le_number_of
  2705   ceiling_le_zero_eq         ~> ceiling_le_zero
  2706   ceiling_le_eq_one          ~> ceiling_le_one
  2707   less_ceiling_eq_number_of  ~> number_of_less_ceiling
  2708   less_ceiling_eq_zero       ~> zero_less_ceiling
  2709   less_ceiling_eq_one        ~> one_less_ceiling
  2710   ceiling_less_eq_number_of  ~> ceiling_less_number_of
  2711   ceiling_less_eq_zero       ~> ceiling_less_zero
  2712   ceiling_less_eq_one        ~> ceiling_less_one
  2713   le_ceiling_eq_number_of    ~> number_of_le_ceiling
  2714   le_ceiling_eq_zero         ~> zero_le_ceiling
  2715   le_ceiling_eq_one          ~> one_le_ceiling
  2716   ceiling_subtract_number_of ~> ceiling_diff_number_of
  2717   ceiling_subtract_one       ~> ceiling_diff_one
  2718 
  2719 * Theory "Finite_Set": various folding_XXX locales facilitate the
  2720 application of the various fold combinators on finite sets.
  2721 
  2722 * Library theory "RBT" renamed to "RBT_Impl"; new library theory "RBT"
  2723 provides abstract red-black tree type which is backed by "RBT_Impl" as
  2724 implementation.  INCOMPATIBILTY.
  2725 
  2726 * Theory Library/Coinductive_List has been removed -- superseded by
  2727 AFP/thys/Coinductive.
  2728 
  2729 * Theory PReal, including the type "preal" and related operations, has
  2730 been removed.  INCOMPATIBILITY.
  2731 
  2732 * Real: new development using Cauchy Sequences.
  2733 
  2734 * Split off theory "Big_Operators" containing setsum, setprod,
  2735 Inf_fin, Sup_fin, Min, Max from theory Finite_Set.  INCOMPATIBILITY.
  2736 
  2737 * Theory "Rational" renamed to "Rat", for consistency with "Nat",
  2738 "Int" etc.  INCOMPATIBILITY.
  2739 
  2740 * Constant Rat.normalize needs to be qualified.  INCOMPATIBILITY.
  2741 
  2742 * New set of rules "ac_simps" provides combined assoc / commute
  2743 rewrites for all interpretations of the appropriate generic locales.
  2744 
  2745 * Renamed theory "OrderedGroup" to "Groups" and split theory
  2746 "Ring_and_Field" into theories "Rings" and "Fields"; for more
  2747 appropriate and more consistent names suitable for name prefixes
  2748 within the HOL theories.  INCOMPATIBILITY.
  2749 
  2750 * Some generic constants have been put to appropriate theories:
  2751   - less_eq, less: Orderings
  2752   - zero, one, plus, minus, uminus, times, abs, sgn: Groups
  2753   - inverse, divide: Rings
  2754 INCOMPATIBILITY.
  2755 
  2756 * More consistent naming of type classes involving orderings (and
  2757 lattices):
  2758 
  2759     lower_semilattice                   ~> semilattice_inf
  2760     upper_semilattice                   ~> semilattice_sup
  2761 
  2762     dense_linear_order                  ~> dense_linorder
  2763 
  2764     pordered_ab_group_add               ~> ordered_ab_group_add
  2765     pordered_ab_group_add_abs           ~> ordered_ab_group_add_abs
  2766     pordered_ab_semigroup_add           ~> ordered_ab_semigroup_add
  2767     pordered_ab_semigroup_add_imp_le    ~> ordered_ab_semigroup_add_imp_le
  2768     pordered_cancel_ab_semigroup_add    ~> ordered_cancel_ab_semigroup_add
  2769     pordered_cancel_comm_semiring       ~> ordered_cancel_comm_semiring
  2770     pordered_cancel_semiring            ~> ordered_cancel_semiring
  2771     pordered_comm_monoid_add            ~> ordered_comm_monoid_add
  2772     pordered_comm_ring                  ~> ordered_comm_ring
  2773     pordered_comm_semiring              ~> ordered_comm_semiring
  2774     pordered_ring                       ~> ordered_ring
  2775     pordered_ring_abs                   ~> ordered_ring_abs
  2776     pordered_semiring                   ~> ordered_semiring
  2777 
  2778     ordered_ab_group_add                ~> linordered_ab_group_add
  2779     ordered_ab_semigroup_add            ~> linordered_ab_semigroup_add
  2780     ordered_cancel_ab_semigroup_add     ~> linordered_cancel_ab_semigroup_add
  2781     ordered_comm_semiring_strict        ~> linordered_comm_semiring_strict
  2782     ordered_field                       ~> linordered_field
  2783     ordered_field_no_lb                 ~> linordered_field_no_lb
  2784     ordered_field_no_ub                 ~> linordered_field_no_ub
  2785     ordered_field_dense_linear_order    ~> dense_linordered_field
  2786     ordered_idom                        ~> linordered_idom
  2787     ordered_ring                        ~> linordered_ring
  2788     ordered_ring_le_cancel_factor       ~> linordered_ring_le_cancel_factor
  2789     ordered_ring_less_cancel_factor     ~> linordered_ring_less_cancel_factor
  2790     ordered_ring_strict                 ~> linordered_ring_strict
  2791     ordered_semidom                     ~> linordered_semidom
  2792     ordered_semiring                    ~> linordered_semiring
  2793     ordered_semiring_1                  ~> linordered_semiring_1
  2794     ordered_semiring_1_strict           ~> linordered_semiring_1_strict
  2795     ordered_semiring_strict             ~> linordered_semiring_strict
  2796 
  2797   The following slightly odd type classes have been moved to a
  2798   separate theory Library/Lattice_Algebras:
  2799 
  2800     lordered_ab_group_add               ~> lattice_ab_group_add
  2801     lordered_ab_group_add_abs           ~> lattice_ab_group_add_abs
  2802     lordered_ab_group_add_meet          ~> semilattice_inf_ab_group_add
  2803     lordered_ab_group_add_join          ~> semilattice_sup_ab_group_add
  2804     lordered_ring                       ~> lattice_ring
  2805 
  2806 INCOMPATIBILITY.
  2807 
  2808 * Refined field classes:
  2809   - classes division_ring_inverse_zero, field_inverse_zero,
  2810     linordered_field_inverse_zero include rule inverse 0 = 0 --
  2811     subsumes former division_by_zero class;
  2812   - numerous lemmas have been ported from field to division_ring.
  2813 INCOMPATIBILITY.
  2814 
  2815 * Refined algebra theorem collections:
  2816   - dropped theorem group group_simps, use algebra_simps instead;
  2817   - dropped theorem group ring_simps, use field_simps instead;
  2818   - proper theorem collection field_simps subsumes former theorem
  2819     groups field_eq_simps and field_simps;
  2820   - dropped lemma eq_minus_self_iff which is a duplicate for
  2821     equal_neg_zero.
  2822 INCOMPATIBILITY.
  2823 
  2824 * Theory Finite_Set and List: some lemmas have been generalized from
  2825 sets to lattices:
  2826 
  2827   fun_left_comm_idem_inter      ~> fun_left_comm_idem_inf
  2828   fun_left_comm_idem_union      ~> fun_left_comm_idem_sup
  2829   inter_Inter_fold_inter        ~> inf_Inf_fold_inf
  2830   union_Union_fold_union        ~> sup_Sup_fold_sup
  2831   Inter_fold_inter              ~> Inf_fold_inf
  2832   Union_fold_union              ~> Sup_fold_sup
  2833   inter_INTER_fold_inter        ~> inf_INFI_fold_inf
  2834   union_UNION_fold_union        ~> sup_SUPR_fold_sup
  2835   INTER_fold_inter              ~> INFI_fold_inf
  2836   UNION_fold_union              ~> SUPR_fold_sup
  2837 
  2838 * Theory "Complete_Lattice": lemmas top_def and bot_def have been
  2839 replaced by the more convenient lemmas Inf_empty and Sup_empty.
  2840 Dropped lemmas Inf_insert_simp and Sup_insert_simp, which are subsumed
  2841 by Inf_insert and Sup_insert.  Lemmas Inf_UNIV and Sup_UNIV replace
  2842 former Inf_Univ and Sup_Univ.  Lemmas inf_top_right and sup_bot_right
  2843 subsume inf_top and sup_bot respectively.  INCOMPATIBILITY.
  2844 
  2845 * Reorganized theory Multiset: swapped notation of pointwise and
  2846 multiset order:
  2847 
  2848   - pointwise ordering is instance of class order with standard syntax
  2849     <= and <;
  2850   - multiset ordering has syntax <=# and <#; partial order properties
  2851     are provided by means of interpretation with prefix
  2852     multiset_order;
  2853   - less duplication, less historical organization of sections,
  2854     conversion from associations lists to multisets, rudimentary code
  2855     generation;
  2856   - use insert_DiffM2 [symmetric] instead of elem_imp_eq_diff_union,
  2857     if needed.
  2858 
  2859 Renamed:
  2860 
  2861   multiset_eq_conv_count_eq  ~>  multiset_ext_iff
  2862   multi_count_ext  ~>  multiset_ext
  2863   diff_union_inverse2  ~>  diff_union_cancelR
  2864 
  2865 INCOMPATIBILITY.
  2866 
  2867 * Theory Permutation: replaced local "remove" by List.remove1.
  2868 
  2869 * Code generation: ML and OCaml code is decorated with signatures.
  2870 
  2871 * Theory List: added transpose.
  2872 
  2873 * Library/Nat_Bijection.thy is a collection of bijective functions
  2874 between nat and other types, which supersedes the older libraries
  2875 Library/Nat_Int_Bij.thy and HOLCF/NatIso.thy.  INCOMPATIBILITY.
  2876 
  2877   Constants:
  2878   Nat_Int_Bij.nat2_to_nat         ~> prod_encode
  2879   Nat_Int_Bij.nat_to_nat2         ~> prod_decode
  2880   Nat_Int_Bij.int_to_nat_bij      ~> int_encode
  2881   Nat_Int_Bij.nat_to_int_bij      ~> int_decode
  2882   Countable.pair_encode           ~> prod_encode
  2883   NatIso.prod2nat                 ~> prod_encode
  2884   NatIso.nat2prod                 ~> prod_decode
  2885   NatIso.sum2nat                  ~> sum_encode
  2886   NatIso.nat2sum                  ~> sum_decode
  2887   NatIso.list2nat                 ~> list_encode
  2888   NatIso.nat2list                 ~> list_decode
  2889   NatIso.set2nat                  ~> set_encode
  2890   NatIso.nat2set                  ~> set_decode
  2891 
  2892   Lemmas:
  2893   Nat_Int_Bij.bij_nat_to_int_bij  ~> bij_int_decode
  2894   Nat_Int_Bij.nat2_to_nat_inj     ~> inj_prod_encode
  2895   Nat_Int_Bij.nat2_to_nat_surj    ~> surj_prod_encode
  2896   Nat_Int_Bij.nat_to_nat2_inj     ~> inj_prod_decode
  2897   Nat_Int_Bij.nat_to_nat2_surj    ~> surj_prod_decode
  2898   Nat_Int_Bij.i2n_n2i_id          ~> int_encode_inverse
  2899   Nat_Int_Bij.n2i_i2n_id          ~> int_decode_inverse
  2900   Nat_Int_Bij.surj_nat_to_int_bij ~> surj_int_encode
  2901   Nat_Int_Bij.surj_int_to_nat_bij ~> surj_int_decode
  2902   Nat_Int_Bij.inj_nat_to_int_bij  ~> inj_int_encode
  2903   Nat_Int_Bij.inj_int_to_nat_bij  ~> inj_int_decode
  2904   Nat_Int_Bij.bij_nat_to_int_bij  ~> bij_int_encode
  2905   Nat_Int_Bij.bij_int_to_nat_bij  ~> bij_int_decode
  2906 
  2907 * Sledgehammer:
  2908   - Renamed ATP commands:
  2909     atp_info     ~> sledgehammer running_atps
  2910     atp_kill     ~> sledgehammer kill_atps
  2911     atp_messages ~> sledgehammer messages
  2912     atp_minimize ~> sledgehammer minimize
  2913     print_atps   ~> sledgehammer available_atps
  2914     INCOMPATIBILITY.
  2915   - Added user's manual ("isabelle doc sledgehammer").
  2916   - Added option syntax and "sledgehammer_params" to customize
  2917     Sledgehammer's behavior.  See the manual for details.
  2918   - Modified the Isar proof reconstruction code so that it produces
  2919     direct proofs rather than proofs by contradiction.  (This feature
  2920     is still experimental.)
  2921   - Made Isar proof reconstruction work for SPASS, remote ATPs, and in
  2922     full-typed mode.
  2923   - Added support for TPTP syntax for SPASS via the "spass_tptp" ATP.
  2924 
  2925 * Nitpick:
  2926   - Added and implemented "binary_ints" and "bits" options.
  2927   - Added "std" option and implemented support for nonstandard models.
  2928   - Added and implemented "finitize" option to improve the precision
  2929     of infinite datatypes based on a monotonicity analysis.
  2930   - Added support for quotient types.
  2931   - Added support for "specification" and "ax_specification"
  2932     constructs.
  2933   - Added support for local definitions (for "function" and
  2934     "termination" proofs).
  2935   - Added support for term postprocessors.
  2936   - Optimized "Multiset.multiset" and "FinFun.finfun".
  2937   - Improved efficiency of "destroy_constrs" optimization.
  2938   - Fixed soundness bugs related to "destroy_constrs" optimization and
  2939     record getters.
  2940   - Fixed soundness bug related to higher-order constructors.
  2941   - Fixed soundness bug when "full_descrs" is enabled.
  2942   - Improved precision of set constructs.
  2943   - Added "atoms" option.
  2944   - Added cache to speed up repeated Kodkod invocations on the same
  2945     problems.
  2946   - Renamed "MiniSatJNI", "zChaffJNI", "BerkMinAlloy", and
  2947     "SAT4JLight" to "MiniSat_JNI", "zChaff_JNI", "BerkMin_Alloy", and
  2948     "SAT4J_Light".  INCOMPATIBILITY.
  2949   - Removed "skolemize", "uncurry", "sym_break", "flatten_prop",
  2950     "sharing_depth", and "show_skolems" options.  INCOMPATIBILITY.
  2951   - Removed "nitpick_intro" attribute.  INCOMPATIBILITY.
  2952 
  2953 * Method "induct" now takes instantiations of the form t, where t is not
  2954   a variable, as a shorthand for "x == t", where x is a fresh variable.
  2955   If this is not intended, t has to be enclosed in parentheses.
  2956   By default, the equalities generated by definitional instantiations
  2957   are pre-simplified, which may cause parameters of inductive cases
  2958   to disappear, or may even delete some of the inductive cases.
  2959   Use "induct (no_simp)" instead of "induct" to restore the old
  2960   behaviour. The (no_simp) option is also understood by the "cases"
  2961   and "nominal_induct" methods, which now perform pre-simplification, too.
  2962   INCOMPATIBILITY.
  2963 
  2964 
  2965 *** HOLCF ***
  2966 
  2967 * Variable names in lemmas generated by the domain package have
  2968 changed; the naming scheme is now consistent with the HOL datatype
  2969 package.  Some proof scripts may be affected, INCOMPATIBILITY.
  2970 
  2971 * The domain package no longer defines the function "foo_copy" for
  2972 recursive domain "foo".  The reach lemma is now stated directly in
  2973 terms of "foo_take".  Lemmas and proofs that mention "foo_copy" must
  2974 be reformulated in terms of "foo_take", INCOMPATIBILITY.
  2975 
  2976 * Most definedness lemmas generated by the domain package (previously
  2977 of the form "x ~= UU ==> foo$x ~= UU") now have an if-and-only-if form
  2978 like "foo$x = UU <-> x = UU", which works better as a simp rule.
  2979 Proofs that used definedness lemmas as intro rules may break,
  2980 potential INCOMPATIBILITY.
  2981 
  2982 * Induction and casedist rules generated by the domain package now
  2983 declare proper case_names (one called "bottom", and one named for each
  2984 constructor).  INCOMPATIBILITY.
  2985 
  2986 * For mutually-recursive domains, separate "reach" and "take_lemma"
  2987 rules are generated for each domain, INCOMPATIBILITY.
  2988 
  2989   foo_bar.reach       ~> foo.reach  bar.reach
  2990   foo_bar.take_lemmas ~> foo.take_lemma  bar.take_lemma
  2991 
  2992 * Some lemmas generated by the domain package have been renamed for
  2993 consistency with the datatype package, INCOMPATIBILITY.
  2994 
  2995   foo.ind        ~> foo.induct
  2996   foo.finite_ind ~> foo.finite_induct
  2997   foo.coind      ~> foo.coinduct
  2998   foo.casedist   ~> foo.exhaust
  2999   foo.exhaust    ~> foo.nchotomy
  3000 
  3001 * For consistency with other definition packages, the fixrec package
  3002 now generates qualified theorem names, INCOMPATIBILITY.
  3003 
  3004   foo_simps  ~> foo.simps
  3005   foo_unfold ~> foo.unfold
  3006   foo_induct ~> foo.induct
  3007 
  3008 * The "fixrec_simp" attribute has been removed.  The "fixrec_simp"
  3009 method and internal fixrec proofs now use the default simpset instead.
  3010 INCOMPATIBILITY.
  3011 
  3012 * The "contlub" predicate has been removed.  Proof scripts should use
  3013 lemma contI2 in place of monocontlub2cont, INCOMPATIBILITY.
  3014 
  3015 * The "admw" predicate has been removed, INCOMPATIBILITY.
  3016 
  3017 * The constants cpair, cfst, and csnd have been removed in favor of
  3018 Pair, fst, and snd from Isabelle/HOL, INCOMPATIBILITY.
  3019 
  3020 
  3021 *** ML ***
  3022 
  3023 * Antiquotations for basic formal entities:
  3024 
  3025     @{class NAME}         -- type class
  3026     @{class_syntax NAME}  -- syntax representation of the above
  3027 
  3028     @{type_name NAME}     -- logical type
  3029     @{type_abbrev NAME}   -- type abbreviation
  3030     @{nonterminal NAME}   -- type of concrete syntactic category
  3031     @{type_syntax NAME}   -- syntax representation of any of the above
  3032 
  3033     @{const_name NAME}    -- logical constant (INCOMPATIBILITY)
  3034     @{const_abbrev NAME}  -- abbreviated constant
  3035     @{const_syntax NAME}  -- syntax representation of any of the above
  3036 
  3037 * Antiquotation @{syntax_const NAME} ensures that NAME refers to a raw
  3038 syntax constant (cf. 'syntax' command).
  3039 
  3040 * Antiquotation @{make_string} inlines a function to print arbitrary
  3041 values similar to the ML toplevel.  The result is compiler dependent
  3042 and may fall back on "?" in certain situations.
  3043 
  3044 * Diagnostic commands 'ML_val' and 'ML_command' may refer to
  3045 antiquotations @{Isar.state} and @{Isar.goal}.  This replaces impure
  3046 Isar.state() and Isar.goal(), which belong to the old TTY loop and do
  3047 not work with the asynchronous Isar document model.
  3048 
  3049 * Configuration options now admit dynamic default values, depending on
  3050 the context or even global references.
  3051 
  3052 * SHA1.digest digests strings according to SHA-1 (see RFC 3174).  It
  3053 uses an efficient external library if available (for Poly/ML).
  3054 
  3055 * Renamed some important ML structures, while keeping the old names
  3056 for some time as aliases within the structure Legacy:
  3057 
  3058   OuterKeyword  ~>  Keyword
  3059   OuterLex      ~>  Token
  3060   OuterParse    ~>  Parse
  3061   OuterSyntax   ~>  Outer_Syntax
  3062   PrintMode     ~>  Print_Mode
  3063   SpecParse     ~>  Parse_Spec
  3064   ThyInfo       ~>  Thy_Info
  3065   ThyLoad       ~>  Thy_Load
  3066   ThyOutput     ~>  Thy_Output
  3067   TypeInfer     ~>  Type_Infer
  3068 
  3069 Note that "open Legacy" simplifies porting of sources, but forgetting
  3070 to remove it again will complicate porting again in the future.
  3071 
  3072 * Most operations that refer to a global context are named
  3073 accordingly, e.g. Simplifier.global_context or
  3074 ProofContext.init_global.  There are some situations where a global
  3075 context actually works, but under normal circumstances one needs to
  3076 pass the proper local context through the code!
  3077 
  3078 * Discontinued old TheoryDataFun with its copy/init operation -- data
  3079 needs to be pure.  Functor Theory_Data_PP retains the traditional
  3080 Pretty.pp argument to merge, which is absent in the standard
  3081 Theory_Data version.
  3082 
  3083 * Sorts.certify_sort and derived "cert" operations for types and terms
  3084 no longer minimize sorts.  Thus certification at the boundary of the
  3085 inference kernel becomes invariant under addition of class relations,
  3086 which is an important monotonicity principle.  Sorts are now minimized
  3087 in the syntax layer only, at the boundary between the end-user and the
  3088 system.  Subtle INCOMPATIBILITY, may have to use Sign.minimize_sort
  3089 explicitly in rare situations.
  3090 
  3091 * Renamed old-style Drule.standard to Drule.export_without_context, to
  3092 emphasize that this is in no way a standard operation.
  3093 INCOMPATIBILITY.
  3094 
  3095 * Subgoal.FOCUS (and variants): resulting goal state is normalized as
  3096 usual for resolution.  Rare INCOMPATIBILITY.
  3097 
  3098 * Renamed varify/unvarify operations to varify_global/unvarify_global
  3099 to emphasize that these only work in a global situation (which is
  3100 quite rare).
  3101 
  3102 * Curried take and drop in library.ML; negative length is interpreted
  3103 as infinity (as in chop).  Subtle INCOMPATIBILITY.
  3104 
  3105 * Proof terms: type substitutions on proof constants now use canonical
  3106 order of type variables.  INCOMPATIBILITY for tools working with proof
  3107 terms.
  3108 
  3109 * Raw axioms/defs may no longer carry sort constraints, and raw defs
  3110 may no longer carry premises.  User-level specifications are
  3111 transformed accordingly by Thm.add_axiom/add_def.
  3112 
  3113 
  3114 *** System ***
  3115 
  3116 * Discontinued special HOL_USEDIR_OPTIONS for the main HOL image;
  3117 ISABELLE_USEDIR_OPTIONS applies uniformly to all sessions.  Note that
  3118 proof terms are enabled unconditionally in the new HOL-Proofs image.
  3119 
  3120 * Discontinued old ISABELLE and ISATOOL environment settings (legacy
  3121 feature since Isabelle2009).  Use ISABELLE_PROCESS and ISABELLE_TOOL,
  3122 respectively.
  3123 
  3124 * Old lib/scripts/polyml-platform is superseded by the
  3125 ISABELLE_PLATFORM setting variable, which defaults to the 32 bit
  3126 variant, even on a 64 bit machine.  The following example setting
  3127 prefers 64 bit if available:
  3128 
  3129   ML_PLATFORM="${ISABELLE_PLATFORM64:-$ISABELLE_PLATFORM}"
  3130 
  3131 * The preliminary Isabelle/jEdit application demonstrates the emerging
  3132 Isabelle/Scala layer for advanced prover interaction and integration.
  3133 See src/Tools/jEdit or "isabelle jedit" provided by the properly built
  3134 component.
  3135 
  3136 * "IsabelleText" is a Unicode font derived from Bitstream Vera Mono
  3137 and Bluesky TeX fonts.  It provides the usual Isabelle symbols,
  3138 similar to the default assignment of the document preparation system
  3139 (cf. isabellesym.sty).  The Isabelle/Scala class Isabelle_System
  3140 provides some operations for direct access to the font without asking
  3141 the user for manual installation.
  3142 
  3143 
  3144 
  3145 New in Isabelle2009-1 (December 2009)
  3146 -------------------------------------
  3147 
  3148 *** General ***
  3149 
  3150 * Discontinued old form of "escaped symbols" such as \\<forall>.  Only
  3151 one backslash should be used, even in ML sources.
  3152 
  3153 
  3154 *** Pure ***
  3155 
  3156 * Locale interpretation propagates mixins along the locale hierarchy.
  3157 The currently only available mixins are the equations used to map
  3158 local definitions to terms of the target domain of an interpretation.
  3159 
  3160 * Reactivated diagnostic command 'print_interps'.  Use "print_interps
  3161 loc" to print all interpretations of locale "loc" in the theory.
  3162 Interpretations in proofs are not shown.
  3163 
  3164 * Thoroughly revised locales tutorial.  New section on conditional
  3165 interpretation.
  3166 
  3167 * On instantiation of classes, remaining undefined class parameters
  3168 are formally declared.  INCOMPATIBILITY.
  3169 
  3170 
  3171 *** Document preparation ***
  3172 
  3173 * New generalized style concept for printing terms: @{foo (style) ...}
  3174 instead of @{foo_style style ...}  (old form is still retained for
  3175 backward compatibility).  Styles can be also applied for
  3176 antiquotations prop, term_type and typeof.
  3177 
  3178 
  3179 *** HOL ***
  3180 
  3181 * New proof method "smt" for a combination of first-order logic with
  3182 equality, linear and nonlinear (natural/integer/real) arithmetic, and
  3183 fixed-size bitvectors; there is also basic support for higher-order
  3184 features (esp. lambda abstractions).  It is an incomplete decision
  3185 procedure based on external SMT solvers using the oracle mechanism;
  3186 for the SMT solver Z3, this method is proof-producing.  Certificates
  3187 are provided to avoid calling the external solvers solely for
  3188 re-checking proofs.  Due to a remote SMT service there is no need for
  3189 installing SMT solvers locally.  See src/HOL/SMT.
  3190 
  3191 * New commands to load and prove verification conditions generated by
  3192 the Boogie program verifier or derived systems (e.g. the Verifying C
  3193 Compiler (VCC) or Spec#).  See src/HOL/Boogie.
  3194 
  3195 * New counterexample generator tool 'nitpick' based on the Kodkod
  3196 relational model finder.  See src/HOL/Tools/Nitpick and
  3197 src/HOL/Nitpick_Examples.
  3198 
  3199 * New commands 'code_pred' and 'values' to invoke the predicate
  3200 compiler and to enumerate values of inductive predicates.
  3201 
  3202 * A tabled implementation of the reflexive transitive closure.
  3203 
  3204 * New implementation of quickcheck uses generic code generator;
  3205 default generators are provided for all suitable HOL types, records
  3206 and datatypes.  Old quickcheck can be re-activated importing theory
  3207 Library/SML_Quickcheck.
  3208 
  3209 * New testing tool Mirabelle for automated proof tools.  Applies
  3210 several tools and tactics like sledgehammer, metis, or quickcheck, to
  3211 every proof step in a theory.  To be used in batch mode via the
  3212 "mirabelle" utility.
  3213 
  3214 * New proof method "sos" (sum of squares) for nonlinear real
  3215 arithmetic (originally due to John Harison). It requires theory
  3216 Library/Sum_Of_Squares.  It is not a complete decision procedure but
  3217 works well in practice on quantifier-free real arithmetic with +, -,
  3218 *, ^, =, <= and <, i.e. boolean combinations of equalities and
  3219 inequalities between polynomials.  It makes use of external
  3220 semidefinite programming solvers.  Method "sos" generates a
  3221 certificate that can be pasted into the proof thus avoiding the need
  3222 to call an external tool every time the proof is checked.  See
  3223 src/HOL/Library/Sum_Of_Squares.
  3224 
  3225 * New method "linarith" invokes existing linear arithmetic decision
  3226 procedure only.
  3227 
  3228 * New command 'atp_minimal' reduces result produced by Sledgehammer.
  3229 
  3230 * New Sledgehammer option "Full Types" in Proof General settings menu.
  3231 Causes full type information to be output to the ATPs.  This slows
  3232 ATPs down considerably but eliminates a source of unsound "proofs"
  3233 that fail later.
  3234 
  3235 * New method "metisFT": A version of metis that uses full type
  3236 information in order to avoid failures of proof reconstruction.
  3237 
  3238 * New evaluator "approximate" approximates an real valued term using
  3239 the same method as the approximation method.
  3240 
  3241 * Method "approximate" now supports arithmetic expressions as
  3242 boundaries of intervals and implements interval splitting and Taylor
  3243 series expansion.
  3244 
  3245 * ML antiquotation @{code_datatype} inserts definition of a datatype
  3246 generated by the code generator; e.g. see src/HOL/Predicate.thy.
  3247 
  3248 * New theory SupInf of the supremum and infimum operators for sets of
  3249 reals.
  3250 
  3251 * New theory Probability, which contains a development of measure
  3252 theory, eventually leading to Lebesgue integration and probability.
  3253 
  3254 * Extended Multivariate Analysis to include derivation and Brouwer's
  3255 fixpoint theorem.
  3256 
  3257 * Reorganization of number theory, INCOMPATIBILITY:
  3258   - new number theory development for nat and int, in theories Divides
  3259     and GCD as well as in new session Number_Theory
  3260   - some constants and facts now suffixed with _nat and _int
  3261     accordingly
  3262   - former session NumberTheory now named Old_Number_Theory, including
  3263     theories Legacy_GCD and Primes (prefer Number_Theory if possible)
  3264   - moved theory Pocklington from src/HOL/Library to
  3265     src/HOL/Old_Number_Theory
  3266 
  3267 * Theory GCD includes functions Gcd/GCD and Lcm/LCM for the gcd and
  3268 lcm of finite and infinite sets. It is shown that they form a complete
  3269 lattice.
  3270 
  3271 * Class semiring_div requires superclass no_zero_divisors and proof of
  3272 div_mult_mult1; theorems div_mult_mult1, div_mult_mult2,
  3273 div_mult_mult1_if, div_mult_mult1 and div_mult_mult2 have been
  3274 generalized to class semiring_div, subsuming former theorems
  3275 zdiv_zmult_zmult1, zdiv_zmult_zmult1_if, zdiv_zmult_zmult1 and
  3276 zdiv_zmult_zmult2.  div_mult_mult1 is now [simp] by default.
  3277 INCOMPATIBILITY.
  3278 
  3279 * Refinements to lattice classes and sets:
  3280   - less default intro/elim rules in locale variant, more default
  3281     intro/elim rules in class variant: more uniformity
  3282   - lemma ge_sup_conv renamed to le_sup_iff, in accordance with
  3283     le_inf_iff
  3284   - dropped lemma alias inf_ACI for inf_aci (same for sup_ACI and
  3285     sup_aci)
  3286   - renamed ACI to inf_sup_aci
  3287   - new class "boolean_algebra"
  3288   - class "complete_lattice" moved to separate theory
  3289     "Complete_Lattice"; corresponding constants (and abbreviations)
  3290     renamed and with authentic syntax:
  3291     Set.Inf ~>    Complete_Lattice.Inf
  3292     Set.Sup ~>    Complete_Lattice.Sup
  3293     Set.INFI ~>   Complete_Lattice.INFI
  3294     Set.SUPR ~>   Complete_Lattice.SUPR
  3295     Set.Inter ~>  Complete_Lattice.Inter
  3296     Set.Union ~>  Complete_Lattice.Union
  3297     Set.INTER ~>  Complete_Lattice.INTER
  3298     Set.UNION ~>  Complete_Lattice.UNION
  3299   - authentic syntax for
  3300     Set.Pow
  3301     Set.image
  3302   - mere abbreviations:
  3303     Set.empty               (for bot)
  3304     Set.UNIV                (for top)
  3305     Set.inter               (for inf, formerly Set.Int)
  3306     Set.union               (for sup, formerly Set.Un)
  3307     Complete_Lattice.Inter  (for Inf)
  3308     Complete_Lattice.Union  (for Sup)
  3309     Complete_Lattice.INTER  (for INFI)
  3310     Complete_Lattice.UNION  (for SUPR)
  3311   - object-logic definitions as far as appropriate
  3312 
  3313 INCOMPATIBILITY.  Care is required when theorems Int_subset_iff or
  3314 Un_subset_iff are explicitly deleted as default simp rules; then also
  3315 their lattice counterparts le_inf_iff and le_sup_iff have to be
  3316 deleted to achieve the desired effect.
  3317 
  3318 * Rules inf_absorb1, inf_absorb2, sup_absorb1, sup_absorb2 are no simp
  3319 rules by default any longer; the same applies to min_max.inf_absorb1
  3320 etc.  INCOMPATIBILITY.
  3321 
  3322 * Rules sup_Int_eq and sup_Un_eq are no longer declared as
  3323 pred_set_conv by default.  INCOMPATIBILITY.
  3324 
  3325 * Power operations on relations and functions are now one dedicated
  3326 constant "compow" with infix syntax "^^".  Power operation on
  3327 multiplicative monoids retains syntax "^" and is now defined generic
  3328 in class power.  INCOMPATIBILITY.
  3329 
  3330 * Relation composition "R O S" now has a more standard argument order:
  3331 "R O S = {(x, z). EX y. (x, y) : R & (y, z) : S}".  INCOMPATIBILITY,
  3332 rewrite propositions with "S O R" --> "R O S". Proofs may occasionally
  3333 break, since the O_assoc rule was not rewritten like this.  Fix using
  3334 O_assoc[symmetric].  The same applies to the curried version "R OO S".
  3335 
  3336 * Function "Inv" is renamed to "inv_into" and function "inv" is now an
  3337 abbreviation for "inv_into UNIV".  Lemmas are renamed accordingly.
  3338 INCOMPATIBILITY.
  3339 
  3340 * Most rules produced by inductive and datatype package have mandatory
  3341 prefixes.  INCOMPATIBILITY.
  3342 
  3343 * Changed "DERIV_intros" to a dynamic fact, which can be augmented by
  3344 the attribute of the same name.  Each of the theorems in the list
  3345 DERIV_intros assumes composition with an additional function and
  3346 matches a variable to the derivative, which has to be solved by the
  3347 Simplifier.  Hence (auto intro!: DERIV_intros) computes the derivative
  3348 of most elementary terms.  Former Maclauren.DERIV_tac and
  3349 Maclauren.deriv_tac should be replaced by (auto intro!: DERIV_intros).
  3350 INCOMPATIBILITY.
  3351 
  3352 * Code generator attributes follow the usual underscore convention:
  3353     code_unfold     replaces    code unfold
  3354     code_post       replaces    code post
  3355     etc.
  3356   INCOMPATIBILITY.
  3357 
  3358 * Renamed methods:
  3359     sizechange -> size_change
  3360     induct_scheme -> induction_schema
  3361   INCOMPATIBILITY.
  3362 
  3363 * Discontinued abbreviation "arbitrary" of constant "undefined".
  3364 INCOMPATIBILITY, use "undefined" directly.
  3365 
  3366 * Renamed theorems:
  3367     Suc_eq_add_numeral_1 -> Suc_eq_plus1
  3368     Suc_eq_add_numeral_1_left -> Suc_eq_plus1_left
  3369     Suc_plus1 -> Suc_eq_plus1
  3370     *anti_sym -> *antisym*
  3371     vector_less_eq_def -> vector_le_def
  3372   INCOMPATIBILITY.
  3373 
  3374 * Added theorem List.map_map as [simp].  Removed List.map_compose.
  3375 INCOMPATIBILITY.
  3376 
  3377 * Removed predicate "M hassize n" (<--> card M = n & finite M).
  3378 INCOMPATIBILITY.
  3379 
  3380 
  3381 *** HOLCF ***
  3382 
  3383 * Theory Representable defines a class "rep" of domains that are
  3384 representable (via an ep-pair) in the universal domain type "udom".
  3385 Instances are provided for all type constructors defined in HOLCF.
  3386 
  3387 * The 'new_domain' command is a purely definitional version of the
  3388 domain package, for representable domains.  Syntax is identical to the
  3389 old domain package.  The 'new_domain' package also supports indirect
  3390 recursion using previously-defined type constructors.  See
  3391 src/HOLCF/ex/New_Domain.thy for examples.
  3392 
  3393 * Method "fixrec_simp" unfolds one step of a fixrec-defined constant
  3394 on the left-hand side of an equation, and then performs
  3395 simplification.  Rewriting is done using rules declared with the
  3396 "fixrec_simp" attribute.  The "fixrec_simp" method is intended as a
  3397 replacement for "fixpat"; see src/HOLCF/ex/Fixrec_ex.thy for examples.
  3398 
  3399 * The pattern-match compiler in 'fixrec' can now handle constructors
  3400 with HOL function types.  Pattern-match combinators for the Pair
  3401 constructor are pre-configured.
  3402 
  3403 * The 'fixrec' package now produces better fixed-point induction rules
  3404 for mutually-recursive definitions:  Induction rules have conclusions
  3405 of the form "P foo bar" instead of "P <foo, bar>".
  3406 
  3407 * The constant "sq_le" (with infix syntax "<<" or "\<sqsubseteq>") has
  3408 been renamed to "below".  The name "below" now replaces "less" in many
  3409 theorem names.  (Legacy theorem names using "less" are still supported
  3410 as well.)
  3411 
  3412 * The 'fixrec' package now supports "bottom patterns".  Bottom
  3413 patterns can be used to generate strictness rules, or to make
  3414 functions more strict (much like the bang-patterns supported by the
  3415 Glasgow Haskell Compiler).  See src/HOLCF/ex/Fixrec_ex.thy for
  3416 examples.
  3417 
  3418 
  3419 *** ML ***
  3420 
  3421 * Support for Poly/ML 5.3.0, with improved reporting of compiler
  3422 errors and run-time exceptions, including detailed source positions.
  3423 
  3424 * Structure Name_Space (formerly NameSpace) now manages uniquely
  3425 identified entries, with some additional information such as source
  3426 position, logical grouping etc.
  3427 
  3428 * Theory and context data is now introduced by the simplified and
  3429 modernized functors Theory_Data, Proof_Data, Generic_Data.  Data needs
  3430 to be pure, but the old TheoryDataFun for mutable data (with explicit
  3431 copy operation) is still available for some time.
  3432 
  3433 * Structure Synchronized (cf. src/Pure/Concurrent/synchronized.ML)
  3434 provides a high-level programming interface to synchronized state
  3435 variables with atomic update.  This works via pure function
  3436 application within a critical section -- its runtime should be as
  3437 short as possible; beware of deadlocks if critical code is nested,
  3438 either directly or indirectly via other synchronized variables!
  3439 
  3440 * Structure Unsynchronized (cf. src/Pure/ML-Systems/unsynchronized.ML)
  3441 wraps raw ML references, explicitly indicating their non-thread-safe
  3442 behaviour.  The Isar toplevel keeps this structure open, to
  3443 accommodate Proof General as well as quick and dirty interactive
  3444 experiments with references.
  3445 
  3446 * PARALLEL_CHOICE and PARALLEL_GOALS provide basic support for
  3447 parallel tactical reasoning.
  3448 
  3449 * Tacticals Subgoal.FOCUS, Subgoal.FOCUS_PREMS, Subgoal.FOCUS_PARAMS
  3450 are similar to SUBPROOF, but are slightly more flexible: only the
  3451 specified parts of the subgoal are imported into the context, and the
  3452 body tactic may introduce new subgoals and schematic variables.
  3453 
  3454 * Old tactical METAHYPS, which does not observe the proof context, has
  3455 been renamed to Old_Goals.METAHYPS and awaits deletion.  Use SUBPROOF
  3456 or Subgoal.FOCUS etc.
  3457 
  3458 * Renamed functor TableFun to Table, and GraphFun to Graph.  (Since
  3459 functors have their own ML name space there is no point to mark them
  3460 separately.)  Minor INCOMPATIBILITY.
  3461 
  3462 * Renamed NamedThmsFun to Named_Thms.  INCOMPATIBILITY.
  3463 
  3464 * Renamed several structures FooBar to Foo_Bar.  Occasional,
  3465 INCOMPATIBILITY.
  3466 
  3467 * Operations of structure Skip_Proof no longer require quick_and_dirty
  3468 mode, which avoids critical setmp.
  3469 
  3470 * Eliminated old Attrib.add_attributes, Method.add_methods and related
  3471 combinators for "args".  INCOMPATIBILITY, need to use simplified
  3472 Attrib/Method.setup introduced in Isabelle2009.
  3473 
  3474 * Proper context for simpset_of, claset_of, clasimpset_of.  May fall
  3475 back on global_simpset_of, global_claset_of, global_clasimpset_of as
  3476 last resort.  INCOMPATIBILITY.
  3477 
  3478 * Display.pretty_thm now requires a proper context (cf. former
  3479 ProofContext.pretty_thm).  May fall back on Display.pretty_thm_global
  3480 or even Display.pretty_thm_without_context as last resort.
  3481 INCOMPATIBILITY.
  3482 
  3483 * Discontinued Display.pretty_ctyp/cterm etc.  INCOMPATIBILITY, use
  3484 Syntax.pretty_typ/term directly, preferably with proper context
  3485 instead of global theory.
  3486 
  3487 
  3488 *** System ***
  3489 
  3490 * Further fine tuning of parallel proof checking, scales up to 8 cores
  3491 (max. speedup factor 5.0).  See also Goal.parallel_proofs in ML and
  3492 usedir option -q.
  3493 
  3494 * Support for additional "Isabelle components" via etc/components, see
  3495 also the system manual.
  3496 
  3497 * The isabelle makeall tool now operates on all components with
  3498 IsaMakefile, not just hardwired "logics".
  3499 
  3500 * Removed "compress" option from isabelle-process and isabelle usedir;
  3501 this is always enabled.
  3502 
  3503 * Discontinued support for Poly/ML 4.x versions.
  3504 
  3505 * Isabelle tool "wwwfind" provides web interface for 'find_theorems'
  3506 on a given logic image.  This requires the lighttpd webserver and is
  3507 currently supported on Linux only.
  3508 
  3509 
  3510 
  3511 New in Isabelle2009 (April 2009)
  3512 --------------------------------
  3513 
  3514 *** General ***
  3515 
  3516 * Simplified main Isabelle executables, with less surprises on
  3517 case-insensitive file-systems (such as Mac OS).
  3518 
  3519   - The main Isabelle tool wrapper is now called "isabelle" instead of
  3520     "isatool."
  3521 
  3522   - The former "isabelle" alias for "isabelle-process" has been
  3523     removed (should rarely occur to regular users).
  3524 
  3525   - The former "isabelle-interface" and its alias "Isabelle" have been
  3526     removed (interfaces are now regular Isabelle tools).
  3527 
  3528 Within scripts and make files, the Isabelle environment variables
  3529 ISABELLE_TOOL and ISABELLE_PROCESS replace old ISATOOL and ISABELLE,
  3530 respectively.  (The latter are still available as legacy feature.)
  3531 
  3532 The old isabelle-interface wrapper could react in confusing ways if
  3533 the interface was uninstalled or changed otherwise.  Individual
  3534 interface tool configuration is now more explicit, see also the
  3535 Isabelle system manual.  In particular, Proof General is now available
  3536 via "isabelle emacs".
  3537 
  3538 INCOMPATIBILITY, need to adapt derivative scripts.  Users may need to
  3539 purge installed copies of Isabelle executables and re-run "isabelle
  3540 install -p ...", or use symlinks.
  3541 
  3542 * The default for ISABELLE_HOME_USER is now ~/.isabelle instead of the
  3543 old ~/isabelle, which was slightly non-standard and apt to cause
  3544 surprises on case-insensitive file-systems (such as Mac OS).
  3545 
  3546 INCOMPATIBILITY, need to move existing ~/isabelle/etc,
  3547 ~/isabelle/heaps, ~/isabelle/browser_info to the new place.  Special
  3548 care is required when using older releases of Isabelle.  Note that
  3549 ISABELLE_HOME_USER can be changed in Isabelle/etc/settings of any
  3550 Isabelle distribution, in order to use the new ~/.isabelle uniformly.
  3551 
  3552 * Proofs of fully specified statements are run in parallel on
  3553 multi-core systems.  A speedup factor of 2.5 to 3.2 can be expected on
  3554 a regular 4-core machine, if the initial heap space is made reasonably
  3555 large (cf. Poly/ML option -H).  (Requires Poly/ML 5.2.1 or later.)
  3556 
  3557 * The main reference manuals ("isar-ref", "implementation", and
  3558 "system") have been updated and extended.  Formally checked references
  3559 as hyperlinks are now available uniformly.
  3560 
  3561 
  3562 *** Pure ***
  3563 
  3564 * Complete re-implementation of locales.  INCOMPATIBILITY in several
  3565 respects.  The most important changes are listed below.  See the
  3566 Tutorial on Locales ("locales" manual) for details.
  3567 
  3568 - In locale expressions, instantiation replaces renaming.  Parameters
  3569 must be declared in a for clause.  To aid compatibility with previous
  3570 parameter inheritance, in locale declarations, parameters that are not
  3571 'touched' (instantiation position "_" or omitted) are implicitly added
  3572 with their syntax at the beginning of the for clause.
  3573 
  3574 - Syntax from abbreviations and definitions in locales is available in
  3575 locale expressions and context elements.  The latter is particularly
  3576 useful in locale declarations.
  3577 
  3578 - More flexible mechanisms to qualify names generated by locale
  3579 expressions.  Qualifiers (prefixes) may be specified in locale
  3580 expressions, and can be marked as mandatory (syntax: "name!:") or
  3581 optional (syntax "name?:").  The default depends for plain "name:"
  3582 depends on the situation where a locale expression is used: in
  3583 commands 'locale' and 'sublocale' prefixes are optional, in
  3584 'interpretation' and 'interpret' prefixes are mandatory.  The old
  3585 implicit qualifiers derived from the parameter names of a locale are
  3586 no longer generated.
  3587 
  3588 - Command "sublocale l < e" replaces "interpretation l < e".  The
  3589 instantiation clause in "interpretation" and "interpret" (square
  3590 brackets) is no longer available.  Use locale expressions.
  3591 
  3592 - When converting proof scripts, mandatory qualifiers in
  3593 'interpretation' and 'interpret' should be retained by default, even
  3594 if this is an INCOMPATIBILITY compared to former behavior.  In the
  3595 worst case, use the "name?:" form for non-mandatory ones.  Qualifiers
  3596 in locale expressions range over a single locale instance only.
  3597 
  3598 - Dropped locale element "includes".  This is a major INCOMPATIBILITY.
  3599 In existing theorem specifications replace the includes element by the
  3600 respective context elements of the included locale, omitting those
  3601 that are already present in the theorem specification.  Multiple
  3602 assume elements of a locale should be replaced by a single one
  3603 involving the locale predicate.  In the proof body, declarations (most
  3604 notably theorems) may be regained by interpreting the respective
  3605 locales in the proof context as required (command "interpret").
  3606 
  3607 If using "includes" in replacement of a target solely because the
  3608 parameter types in the theorem are not as general as in the target,
  3609 consider declaring a new locale with additional type constraints on
  3610 the parameters (context element "constrains").
  3611 
  3612 - Discontinued "locale (open)".  INCOMPATIBILITY.
  3613 
  3614 - Locale interpretation commands no longer attempt to simplify goal.
  3615 INCOMPATIBILITY: in rare situations the generated goal differs.  Use
  3616 methods intro_locales and unfold_locales to clarify.
  3617 
  3618 - Locale interpretation commands no longer accept interpretation
  3619 attributes.  INCOMPATIBILITY.
  3620 
  3621 * Class declaration: so-called "base sort" must not be given in import
  3622 list any longer, but is inferred from the specification.  Particularly
  3623 in HOL, write
  3624 
  3625     class foo = ...
  3626 
  3627 instead of
  3628 
  3629     class foo = type + ...
  3630 
  3631 * Class target: global versions of theorems stemming do not carry a
  3632 parameter prefix any longer.  INCOMPATIBILITY.
  3633 
  3634 * Class 'instance' command no longer accepts attached definitions.
  3635 INCOMPATIBILITY, use proper 'instantiation' target instead.
  3636 
  3637 * Recovered hiding of consts, which was accidentally broken in
  3638 Isabelle2007.  Potential INCOMPATIBILITY, ``hide const c'' really
  3639 makes c inaccessible; consider using ``hide (open) const c'' instead.
  3640 
  3641 * Slightly more coherent Pure syntax, with updated documentation in
  3642 isar-ref manual.  Removed locales meta_term_syntax and
  3643 meta_conjunction_syntax: TERM and &&& (formerly &&) are now permanent,
  3644 INCOMPATIBILITY in rare situations.  Note that &&& should not be used
  3645 directly in regular applications.
  3646 
  3647 * There is a new syntactic category "float_const" for signed decimal
  3648 fractions (e.g. 123.45 or -123.45).
  3649 
  3650 * Removed exotic 'token_translation' command.  INCOMPATIBILITY, use ML
  3651 interface with 'setup' command instead.
  3652 
  3653 * Command 'local_setup' is similar to 'setup', but operates on a local
  3654 theory context.
  3655 
  3656 * The 'axiomatization' command now only works within a global theory
  3657 context.  INCOMPATIBILITY.
  3658 
  3659 * Goal-directed proof now enforces strict proof irrelevance wrt. sort
  3660 hypotheses.  Sorts required in the course of reasoning need to be
  3661 covered by the constraints in the initial statement, completed by the
  3662 type instance information of the background theory.  Non-trivial sort
  3663 hypotheses, which rarely occur in practice, may be specified via
  3664 vacuous propositions of the form SORT_CONSTRAINT('a::c).  For example:
  3665 
  3666   lemma assumes "SORT_CONSTRAINT('a::empty)" shows False ...
  3667 
  3668 The result contains an implicit sort hypotheses as before --
  3669 SORT_CONSTRAINT premises are eliminated as part of the canonical rule
  3670 normalization.
  3671 
  3672 * Generalized Isar history, with support for linear undo, direct state
  3673 addressing etc.
  3674 
  3675 * Changed defaults for unify configuration options:
  3676 
  3677   unify_trace_bound = 50 (formerly 25)
  3678   unify_search_bound = 60 (formerly 30)
  3679 
  3680 * Different bookkeeping for code equations (INCOMPATIBILITY):
  3681 
  3682   a) On theory merge, the last set of code equations for a particular
  3683      constant is taken (in accordance with the policy applied by other
  3684      parts of the code generator framework).
  3685 
  3686   b) Code equations stemming from explicit declarations (e.g. code
  3687      attribute) gain priority over default code equations stemming
  3688      from definition, primrec, fun etc.
  3689 
  3690 * Keyword 'code_exception' now named 'code_abort'.  INCOMPATIBILITY.
  3691 
  3692 * Unified theorem tables for both code generators.  Thus [code
  3693 func] has disappeared and only [code] remains.  INCOMPATIBILITY.
  3694 
  3695 * Command 'find_consts' searches for constants based on type and name
  3696 patterns, e.g.
  3697 
  3698     find_consts "_ => bool"
  3699 
  3700 By default, matching is against subtypes, but it may be restricted to
  3701 the whole type.  Searching by name is possible.  Multiple queries are
  3702 conjunctive and queries may be negated by prefixing them with a
  3703 hyphen:
  3704 
  3705     find_consts strict: "_ => bool" name: "Int" -"int => int"
  3706 
  3707 * New 'find_theorems' criterion "solves" matches theorems that
  3708 directly solve the current goal (modulo higher-order unification).
  3709 
  3710 * Auto solve feature for main theorem statements: whenever a new goal
  3711 is stated, "find_theorems solves" is called; any theorems that could
  3712 solve the lemma directly are listed as part of the goal state.
  3713 Cf. associated options in Proof General Isabelle settings menu,
  3714 enabled by default, with reasonable timeout for pathological cases of
  3715 higher-order unification.
  3716 
  3717 
  3718 *** Document preparation ***
  3719 
  3720 * Antiquotation @{lemma} now imitates a regular terminal proof,
  3721 demanding keyword 'by' and supporting the full method expression
  3722 syntax just like the Isar command 'by'.
  3723 
  3724 
  3725 *** HOL ***
  3726 
  3727 * Integrated main parts of former image HOL-Complex with HOL.  Entry
  3728 points Main and Complex_Main remain as before.
  3729 
  3730 * Logic image HOL-Plain provides a minimal HOL with the most important
  3731 tools available (inductive, datatype, primrec, ...).  This facilitates
  3732 experimentation and tool development.  Note that user applications
  3733 (and library theories) should never refer to anything below theory
  3734 Main, as before.
  3735 
  3736 * Logic image HOL-Main stops at theory Main, and thus facilitates
  3737 experimentation due to shorter build times.
  3738 
  3739 * Logic image HOL-NSA contains theories of nonstandard analysis which
  3740 were previously part of former HOL-Complex.  Entry point Hyperreal
  3741 remains valid, but theories formerly using Complex_Main should now use
  3742 new entry point Hypercomplex.
  3743 
  3744 * Generic ATP manager for Sledgehammer, based on ML threads instead of
  3745 Posix processes.  Avoids potentially expensive forking of the ML
  3746 process.  New thread-based implementation also works on non-Unix
  3747 platforms (Cygwin).  Provers are no longer hardwired, but defined
  3748 within the theory via plain ML wrapper functions.  Basic Sledgehammer
  3749 commands are covered in the isar-ref manual.
  3750 
  3751 * Wrapper scripts for remote SystemOnTPTP service allows to use
  3752 sledgehammer without local ATP installation (Vampire etc.). Other
  3753 provers may be included via suitable ML wrappers, see also
  3754 src/HOL/ATP_Linkup.thy.
  3755 
  3756 * ATP selection (E/Vampire/Spass) is now via Proof General's settings
  3757 menu.
  3758 
  3759 * The metis method no longer fails because the theorem is too trivial
  3760 (contains the empty clause).
  3761 
  3762 * The metis method now fails in the usual manner, rather than raising
  3763 an exception, if it determines that it cannot prove the theorem.
  3764 
  3765 * Method "coherent" implements a prover for coherent logic (see also
  3766 src/Tools/coherent.ML).
  3767 
  3768 * Constants "undefined" and "default" replace "arbitrary".  Usually
  3769 "undefined" is the right choice to replace "arbitrary", though
  3770 logically there is no difference.  INCOMPATIBILITY.
  3771 
  3772 * Command "value" now integrates different evaluation mechanisms.  The
  3773 result of the first successful evaluation mechanism is printed.  In
  3774 square brackets a particular named evaluation mechanisms may be
  3775 specified (currently, [SML], [code] or [nbe]).  See further
  3776 src/HOL/ex/Eval_Examples.thy.
  3777 
  3778 * Normalization by evaluation now allows non-leftlinear equations.
  3779 Declare with attribute [code nbe].
  3780 
  3781 * Methods "case_tac" and "induct_tac" now refer to the very same rules
  3782 as the structured Isar versions "cases" and "induct", cf. the
  3783 corresponding "cases" and "induct" attributes.  Mutual induction rules
  3784 are now presented as a list of individual projections
  3785 (e.g. foo_bar.inducts for types foo and bar); the old format with
  3786 explicit HOL conjunction is no longer supported.  INCOMPATIBILITY, in
  3787 rare situations a different rule is selected --- notably nested tuple
  3788 elimination instead of former prod.exhaust: use explicit (case_tac t
  3789 rule: prod.exhaust) here.
  3790 
  3791 * Attributes "cases", "induct", "coinduct" support "del" option.
  3792 
  3793 * Removed fact "case_split_thm", which duplicates "case_split".
  3794 
  3795 * The option datatype has been moved to a new theory Option.  Renamed
  3796 option_map to Option.map, and o2s to Option.set, INCOMPATIBILITY.
  3797 
  3798 * New predicate "strict_mono" classifies strict functions on partial
  3799 orders.  With strict functions on linear orders, reasoning about
  3800 (in)equalities is facilitated by theorems "strict_mono_eq",
  3801 "strict_mono_less_eq" and "strict_mono_less".
  3802 
  3803 * Some set operations are now proper qualified constants with
  3804 authentic syntax.  INCOMPATIBILITY:
  3805 
  3806     op Int ~>   Set.Int
  3807     op Un ~>    Set.Un
  3808     INTER ~>    Set.INTER
  3809     UNION ~>    Set.UNION
  3810     Inter ~>    Set.Inter
  3811     Union ~>    Set.Union
  3812     {} ~>       Set.empty
  3813     UNIV ~>     Set.UNIV
  3814 
  3815 * Class complete_lattice with operations Inf, Sup, INFI, SUPR now in
  3816 theory Set.
  3817 
  3818 * Auxiliary class "itself" has disappeared -- classes without any
  3819 parameter are treated as expected by the 'class' command.
  3820 
  3821 * Leibnitz's Series for Pi and the arcus tangens and logarithm series.
  3822 
  3823 * Common decision procedures (Cooper, MIR, Ferrack, Approximation,
  3824 Dense_Linear_Order) are now in directory HOL/Decision_Procs.
  3825 
  3826 * Theory src/HOL/Decision_Procs/Approximation provides the new proof
  3827 method "approximation".  It proves formulas on real values by using
  3828 interval arithmetic.  In the formulas are also the transcendental
  3829 functions sin, cos, tan, atan, ln, exp and the constant pi are
  3830 allowed. For examples see
  3831 src/HOL/Descision_Procs/ex/Approximation_Ex.thy.
  3832 
  3833 * Theory "Reflection" now resides in HOL/Library.
  3834 
  3835 * Entry point to Word library now simply named "Word".
  3836 INCOMPATIBILITY.
  3837 
  3838 * Made source layout more coherent with logical distribution
  3839 structure:
  3840 
  3841     src/HOL/Library/RType.thy ~> src/HOL/Typerep.thy
  3842     src/HOL/Library/Code_Message.thy ~> src/HOL/
  3843     src/HOL/Library/GCD.thy ~> src/HOL/
  3844     src/HOL/Library/Order_Relation.thy ~> src/HOL/
  3845     src/HOL/Library/Parity.thy ~> src/HOL/
  3846     src/HOL/Library/Univ_Poly.thy ~> src/HOL/
  3847     src/HOL/Real/ContNotDenum.thy ~> src/HOL/Library/
  3848     src/HOL/Real/Lubs.thy ~> src/HOL/
  3849     src/HOL/Real/PReal.thy ~> src/HOL/
  3850     src/HOL/Real/Rational.thy ~> src/HOL/
  3851     src/HOL/Real/RComplete.thy ~> src/HOL/
  3852     src/HOL/Real/RealDef.thy ~> src/HOL/
  3853     src/HOL/Real/RealPow.thy ~> src/HOL/
  3854     src/HOL/Real/Real.thy ~> src/HOL/
  3855     src/HOL/Complex/Complex_Main.thy ~> src/HOL/
  3856     src/HOL/Complex/Complex.thy ~> src/HOL/
  3857     src/HOL/Complex/FrechetDeriv.thy ~> src/HOL/Library/
  3858     src/HOL/Complex/Fundamental_Theorem_Algebra.thy ~> src/HOL/Library/
  3859     src/HOL/Hyperreal/Deriv.thy ~> src/HOL/
  3860     src/HOL/Hyperreal/Fact.thy ~> src/HOL/
  3861     src/HOL/Hyperreal/Integration.thy ~> src/HOL/
  3862     src/HOL/Hyperreal/Lim.thy ~> src/HOL/
  3863     src/HOL/Hyperreal/Ln.thy ~> src/HOL/
  3864     src/HOL/Hyperreal/Log.thy ~> src/HOL/
  3865     src/HOL/Hyperreal/MacLaurin.thy ~> src/HOL/
  3866     src/HOL/Hyperreal/NthRoot.thy ~> src/HOL/
  3867     src/HOL/Hyperreal/Series.thy ~> src/HOL/
  3868     src/HOL/Hyperreal/SEQ.thy ~> src/HOL/
  3869     src/HOL/Hyperreal/Taylor.thy ~> src/HOL/
  3870     src/HOL/Hyperreal/Transcendental.thy ~> src/HOL/
  3871     src/HOL/Real/Float ~> src/HOL/Library/
  3872     src/HOL/Real/HahnBanach ~> src/HOL/HahnBanach
  3873     src/HOL/Real/RealVector.thy ~> src/HOL/
  3874 
  3875     src/HOL/arith_data.ML ~> src/HOL/Tools
  3876     src/HOL/hologic.ML ~> src/HOL/Tools
  3877     src/HOL/simpdata.ML ~> src/HOL/Tools
  3878     src/HOL/int_arith1.ML ~> src/HOL/Tools/int_arith.ML
  3879     src/HOL/int_factor_simprocs.ML ~> src/HOL/Tools
  3880     src/HOL/nat_simprocs.ML ~> src/HOL/Tools
  3881     src/HOL/Real/float_arith.ML ~> src/HOL/Tools
  3882     src/HOL/Real/float_syntax.ML ~> src/HOL/Tools
  3883     src/HOL/Real/rat_arith.ML ~> src/HOL/Tools
  3884     src/HOL/Real/real_arith.ML ~> src/HOL/Tools
  3885 
  3886     src/HOL/Library/Array.thy ~> src/HOL/Imperative_HOL
  3887     src/HOL/Library/Heap_Monad.thy ~> src/HOL/Imperative_HOL
  3888     src/HOL/Library/Heap.thy ~> src/HOL/Imperative_HOL
  3889     src/HOL/Library/Imperative_HOL.thy ~> src/HOL/Imperative_HOL
  3890     src/HOL/Library/Ref.thy ~> src/HOL/Imperative_HOL
  3891     src/HOL/Library/Relational.thy ~> src/HOL/Imperative_HOL
  3892 
  3893 * If methods "eval" and "evaluation" encounter a structured proof
  3894 state with !!/==>, only the conclusion is evaluated to True (if
  3895 possible), avoiding strange error messages.
  3896 
  3897 * Method "sizechange" automates termination proofs using (a
  3898 modification of) the size-change principle.  Requires SAT solver.  See
  3899 src/HOL/ex/Termination.thy for examples.
  3900 
  3901 * Simplifier: simproc for let expressions now unfolds if bound
  3902 variable occurs at most once in let expression body.  INCOMPATIBILITY.
  3903 
  3904 * Method "arith": Linear arithmetic now ignores all inequalities when
  3905 fast_arith_neq_limit is exceeded, instead of giving up entirely.
  3906 
  3907 * New attribute "arith" for facts that should always be used
  3908 automatically by arithmetic. It is intended to be used locally in
  3909 proofs, e.g.
  3910 
  3911   assumes [arith]: "x > 0"
  3912 
  3913 Global usage is discouraged because of possible performance impact.
  3914 
  3915 * New classes "top" and "bot" with corresponding operations "top" and
  3916 "bot" in theory Orderings; instantiation of class "complete_lattice"
  3917 requires instantiation of classes "top" and "bot".  INCOMPATIBILITY.
  3918 
  3919 * Changed definition lemma "less_fun_def" in order to provide an
  3920 instance for preorders on functions; use lemma "less_le" instead.
  3921 INCOMPATIBILITY.
  3922 
  3923 * Theory Orderings: class "wellorder" moved here, with explicit
  3924 induction rule "less_induct" as assumption.  For instantiation of
  3925 "wellorder" by means of predicate "wf", use rule wf_wellorderI.
  3926 INCOMPATIBILITY.
  3927 
  3928 * Theory Orderings: added class "preorder" as superclass of "order".
  3929 INCOMPATIBILITY: Instantiation proofs for order, linorder
  3930 etc. slightly changed.  Some theorems named order_class.* now named
  3931 preorder_class.*.
  3932 
  3933 * Theory Relation: renamed "refl" to "refl_on", "reflexive" to "refl,
  3934 "diag" to "Id_on".
  3935 
  3936 * Theory Finite_Set: added a new fold combinator of type
  3937 
  3938   ('a => 'b => 'b) => 'b => 'a set => 'b
  3939 
  3940 Occasionally this is more convenient than the old fold combinator
  3941 which is now defined in terms of the new one and renamed to
  3942 fold_image.
  3943 
  3944 * Theories Ring_and_Field and OrderedGroup: The lemmas "group_simps"
  3945 and "ring_simps" have been replaced by "algebra_simps" (which can be
  3946 extended with further lemmas!).  At the moment both still exist but
  3947 the former will disappear at some point.
  3948 
  3949 * Theory Power: Lemma power_Suc is now declared as a simp rule in
  3950 class recpower.  Type-specific simp rules for various recpower types
  3951 have been removed.  INCOMPATIBILITY, rename old lemmas as follows:
  3952 
  3953 rat_power_0    -> power_0
  3954 rat_power_Suc  -> power_Suc
  3955 realpow_0      -> power_0
  3956 realpow_Suc    -> power_Suc
  3957 complexpow_0   -> power_0
  3958 complexpow_Suc -> power_Suc
  3959 power_poly_0   -> power_0
  3960 power_poly_Suc -> power_Suc
  3961 
  3962 * Theories Ring_and_Field and Divides: Definition of "op dvd" has been
  3963 moved to separate class dvd in Ring_and_Field; a couple of lemmas on
  3964 dvd has been generalized to class comm_semiring_1.  Likewise a bunch
  3965 of lemmas from Divides has been generalized from nat to class
  3966 semiring_div.  INCOMPATIBILITY.  This involves the following theorem
  3967 renames resulting from duplicate elimination:
  3968 
  3969     dvd_def_mod ~>          dvd_eq_mod_eq_0
  3970     zero_dvd_iff ~>         dvd_0_left_iff
  3971     dvd_0 ~>                dvd_0_right
  3972     DIVISION_BY_ZERO_DIV ~> div_by_0
  3973     DIVISION_BY_ZERO_MOD ~> mod_by_0
  3974     mult_div ~>             div_mult_self2_is_id
  3975     mult_mod ~>             mod_mult_self2_is_0
  3976 
  3977 * Theory IntDiv: removed many lemmas that are instances of class-based
  3978 generalizations (from Divides and Ring_and_Field).  INCOMPATIBILITY,
  3979 rename old lemmas as follows:
  3980 
  3981 dvd_diff               -> nat_dvd_diff
  3982 dvd_zminus_iff         -> dvd_minus_iff
  3983 mod_add1_eq            -> mod_add_eq
  3984 mod_mult1_eq           -> mod_mult_right_eq
  3985 mod_mult1_eq'          -> mod_mult_left_eq
  3986 mod_mult_distrib_mod   -> mod_mult_eq
  3987 nat_mod_add_left_eq    -> mod_add_left_eq
  3988 nat_mod_add_right_eq   -> mod_add_right_eq
  3989 nat_mod_div_trivial    -> mod_div_trivial
  3990 nat_mod_mod_trivial    -> mod_mod_trivial
  3991 zdiv_zadd_self1        -> div_add_self1
  3992 zdiv_zadd_self2        -> div_add_self2
  3993 zdiv_zmult_self1       -> div_mult_self2_is_id
  3994 zdiv_zmult_self2       -> div_mult_self1_is_id
  3995 zdvd_triv_left         -> dvd_triv_left
  3996 zdvd_triv_right        -> dvd_triv_right
  3997 zdvd_zmult_cancel_disj -> dvd_mult_cancel_left
  3998 zmod_eq0_zdvd_iff      -> dvd_eq_mod_eq_0[symmetric]
  3999 zmod_zadd_left_eq      -> mod_add_left_eq
  4000 zmod_zadd_right_eq     -> mod_add_right_eq
  4001 zmod_zadd_self1        -> mod_add_self1
  4002 zmod_zadd_self2        -> mod_add_self2
  4003 zmod_zadd1_eq          -> mod_add_eq
  4004 zmod_zdiff1_eq         -> mod_diff_eq
  4005 zmod_zdvd_zmod         -> mod_mod_cancel
  4006 zmod_zmod_cancel       -> mod_mod_cancel
  4007 zmod_zmult_self1       -> mod_mult_self2_is_0
  4008 zmod_zmult_self2       -> mod_mult_self1_is_0
  4009 zmod_1                 -> mod_by_1
  4010 zdiv_1                 -> div_by_1
  4011 zdvd_abs1              -> abs_dvd_iff
  4012 zdvd_abs2              -> dvd_abs_iff
  4013 zdvd_refl              -> dvd_refl
  4014 zdvd_trans             -> dvd_trans
  4015 zdvd_zadd              -> dvd_add
  4016 zdvd_zdiff             -> dvd_diff
  4017 zdvd_zminus_iff        -> dvd_minus_iff
  4018 zdvd_zminus2_iff       -> minus_dvd_iff
  4019 zdvd_zmultD            -> dvd_mult_right
  4020 zdvd_zmultD2           -> dvd_mult_left
  4021 zdvd_zmult_mono        -> mult_dvd_mono
  4022 zdvd_0_right           -> dvd_0_right
  4023 zdvd_0_left            -> dvd_0_left_iff
  4024 zdvd_1_left            -> one_dvd
  4025 zminus_dvd_iff         -> minus_dvd_iff
  4026 
  4027 * Theory Rational: 'Fract k 0' now equals '0'.  INCOMPATIBILITY.
  4028 
  4029 * The real numbers offer decimal input syntax: 12.34 is translated
  4030 into 1234/10^2. This translation is not reversed upon output.
  4031 
  4032 * Theory Library/Polynomial defines an abstract type 'a poly of
  4033 univariate polynomials with coefficients of type 'a.  In addition to
  4034 the standard ring operations, it also supports div and mod.  Code
  4035 generation is also supported, using list-style constructors.
  4036 
  4037 * Theory Library/Inner_Product defines a class of real_inner for real
  4038 inner product spaces, with an overloaded operation inner :: 'a => 'a
  4039 => real.  Class real_inner is a subclass of real_normed_vector from
  4040 theory RealVector.
  4041 
  4042 * Theory Library/Product_Vector provides instances for the product
  4043 type 'a * 'b of several classes from RealVector and Inner_Product.
  4044 Definitions of addition, subtraction, scalar multiplication, norms,
  4045 and inner products are included.
  4046 
  4047 * Theory Library/Bit defines the field "bit" of integers modulo 2.  In
  4048 addition to the field operations, numerals and case syntax are also
  4049 supported.
  4050 
  4051 * Theory Library/Diagonalize provides constructive version of Cantor's
  4052 first diagonalization argument.
  4053 
  4054 * Theory Library/GCD: Curried operations gcd, lcm (for nat) and zgcd,
  4055 zlcm (for int); carried together from various gcd/lcm developements in
  4056 the HOL Distribution.  Constants zgcd and zlcm replace former igcd and
  4057 ilcm; corresponding theorems renamed accordingly.  INCOMPATIBILITY,
  4058 may recover tupled syntax as follows:
  4059 
  4060     hide (open) const gcd
  4061     abbreviation gcd where
  4062       "gcd == (%(a, b). GCD.gcd a b)"
  4063     notation (output)
  4064       GCD.gcd ("gcd '(_, _')")
  4065 
  4066 The same works for lcm, zgcd, zlcm.
  4067 
  4068 * Theory Library/Nat_Infinity: added addition, numeral syntax and more
  4069 instantiations for algebraic structures.  Removed some duplicate
  4070 theorems.  Changes in simp rules.  INCOMPATIBILITY.
  4071 
  4072 * ML antiquotation @{code} takes a constant as argument and generates
  4073 corresponding code in background and inserts name of the corresponding
  4074 resulting ML value/function/datatype constructor binding in place.
  4075 All occurrences of @{code} with a single ML block are generated
  4076 simultaneously.  Provides a generic and safe interface for
  4077 instrumentalizing code generation.  See
  4078 src/HOL/Decision_Procs/Ferrack.thy for a more ambitious application.
  4079 In future you ought to refrain from ad-hoc compiling generated SML
  4080 code on the ML toplevel.  Note that (for technical reasons) @{code}
  4081 cannot refer to constants for which user-defined serializations are
  4082 set.  Refer to the corresponding ML counterpart directly in that
  4083 cases.
  4084 
  4085 * Command 'rep_datatype': instead of theorem names the command now
  4086 takes a list of terms denoting the constructors of the type to be
  4087 represented as datatype.  The characteristic theorems have to be
  4088 proven.  INCOMPATIBILITY.  Also observe that the following theorems
  4089 have disappeared in favour of existing ones:
  4090 
  4091     unit_induct                 ~> unit.induct
  4092     prod_induct                 ~> prod.induct
  4093     sum_induct                  ~> sum.induct
  4094     Suc_Suc_eq                  ~> nat.inject
  4095     Suc_not_Zero Zero_not_Suc   ~> nat.distinct
  4096 
  4097 
  4098 *** HOL-Algebra ***
  4099 
  4100 * New locales for orders and lattices where the equivalence relation
  4101 is not restricted to equality.  INCOMPATIBILITY: all order and lattice
  4102 locales use a record structure with field eq for the equivalence.
  4103 
  4104 * New theory of factorial domains.
  4105 
  4106 * Units_l_inv and Units_r_inv are now simp rules by default.
  4107 INCOMPATIBILITY.  Simplifier proof that require deletion of l_inv
  4108 and/or r_inv will now also require deletion of these lemmas.
  4109 
  4110 * Renamed the following theorems, INCOMPATIBILITY:
  4111 
  4112 UpperD ~> Upper_memD
  4113 LowerD ~> Lower_memD
  4114 least_carrier ~> least_closed
  4115 greatest_carrier ~> greatest_closed
  4116 greatest_Lower_above ~> greatest_Lower_below
  4117 one_zero ~> carrier_one_zero
  4118 one_not_zero ~> carrier_one_not_zero  (collision with assumption)
  4119 
  4120 
  4121 *** HOL-Nominal ***
  4122 
  4123 * Nominal datatypes can now contain type-variables.
  4124 
  4125 * Commands 'nominal_inductive' and 'equivariance' work with local
  4126 theory targets.
  4127 
  4128 * Nominal primrec can now works with local theory targets and its
  4129 specification syntax now conforms to the general format as seen in
  4130 'inductive' etc.
  4131 
  4132 * Method "perm_simp" honours the standard simplifier attributes
  4133 (no_asm), (no_asm_use) etc.
  4134 
  4135 * The new predicate #* is defined like freshness, except that on the
  4136 left hand side can be a set or list of atoms.
  4137 
  4138 * Experimental command 'nominal_inductive2' derives strong induction
  4139 principles for inductive definitions.  In contrast to
  4140 'nominal_inductive', which can only deal with a fixed number of
  4141 binders, it can deal with arbitrary expressions standing for sets of
  4142 atoms to be avoided.  The only inductive definition we have at the
  4143 moment that needs this generalisation is the typing rule for Lets in
  4144 the algorithm W:
  4145 
  4146  Gamma |- t1 : T1   (x,close Gamma T1)::Gamma |- t2 : T2   x#Gamma
  4147  -----------------------------------------------------------------
  4148          Gamma |- Let x be t1 in t2 : T2
  4149 
  4150 In this rule one wants to avoid all the binders that are introduced by
  4151 "close Gamma T1".  We are looking for other examples where this
  4152 feature might be useful.  Please let us know.
  4153 
  4154 
  4155 *** HOLCF ***
  4156 
  4157 * Reimplemented the simplification procedure for proving continuity
  4158 subgoals.  The new simproc is extensible; users can declare additional
  4159 continuity introduction rules with the attribute [cont2cont].
  4160 
  4161 * The continuity simproc now uses a different introduction rule for
  4162 solving continuity subgoals on terms with lambda abstractions.  In
  4163 some rare cases the new simproc may fail to solve subgoals that the
  4164 old one could solve, and "simp add: cont2cont_LAM" may be necessary.
  4165 Potential INCOMPATIBILITY.
  4166 
  4167 * Command 'fixrec': specification syntax now conforms to the general
  4168 format as seen in 'inductive' etc.  See src/HOLCF/ex/Fixrec_ex.thy for
  4169 examples.  INCOMPATIBILITY.
  4170 
  4171 
  4172 *** ZF ***
  4173 
  4174 * Proof of Zorn's Lemma for partial orders.
  4175 
  4176 
  4177 *** ML ***
  4178 
  4179 * Multithreading for Poly/ML 5.1/5.2 is no longer supported, only for
  4180 Poly/ML 5.2.1 or later.  Important note: the TimeLimit facility
  4181 depends on multithreading, so timouts will not work before Poly/ML
  4182 5.2.1!
  4183 
  4184 * High-level support for concurrent ML programming, see
  4185 src/Pure/Cuncurrent.  The data-oriented model of "future values" is
  4186 particularly convenient to organize independent functional
  4187 computations.  The concept of "synchronized variables" provides a
  4188 higher-order interface for components with shared state, avoiding the
  4189 delicate details of mutexes and condition variables.  (Requires
  4190 Poly/ML 5.2.1 or later.)
  4191 
  4192 * ML bindings produced via Isar commands are stored within the Isar
  4193 context (theory or proof).  Consequently, commands like 'use' and 'ML'
  4194 become thread-safe and work with undo as expected (concerning
  4195 top-level bindings, not side-effects on global references).
  4196 INCOMPATIBILITY, need to provide proper Isar context when invoking the
  4197 compiler at runtime; really global bindings need to be given outside a
  4198 theory.  (Requires Poly/ML 5.2 or later.)
  4199 
  4200 * Command 'ML_prf' is analogous to 'ML' but works within a proof
  4201 context.  Top-level ML bindings are stored within the proof context in
  4202 a purely sequential fashion, disregarding the nested proof structure.
  4203 ML bindings introduced by 'ML_prf' are discarded at the end of the
  4204 proof.  (Requires Poly/ML 5.2 or later.)
  4205 
  4206 * Simplified ML attribute and method setup, cf. functions Attrib.setup
  4207 and Method.setup, as well as Isar commands 'attribute_setup' and
  4208 'method_setup'.  INCOMPATIBILITY for 'method_setup', need to simplify
  4209 existing code accordingly, or use plain 'setup' together with old
  4210 Method.add_method.
  4211 
  4212 * Simplified ML oracle interface Thm.add_oracle promotes 'a -> cterm
  4213 to 'a -> thm, while results are always tagged with an authentic oracle
  4214 name.  The Isar command 'oracle' is now polymorphic, no argument type
  4215 is specified.  INCOMPATIBILITY, need to simplify existing oracle code
  4216 accordingly.  Note that extra performance may be gained by producing
  4217 the cterm carefully, avoiding slow Thm.cterm_of.
  4218 
  4219 * Simplified interface for defining document antiquotations via
  4220 ThyOutput.antiquotation, ThyOutput.output, and optionally
  4221 ThyOutput.maybe_pretty_source.  INCOMPATIBILITY, need to simplify user
  4222 antiquotations accordingly, see src/Pure/Thy/thy_output.ML for common
  4223 examples.
  4224 
  4225 * More systematic treatment of long names, abstract name bindings, and
  4226 name space operations.  Basic operations on qualified names have been
  4227 move from structure NameSpace to Long_Name, e.g. Long_Name.base_name,
  4228 Long_Name.append.  Old type bstring has been mostly replaced by
  4229 abstract type binding (see structure Binding), which supports precise
  4230 qualification by packages and local theory targets, as well as proper
  4231 tracking of source positions.  INCOMPATIBILITY, need to wrap old
  4232 bstring values into Binding.name, or better pass through abstract
  4233 bindings everywhere.  See further src/Pure/General/long_name.ML,
  4234 src/Pure/General/binding.ML and src/Pure/General/name_space.ML
  4235 
  4236 * Result facts (from PureThy.note_thms, ProofContext.note_thms,
  4237 LocalTheory.note etc.) now refer to the *full* internal name, not the
  4238 bstring as before.  INCOMPATIBILITY, not detected by ML type-checking!
  4239 
  4240 * Disposed old type and term read functions (Sign.read_def_typ,
  4241 Sign.read_typ, Sign.read_def_terms, Sign.read_term,
  4242 Thm.read_def_cterms, Thm.read_cterm etc.).  INCOMPATIBILITY, should
  4243 use regular Syntax.read_typ, Syntax.read_term, Syntax.read_typ_global,
  4244 Syntax.read_term_global etc.; see also OldGoals.read_term as last
  4245 resort for legacy applications.
  4246 
  4247 * Disposed old declarations, tactics, tactic combinators that refer to
  4248 the simpset or claset of an implicit theory (such as Addsimps,
  4249 Simp_tac, SIMPSET).  INCOMPATIBILITY, should use @{simpset} etc. in
  4250 embedded ML text, or local_simpset_of with a proper context passed as
  4251 explicit runtime argument.
  4252 
  4253 * Rules and tactics that read instantiations (read_instantiate,
  4254 res_inst_tac, thin_tac, subgoal_tac etc.) now demand a proper proof
  4255 context, which is required for parsing and type-checking.  Moreover,
  4256 the variables are specified as plain indexnames, not string encodings
  4257 thereof.  INCOMPATIBILITY.
  4258 
  4259 * Generic Toplevel.add_hook interface allows to analyze the result of
  4260 transactions.  E.g. see src/Pure/ProofGeneral/proof_general_pgip.ML
  4261 for theorem dependency output of transactions resulting in a new
  4262 theory state.
  4263 
  4264 * ML antiquotations: block-structured compilation context indicated by
  4265 \<lbrace> ... \<rbrace>; additional antiquotation forms:
  4266 
  4267   @{binding name}                         - basic name binding
  4268   @{let ?pat = term}                      - term abbreviation (HO matching)
  4269   @{note name = fact}                     - fact abbreviation
  4270   @{thm fact}                             - singleton fact (with attributes)
  4271   @{thms fact}                            - general fact (with attributes)
  4272   @{lemma prop by method}                 - singleton goal
  4273   @{lemma prop by meth1 meth2}            - singleton goal
  4274   @{lemma prop1 ... propN by method}      - general goal
  4275   @{lemma prop1 ... propN by meth1 meth2} - general goal
  4276   @{lemma (open) ...}                     - open derivation
  4277 
  4278 
  4279 *** System ***
  4280 
  4281 * The Isabelle "emacs" tool provides a specific interface to invoke
  4282 Proof General / Emacs, with more explicit failure if that is not
  4283 installed (the old isabelle-interface script silently falls back on
  4284 isabelle-process).  The PROOFGENERAL_HOME setting determines the
  4285 installation location of the Proof General distribution.
  4286 
  4287 * Isabelle/lib/classes/Pure.jar provides basic support to integrate
  4288 the Isabelle process into a JVM/Scala application.  See
  4289 Isabelle/lib/jedit/plugin for a minimal example.  (The obsolete Java
  4290 process wrapper has been discontinued.)
  4291 
  4292 * Added homegrown Isabelle font with unicode layout, see lib/fonts.
  4293 
  4294 * Various status messages (with exact source position information) are
  4295 emitted, if proper markup print mode is enabled.  This allows
  4296 user-interface components to provide detailed feedback on internal
  4297 prover operations.
  4298 
  4299 
  4300 
  4301 New in Isabelle2008 (June 2008)
  4302 -------------------------------
  4303 
  4304 *** General ***
  4305 
  4306 * The Isabelle/Isar Reference Manual (isar-ref) has been reorganized
  4307 and updated, with formally checked references as hyperlinks.
  4308 
  4309 * Theory loader: use_thy (and similar operations) no longer set the
  4310 implicit ML context, which was occasionally hard to predict and in
  4311 conflict with concurrency.  INCOMPATIBILITY, use ML within Isar which
  4312 provides a proper context already.
  4313 
  4314 * Theory loader: old-style ML proof scripts being *attached* to a thy
  4315 file are no longer supported.  INCOMPATIBILITY, regular 'uses' and
  4316 'use' within a theory file will do the job.
  4317 
  4318 * Name space merge now observes canonical order, i.e. the second space
  4319 is inserted into the first one, while existing entries in the first
  4320 space take precedence.  INCOMPATIBILITY in rare situations, may try to
  4321 swap theory imports.
  4322 
  4323 * Syntax: symbol \<chi> is now considered a letter.  Potential
  4324 INCOMPATIBILITY in identifier syntax etc.
  4325 
  4326 * Outer syntax: string tokens no longer admit escaped white space,
  4327 which was an accidental (undocumented) feature.  INCOMPATIBILITY, use
  4328 white space without escapes.
  4329 
  4330 * Outer syntax: string tokens may contain arbitrary character codes
  4331 specified via 3 decimal digits (as in SML).  E.g. "foo\095bar" for
  4332 "foo_bar".
  4333 
  4334 
  4335 *** Pure ***
  4336 
  4337 * Context-dependent token translations.  Default setup reverts locally
  4338 fixed variables, and adds hilite markup for undeclared frees.
  4339 
  4340 * Unused theorems can be found using the new command 'unused_thms'.
  4341 There are three ways of invoking it:
  4342 
  4343 (1) unused_thms
  4344      Only finds unused theorems in the current theory.
  4345 
  4346 (2) unused_thms thy_1 ... thy_n -
  4347      Finds unused theorems in the current theory and all of its ancestors,
  4348      excluding the theories thy_1 ... thy_n and all of their ancestors.
  4349 
  4350 (3) unused_thms thy_1 ... thy_n - thy'_1 ... thy'_m
  4351      Finds unused theorems in the theories thy'_1 ... thy'_m and all of
  4352      their ancestors, excluding the theories thy_1 ... thy_n and all of
  4353      their ancestors.
  4354 
  4355 In order to increase the readability of the list produced by
  4356 unused_thms, theorems that have been created by a particular instance
  4357 of a theory command such as 'inductive' or 'function' are considered
  4358 to belong to the same "group", meaning that if at least one theorem in
  4359 this group is used, the other theorems in the same group are no longer
  4360 reported as unused.  Moreover, if all theorems in the group are
  4361 unused, only one theorem in the group is displayed.
  4362 
  4363 Note that proof objects have to be switched on in order for
  4364 unused_thms to work properly (i.e. !proofs must be >= 1, which is
  4365 usually the case when using Proof General with the default settings).
  4366 
  4367 * Authentic naming of facts disallows ad-hoc overwriting of previous
  4368 theorems within the same name space.  INCOMPATIBILITY, need to remove
  4369 duplicate fact bindings, or even accidental fact duplications.  Note
  4370 that tools may maintain dynamically scoped facts systematically, using
  4371 PureThy.add_thms_dynamic.
  4372 
  4373 * Command 'hide' now allows to hide from "fact" name space as well.
  4374 
  4375 * Eliminated destructive theorem database, simpset, claset, and
  4376 clasimpset.  Potential INCOMPATIBILITY, really need to observe linear
  4377 update of theories within ML code.
  4378 
  4379 * Eliminated theory ProtoPure and CPure, leaving just one Pure theory.
  4380 INCOMPATIBILITY, object-logics depending on former Pure require
  4381 additional setup PureThy.old_appl_syntax_setup; object-logics
  4382 depending on former CPure need to refer to Pure.
  4383 
  4384 * Commands 'use' and 'ML' are now purely functional, operating on
  4385 theory/local_theory.  Removed former 'ML_setup' (on theory), use 'ML'
  4386 instead.  Added 'ML_val' as mere diagnostic replacement for 'ML'.
  4387 INCOMPATIBILITY.
  4388 
  4389 * Command 'setup': discontinued implicit version with ML reference.
  4390 
  4391 * Instantiation target allows for simultaneous specification of class
  4392 instance operations together with an instantiation proof.
  4393 Type-checking phase allows to refer to class operations uniformly.
  4394 See src/HOL/Complex/Complex.thy for an Isar example and
  4395 src/HOL/Library/Eval.thy for an ML example.
  4396 
  4397 * Indexing of literal facts: be more serious about including only
  4398 facts from the visible specification/proof context, but not the
  4399 background context (locale etc.).  Affects `prop` notation and method
  4400 "fact".  INCOMPATIBILITY: need to name facts explicitly in rare
  4401 situations.
  4402 
  4403 * Method "cases", "induct", "coinduct": removed obsolete/undocumented
  4404 "(open)" option, which used to expose internal bound variables to the
  4405 proof text.
  4406 
  4407 * Isar statements: removed obsolete case "rule_context".
  4408 INCOMPATIBILITY, better use explicit fixes/assumes.
  4409 
  4410 * Locale proofs: default proof step now includes 'unfold_locales';
  4411 hence 'proof' without argument may be used to unfold locale
  4412 predicates.
  4413 
  4414 
  4415 *** Document preparation ***
  4416 
  4417 * Simplified pdfsetup.sty: color/hyperref is used unconditionally for
  4418 both pdf and dvi (hyperlinks usually work in xdvi as well); removed
  4419 obsolete thumbpdf setup (contemporary PDF viewers do this on the
  4420 spot); renamed link color from "darkblue" to "linkcolor" (default
  4421 value unchanged, can be redefined via \definecolor); no longer sets
  4422 "a4paper" option (unnecessary or even intrusive).
  4423 
  4424 * Antiquotation @{lemma A method} proves proposition A by the given
  4425 method (either a method name or a method name plus (optional) method
  4426 arguments in parentheses) and prints A just like @{prop A}.
  4427 
  4428 
  4429 *** HOL ***
  4430 
  4431 * New primrec package.  Specification syntax conforms in style to
  4432 definition/function/....  No separate induction rule is provided.  The
  4433 "primrec" command distinguishes old-style and new-style specifications
  4434 by syntax.  The former primrec package is now named OldPrimrecPackage.
  4435 When adjusting theories, beware: constants stemming from new-style
  4436 primrec specifications have authentic syntax.
  4437 
  4438 * Metis prover is now an order of magnitude faster, and also works
  4439 with multithreading.
  4440 
  4441 * Metis: the maximum number of clauses that can be produced from a
  4442 theorem is now given by the attribute max_clauses.  Theorems that
  4443 exceed this number are ignored, with a warning printed.
  4444 
  4445 * Sledgehammer no longer produces structured proofs by default. To
  4446 enable, declare [[sledgehammer_full = true]].  Attributes
  4447 reconstruction_modulus, reconstruction_sorts renamed
  4448 sledgehammer_modulus, sledgehammer_sorts.  INCOMPATIBILITY.
  4449 
  4450 * Method "induct_scheme" derives user-specified induction rules
  4451 from well-founded induction and completeness of patterns. This factors
  4452 out some operations that are done internally by the function package
  4453 and makes them available separately.  See
  4454 src/HOL/ex/Induction_Scheme.thy for examples.
  4455 
  4456 * More flexible generation of measure functions for termination
  4457 proofs: Measure functions can be declared by proving a rule of the
  4458 form "is_measure f" and giving it the [measure_function] attribute.
  4459 The "is_measure" predicate is logically meaningless (always true), and
  4460 just guides the heuristic.  To find suitable measure functions, the
  4461 termination prover sets up the goal "is_measure ?f" of the appropriate
  4462 type and generates all solutions by prolog-style backwards proof using
  4463 the declared rules.
  4464 
  4465 This setup also deals with rules like 
  4466 
  4467   "is_measure f ==> is_measure (list_size f)"
  4468 
  4469 which accommodates nested datatypes that recurse through lists.
  4470 Similar rules are predeclared for products and option types.
  4471 
  4472 * Turned the type of sets "'a set" into an abbreviation for "'a => bool"
  4473 
  4474   INCOMPATIBILITIES:
  4475 
  4476   - Definitions of overloaded constants on sets have to be replaced by
  4477     definitions on => and bool.
  4478 
  4479   - Some definitions of overloaded operators on sets can now be proved
  4480     using the definitions of the operators on => and bool.  Therefore,
  4481     the following theorems have been renamed:
  4482 
  4483       subset_def   -> subset_eq
  4484       psubset_def  -> psubset_eq
  4485       set_diff_def -> set_diff_eq
  4486       Compl_def    -> Compl_eq
  4487       Sup_set_def  -> Sup_set_eq
  4488       Inf_set_def  -> Inf_set_eq
  4489       sup_set_def  -> sup_set_eq
  4490       inf_set_def  -> inf_set_eq
  4491 
  4492   - Due to the incompleteness of the HO unification algorithm, some
  4493     rules such as subst may require manual instantiation, if some of
  4494     the unknowns in the rule is a set.
  4495 
  4496   - Higher order unification and forward proofs:
  4497     The proof pattern
  4498 
  4499       have "P (S::'a set)" <...>
  4500       then have "EX S. P S" ..
  4501 
  4502     no longer works (due to the incompleteness of the HO unification
  4503     algorithm) and must be replaced by the pattern
  4504 
  4505       have "EX S. P S"
  4506       proof
  4507         show "P S" <...>
  4508       qed
  4509 
  4510   - Calculational reasoning with subst (or similar rules):
  4511     The proof pattern
  4512 
  4513       have "P (S::'a set)" <...>
  4514       also have "S = T" <...>
  4515       finally have "P T" .
  4516 
  4517     no longer works (for similar reasons as the previous example) and
  4518     must be replaced by something like
  4519 
  4520       have "P (S::'a set)" <...>
  4521       moreover have "S = T" <...>
  4522       ultimately have "P T" by simp
  4523 
  4524   - Tactics or packages written in ML code:
  4525     Code performing pattern matching on types via
  4526 
  4527       Type ("set", [T]) => ...
  4528 
  4529     must be rewritten. Moreover, functions like strip_type or
  4530     binder_types no longer return the right value when applied to a
  4531     type of the form
  4532 
  4533       T1 => ... => Tn => U => bool
  4534 
  4535     rather than
  4536 
  4537       T1 => ... => Tn => U set
  4538 
  4539 * Merged theories Wellfounded_Recursion, Accessible_Part and
  4540 Wellfounded_Relations to theory Wellfounded.
  4541 
  4542 * Explicit class "eq" for executable equality.  INCOMPATIBILITY.
  4543 
  4544 * Class finite no longer treats UNIV as class parameter.  Use class
  4545 enum from theory Library/Enum instead to achieve a similar effect.
  4546 INCOMPATIBILITY.
  4547 
  4548 * Theory List: rule list_induct2 now has explicitly named cases "Nil"
  4549 and "Cons".  INCOMPATIBILITY.
  4550 
  4551 * HOL (and FOL): renamed variables in rules imp_elim and swap.
  4552 Potential INCOMPATIBILITY.
  4553 
  4554 * Theory Product_Type: duplicated lemmas split_Pair_apply and
  4555 injective_fst_snd removed, use split_eta and prod_eqI instead.
  4556 Renamed upd_fst to apfst and upd_snd to apsnd.  INCOMPATIBILITY.
  4557 
  4558 * Theory Nat: removed redundant lemmas that merely duplicate lemmas of
  4559 the same name in theory Orderings:
  4560 
  4561   less_trans
  4562   less_linear
  4563   le_imp_less_or_eq
  4564   le_less_trans
  4565   less_le_trans
  4566   less_not_sym
  4567   less_asym
  4568 
  4569 Renamed less_imp_le to less_imp_le_nat, and less_irrefl to
  4570 less_irrefl_nat.  Potential INCOMPATIBILITY due to more general types
  4571 and different variable names.
  4572 
  4573 * Library/Option_ord.thy: Canonical order on option type.
  4574 
  4575 * Library/RBT.thy: Red-black trees, an efficient implementation of
  4576 finite maps.
  4577 
  4578 * Library/Countable.thy: Type class for countable types.
  4579 
  4580 * Theory Int: The representation of numerals has changed.  The infix
  4581 operator BIT and the bit datatype with constructors B0 and B1 have
  4582 disappeared.  INCOMPATIBILITY, use "Int.Bit0 x" and "Int.Bit1 y" in
  4583 place of "x BIT bit.B0" and "y BIT bit.B1", respectively.  Theorems
  4584 involving BIT, B0, or B1 have been renamed with "Bit0" or "Bit1"
  4585 accordingly.
  4586 
  4587 * Theory Nat: definition of <= and < on natural numbers no longer
  4588 depend on well-founded relations.  INCOMPATIBILITY.  Definitions
  4589 le_def and less_def have disappeared.  Consider lemmas not_less
  4590 [symmetric, where ?'a = nat] and less_eq [symmetric] instead.
  4591 
  4592 * Theory Finite_Set: locales ACf, ACe, ACIf, ACIfSL and ACIfSLlin
  4593 (whose purpose mainly is for various fold_set functionals) have been
  4594 abandoned in favor of the existing algebraic classes
  4595 ab_semigroup_mult, comm_monoid_mult, ab_semigroup_idem_mult,
  4596 lower_semilattice (resp. upper_semilattice) and linorder.
  4597 INCOMPATIBILITY.
  4598 
  4599 * Theory Transitive_Closure: induct and cases rules now declare proper
  4600 case_names ("base" and "step").  INCOMPATIBILITY.
  4601 
  4602 * Theorem Inductive.lfp_ordinal_induct generalized to complete
  4603 lattices.  The form set-specific version is available as
  4604 Inductive.lfp_ordinal_induct_set.
  4605 
  4606 * Renamed theorems "power.simps" to "power_int.simps".
  4607 INCOMPATIBILITY.
  4608 
  4609 * Class semiring_div provides basic abstract properties of semirings
  4610 with division and modulo operations.  Subsumes former class dvd_mod.
  4611 
  4612 * Merged theories IntDef, Numeral and IntArith into unified theory
  4613 Int.  INCOMPATIBILITY.
  4614 
  4615 * Theory Library/Code_Index: type "index" now represents natural
  4616 numbers rather than integers.  INCOMPATIBILITY.
  4617 
  4618 * New class "uminus" with operation "uminus" (split of from class
  4619 "minus" which now only has operation "minus", binary).
  4620 INCOMPATIBILITY.
  4621 
  4622 * Constants "card", "internal_split", "option_map" now with authentic
  4623 syntax.  INCOMPATIBILITY.
  4624 
  4625 * Definitions subset_def, psubset_def, set_diff_def, Compl_def,
  4626 le_bool_def, less_bool_def, le_fun_def, less_fun_def, inf_bool_def,
  4627 sup_bool_def, Inf_bool_def, Sup_bool_def, inf_fun_def, sup_fun_def,
  4628 Inf_fun_def, Sup_fun_def, inf_set_def, sup_set_def, Inf_set_def,
  4629 Sup_set_def, le_def, less_def, option_map_def now with object
  4630 equality.  INCOMPATIBILITY.
  4631 
  4632 * Records. Removed K_record, and replaced it by pure lambda term
  4633 %x. c. The simplifier setup is now more robust against eta expansion.
  4634 INCOMPATIBILITY: in cases explicitly referring to K_record.
  4635 
  4636 * Library/Multiset: {#a, b, c#} abbreviates {#a#} + {#b#} + {#c#}.
  4637 
  4638 * Library/ListVector: new theory of arithmetic vector operations.
  4639 
  4640 * Library/Order_Relation: new theory of various orderings as sets of
  4641 pairs.  Defines preorders, partial orders, linear orders and
  4642 well-orders on sets and on types.
  4643 
  4644 
  4645 *** ZF ***
  4646 
  4647 * Renamed some theories to allow to loading both ZF and HOL in the
  4648 same session:
  4649 
  4650   Datatype  -> Datatype_ZF
  4651   Inductive -> Inductive_ZF
  4652   Int       -> Int_ZF
  4653   IntDiv    -> IntDiv_ZF
  4654   Nat       -> Nat_ZF
  4655   List      -> List_ZF
  4656   Main      -> Main_ZF
  4657 
  4658 INCOMPATIBILITY: ZF theories that import individual theories below
  4659 Main might need to be adapted.  Regular theory Main is still
  4660 available, as trivial extension of Main_ZF.
  4661 
  4662 
  4663 *** ML ***
  4664 
  4665 * ML within Isar: antiquotation @{const name} or @{const
  4666 name(typargs)} produces statically-checked Const term.
  4667 
  4668 * Functor NamedThmsFun: data is available to the user as dynamic fact
  4669 (of the same name).  Removed obsolete print command.
  4670 
  4671 * Removed obsolete "use_legacy_bindings" function.
  4672 
  4673 * The ``print mode'' is now a thread-local value derived from a global
  4674 template (the former print_mode reference), thus access becomes
  4675 non-critical.  The global print_mode reference is for session
  4676 management only; user-code should use print_mode_value,
  4677 print_mode_active, PrintMode.setmp etc.  INCOMPATIBILITY.
  4678 
  4679 * Functions system/system_out provide a robust way to invoke external
  4680 shell commands, with propagation of interrupts (requires Poly/ML
  4681 5.2.1).  Do not use OS.Process.system etc. from the basis library!
  4682 
  4683 
  4684 *** System ***
  4685 
  4686 * Default settings: PROOFGENERAL_OPTIONS no longer impose xemacs ---
  4687 in accordance with Proof General 3.7, which prefers GNU emacs.
  4688 
  4689 * isatool tty runs Isabelle process with plain tty interaction;
  4690 optional line editor may be specified via ISABELLE_LINE_EDITOR
  4691 setting, the default settings attempt to locate "ledit" and "rlwrap".
  4692 
  4693 * isatool browser now works with Cygwin as well, using general
  4694 "javapath" function defined in Isabelle process environment.
  4695 
  4696 * YXML notation provides a simple and efficient alternative to
  4697 standard XML transfer syntax.  See src/Pure/General/yxml.ML and
  4698 isatool yxml as described in the Isabelle system manual.
  4699 
  4700 * JVM class isabelle.IsabelleProcess (located in Isabelle/lib/classes)
  4701 provides general wrapper for managing an Isabelle process in a robust
  4702 fashion, with ``cooked'' output from stdin/stderr.
  4703 
  4704 * Rudimentary Isabelle plugin for jEdit (see Isabelle/lib/jedit),
  4705 based on Isabelle/JVM process wrapper (see Isabelle/lib/classes).
  4706 
  4707 * Removed obsolete THIS_IS_ISABELLE_BUILD feature.  NB: the documented
  4708 way of changing the user's settings is via
  4709 ISABELLE_HOME_USER/etc/settings, which is a fully featured bash
  4710 script.
  4711 
  4712 * Multithreading.max_threads := 0 refers to the number of actual CPU
  4713 cores of the underlying machine, which is a good starting point for
  4714 optimal performance tuning.  The corresponding usedir option -M allows
  4715 "max" as an alias for "0".  WARNING: does not work on certain versions
  4716 of Mac OS (with Poly/ML 5.1).
  4717 
  4718 * isabelle-process: non-ML sessions are run with "nice", to reduce the
  4719 adverse effect of Isabelle flooding interactive front-ends (notably
  4720 ProofGeneral / XEmacs).
  4721 
  4722 
  4723 
  4724 New in Isabelle2007 (November 2007)
  4725 -----------------------------------
  4726 
  4727 *** General ***
  4728 
  4729 * More uniform information about legacy features, notably a
  4730 warning/error of "Legacy feature: ...", depending on the state of the
  4731 tolerate_legacy_features flag (default true). FUTURE INCOMPATIBILITY:
  4732 legacy features will disappear eventually.
  4733 
  4734 * Theory syntax: the header format ``theory A = B + C:'' has been
  4735 discontinued in favour of ``theory A imports B C begin''.  Use isatool
  4736 fixheaders to convert existing theory files.  INCOMPATIBILITY.
  4737 
  4738 * Theory syntax: the old non-Isar theory file format has been
  4739 discontinued altogether.  Note that ML proof scripts may still be used
  4740 with Isar theories; migration is usually quite simple with the ML
  4741 function use_legacy_bindings.  INCOMPATIBILITY.
  4742 
  4743 * Theory syntax: some popular names (e.g. 'class', 'declaration',
  4744 'fun', 'help', 'if') are now keywords.  INCOMPATIBILITY, use double
  4745 quotes.
  4746 
  4747 * Theory loader: be more serious about observing the static theory
  4748 header specifications (including optional directories), but not the
  4749 accidental file locations of previously successful loads.  The strict
  4750 update policy of former update_thy is now already performed by
  4751 use_thy, so the former has been removed; use_thys updates several
  4752 theories simultaneously, just as 'imports' within a theory header
  4753 specification, but without merging the results.  Potential
  4754 INCOMPATIBILITY: may need to refine theory headers and commands
  4755 ROOT.ML which depend on load order.
  4756 
  4757 * Theory loader: optional support for content-based file
  4758 identification, instead of the traditional scheme of full physical
  4759 path plus date stamp; configured by the ISABELLE_FILE_IDENT setting
  4760 (cf. the system manual).  The new scheme allows to work with
  4761 non-finished theories in persistent session images, such that source
  4762 files may be moved later on without requiring reloads.
  4763 
  4764 * Theory loader: old-style ML proof scripts being *attached* to a thy
  4765 file (with the same base name as the theory) are considered a legacy
  4766 feature, which will disappear eventually. Even now, the theory loader
  4767 no longer maintains dependencies on such files.
  4768 
  4769 * Syntax: the scope for resolving ambiguities via type-inference is
  4770 now limited to individual terms, instead of whole simultaneous
  4771 specifications as before. This greatly reduces the complexity of the
  4772 syntax module and improves flexibility by separating parsing and
  4773 type-checking. INCOMPATIBILITY: additional type-constraints (explicit
  4774 'fixes' etc.) are required in rare situations.
  4775 
  4776 * Syntax: constants introduced by new-style packages ('definition',
  4777 'abbreviation' etc.) are passed through the syntax module in
  4778 ``authentic mode''. This means that associated mixfix annotations
  4779 really stick to such constants, independently of potential name space
  4780 ambiguities introduced later on. INCOMPATIBILITY: constants in parse
  4781 trees are represented slightly differently, may need to adapt syntax
  4782 translations accordingly. Use CONST marker in 'translations' and
  4783 @{const_syntax} antiquotation in 'parse_translation' etc.
  4784 
  4785 * Legacy goal package: reduced interface to the bare minimum required
  4786 to keep existing proof scripts running.  Most other user-level
  4787 functions are now part of the OldGoals structure, which is *not* open
  4788 by default (consider isatool expandshort before open OldGoals).
  4789 Removed top_sg, prin, printyp, pprint_term/typ altogether, because
  4790 these tend to cause confusion about the actual goal (!) context being
  4791 used here, which is not necessarily the same as the_context().
  4792 
  4793 * Command 'find_theorems': supports "*" wild-card in "name:"
  4794 criterion; "with_dups" option.  Certain ProofGeneral versions might
  4795 support a specific search form (see ProofGeneral/CHANGES).
  4796 
  4797 * The ``prems limit'' option (cf. ProofContext.prems_limit) is now -1
  4798 by default, which means that "prems" (and also "fixed variables") are
  4799 suppressed from proof state output.  Note that the ProofGeneral
  4800 settings mechanism allows to change and save options persistently, but
  4801 older versions of Isabelle will fail to start up if a negative prems
  4802 limit is imposed.
  4803 
  4804 * Local theory targets may be specified by non-nested blocks of
  4805 ``context/locale/class ... begin'' followed by ``end''.  The body may
  4806 contain definitions, theorems etc., including any derived mechanism
  4807 that has been implemented on top of these primitives.  This concept
  4808 generalizes the existing ``theorem (in ...)'' towards more versatility
  4809 and scalability.
  4810 
  4811 * Proof General interface: proper undo of final 'end' command;
  4812 discontinued Isabelle/classic mode (ML proof scripts).
  4813 
  4814 
  4815 *** Document preparation ***
  4816 
  4817 * Added antiquotation @{theory name} which prints the given name,
  4818 after checking that it refers to a valid ancestor theory in the
  4819 current context.
  4820 
  4821 * Added antiquotations @{ML_type text} and @{ML_struct text} which
  4822 check the given source text as ML type/structure, printing verbatim.
  4823 
  4824 * Added antiquotation @{abbrev "c args"} which prints the abbreviation
  4825 "c args == rhs" given in the current context.  (Any number of
  4826 arguments may be given on the LHS.)
  4827 
  4828 
  4829 *** Pure ***
  4830 
  4831 * The 'class' package offers a combination of axclass and locale to
  4832 achieve Haskell-like type classes in Isabelle.  Definitions and
  4833 theorems within a class context produce both relative results (with
  4834 implicit parameters according to the locale context), and polymorphic
  4835 constants with qualified polymorphism (according to the class
  4836 context).  Within the body context of a 'class' target, a separate
  4837 syntax layer ("user space type system") takes care of converting
  4838 between global polymorphic consts and internal locale representation.
  4839 See src/HOL/ex/Classpackage.thy for examples (as well as main HOL).
  4840 "isatool doc classes" provides a tutorial.
  4841 
  4842 * Generic code generator framework allows to generate executable
  4843 code for ML and Haskell (including Isabelle classes).  A short usage
  4844 sketch:
  4845 
  4846     internal compilation:
  4847         export_code <list of constants (term syntax)> in SML
  4848     writing SML code to a file:
  4849         export_code <list of constants (term syntax)> in SML <filename>
  4850     writing OCaml code to a file:
  4851         export_code <list of constants (term syntax)> in OCaml <filename>
  4852     writing Haskell code to a bunch of files:
  4853         export_code <list of constants (term syntax)> in Haskell <filename>
  4854 
  4855     evaluating closed propositions to True/False using code generation:
  4856         method ``eval''
  4857 
  4858 Reasonable default setup of framework in HOL.
  4859 
  4860 Theorem attributs for selecting and transforming function equations theorems:
  4861 
  4862     [code fun]:        select a theorem as function equation for a specific constant
  4863     [code fun del]:    deselect a theorem as function equation for a specific constant
  4864     [code inline]:     select an equation theorem for unfolding (inlining) in place
  4865     [code inline del]: deselect an equation theorem for unfolding (inlining) in place
  4866 
  4867 User-defined serializations (target in {SML, OCaml, Haskell}):
  4868 
  4869     code_const <and-list of constants (term syntax)>
  4870       {(target) <and-list of const target syntax>}+
  4871 
  4872     code_type <and-list of type constructors>
  4873       {(target) <and-list of type target syntax>}+
  4874 
  4875     code_instance <and-list of instances>
  4876       {(target)}+
  4877         where instance ::= <type constructor> :: <class>
  4878 
  4879     code_class <and_list of classes>
  4880       {(target) <and-list of class target syntax>}+
  4881         where class target syntax ::= <class name> {where {<classop> == <target syntax>}+}?
  4882 
  4883 code_instance and code_class only are effective to target Haskell.
  4884 
  4885 For example usage see src/HOL/ex/Codegenerator.thy and
  4886 src/HOL/ex/Codegenerator_Pretty.thy.  A separate tutorial on code
  4887 generation from Isabelle/HOL theories is available via "isatool doc
  4888 codegen".
  4889 
  4890 * Code generator: consts in 'consts_code' Isar commands are now
  4891 referred to by usual term syntax (including optional type
  4892 annotations).
  4893 
  4894 * Command 'no_translations' removes translation rules from theory
  4895 syntax.
  4896 
  4897 * Overloaded definitions are now actually checked for acyclic
  4898 dependencies.  The overloading scheme is slightly more general than
  4899 that of Haskell98, although Isabelle does not demand an exact
  4900 correspondence to type class and instance declarations.
  4901 INCOMPATIBILITY, use ``defs (unchecked overloaded)'' to admit more
  4902 exotic versions of overloading -- at the discretion of the user!
  4903 
  4904 Polymorphic constants are represented via type arguments, i.e. the
  4905 instantiation that matches an instance against the most general
  4906 declaration given in the signature.  For example, with the declaration
  4907 c :: 'a => 'a => 'a, an instance c :: nat => nat => nat is represented
  4908 as c(nat).  Overloading is essentially simultaneous structural
  4909 recursion over such type arguments.  Incomplete specification patterns
  4910 impose global constraints on all occurrences, e.g. c('a * 'a) on the
  4911 LHS means that more general c('a * 'b) will be disallowed on any RHS.
  4912 Command 'print_theory' outputs the normalized system of recursive
  4913 equations, see section "definitions".
  4914 
  4915 * Configuration options are maintained within the theory or proof
  4916 context (with name and type bool/int/string), providing a very simple
  4917 interface to a poor-man's version of general context data.  Tools may
  4918 declare options in ML (e.g. using Attrib.config_int) and then refer to
  4919 these values using Config.get etc.  Users may change options via an
  4920 associated attribute of the same name.  This form of context
  4921 declaration works particularly well with commands 'declare' or
  4922 'using', for example ``declare [[foo = 42]]''.  Thus it has become
  4923 very easy to avoid global references, which would not observe Isar
  4924 toplevel undo/redo and fail to work with multithreading.
  4925 
  4926 Various global ML references of Pure and HOL have been turned into
  4927 configuration options:
  4928 
  4929   Unify.search_bound		unify_search_bound
  4930   Unify.trace_bound		unify_trace_bound
  4931   Unify.trace_simp		unify_trace_simp
  4932   Unify.trace_types		unify_trace_types
  4933   Simplifier.simp_depth_limit	simp_depth_limit
  4934   Blast.depth_limit		blast_depth_limit
  4935   DatatypeProp.dtK		datatype_distinctness_limit
  4936   fast_arith_neq_limit  	fast_arith_neq_limit
  4937   fast_arith_split_limit	fast_arith_split_limit
  4938 
  4939 * Named collections of theorems may be easily installed as context
  4940 data using the functor NamedThmsFun (see also
  4941 src/Pure/Tools/named_thms.ML).  The user may add or delete facts via
  4942 attributes; there is also a toplevel print command.  This facility is
  4943 just a common case of general context data, which is the preferred way
  4944 for anything more complex than just a list of facts in canonical
  4945 order.
  4946 
  4947 * Isar: command 'declaration' augments a local theory by generic
  4948 declaration functions written in ML.  This enables arbitrary content
  4949 being added to the context, depending on a morphism that tells the
  4950 difference of the original declaration context wrt. the application
  4951 context encountered later on.
  4952 
  4953 * Isar: proper interfaces for simplification procedures.  Command
  4954 'simproc_setup' declares named simprocs (with match patterns, and body
  4955 text in ML).  Attribute "simproc" adds/deletes simprocs in the current
  4956 context.  ML antiquotation @{simproc name} retrieves named simprocs.
  4957 
  4958 * Isar: an extra pair of brackets around attribute declarations
  4959 abbreviates a theorem reference involving an internal dummy fact,
  4960 which will be ignored later --- only the effect of the attribute on
  4961 the background context will persist.  This form of in-place
  4962 declarations is particularly useful with commands like 'declare' and
  4963 'using', for example ``have A using [[simproc a]] by simp''.
  4964 
  4965 * Isar: method "assumption" (and implicit closing of subproofs) now
  4966 takes simple non-atomic goal assumptions into account: after applying
  4967 an assumption as a rule the resulting subgoals are solved by atomic
  4968 assumption steps.  This is particularly useful to finish 'obtain'
  4969 goals, such as "!!x. (!!x. P x ==> thesis) ==> P x ==> thesis",
  4970 without referring to the original premise "!!x. P x ==> thesis" in the
  4971 Isar proof context.  POTENTIAL INCOMPATIBILITY: method "assumption" is
  4972 more permissive.
  4973 
  4974 * Isar: implicit use of prems from the Isar proof context is
  4975 considered a legacy feature.  Common applications like ``have A .''
  4976 may be replaced by ``have A by fact'' or ``note `A`''.  In general,
  4977 referencing facts explicitly here improves readability and
  4978 maintainability of proof texts.
  4979 
  4980 * Isar: improper proof element 'guess' is like 'obtain', but derives
  4981 the obtained context from the course of reasoning!  For example:
  4982 
  4983   assume "EX x y. A x & B y"   -- "any previous fact"
  4984   then guess x and y by clarify
  4985 
  4986 This technique is potentially adventurous, depending on the facts and
  4987 proof tools being involved here.
  4988 
  4989 * Isar: known facts from the proof context may be specified as literal
  4990 propositions, using ASCII back-quote syntax.  This works wherever
  4991 named facts used to be allowed so far, in proof commands, proof
  4992 methods, attributes etc.  Literal facts are retrieved from the context
  4993 according to unification of type and term parameters.  For example,
  4994 provided that "A" and "A ==> B" and "!!x. P x ==> Q x" are known
  4995 theorems in the current context, then these are valid literal facts:
  4996 `A` and `A ==> B` and `!!x. P x ==> Q x" as well as `P a ==> Q a` etc.
  4997 
  4998 There is also a proof method "fact" which does the same composition
  4999 for explicit goal states, e.g. the following proof texts coincide with
  5000 certain special cases of literal facts:
  5001 
  5002   have "A" by fact                 ==  note `A`
  5003   have "A ==> B" by fact           ==  note `A ==> B`
  5004   have "!!x. P x ==> Q x" by fact  ==  note `!!x. P x ==> Q x`
  5005   have "P a ==> Q a" by fact       ==  note `P a ==> Q a`
  5006 
  5007 * Isar: ":" (colon) is no longer a symbolic identifier character in
  5008 outer syntax.  Thus symbolic identifiers may be used without
  5009 additional white space in declarations like this: ``assume *: A''.
  5010 
  5011 * Isar: 'print_facts' prints all local facts of the current context,
  5012 both named and unnamed ones.
  5013 
  5014 * Isar: 'def' now admits simultaneous definitions, e.g.:
  5015 
  5016   def x == "t" and y == "u"
  5017 
  5018 * Isar: added command 'unfolding', which is structurally similar to
  5019 'using', but affects both the goal state and facts by unfolding given
  5020 rewrite rules.  Thus many occurrences of the 'unfold' method or
  5021 'unfolded' attribute may be replaced by first-class proof text.
  5022 
  5023 * Isar: methods 'unfold' / 'fold', attributes 'unfolded' / 'folded',
  5024 and command 'unfolding' now all support object-level equalities
  5025 (potentially conditional).  The underlying notion of rewrite rule is
  5026 analogous to the 'rule_format' attribute, but *not* that of the
  5027 Simplifier (which is usually more generous).
  5028 
  5029 * Isar: the new attribute [rotated n] (default n = 1) rotates the
  5030 premises of a theorem by n. Useful in conjunction with drule.
  5031 
  5032 * Isar: the goal restriction operator [N] (default N = 1) evaluates a
  5033 method expression within a sandbox consisting of the first N
  5034 sub-goals, which need to exist.  For example, ``simp_all [3]''
  5035 simplifies the first three sub-goals, while (rule foo, simp_all)[]
  5036 simplifies all new goals that emerge from applying rule foo to the
  5037 originally first one.
  5038 
  5039 * Isar: schematic goals are no longer restricted to higher-order
  5040 patterns; e.g. ``lemma "?P(?x)" by (rule TrueI)'' now works as
  5041 expected.
  5042 
  5043 * Isar: the conclusion of a long theorem statement is now either
  5044 'shows' (a simultaneous conjunction, as before), or 'obtains'
  5045 (essentially a disjunction of cases with local parameters and
  5046 assumptions).  The latter allows to express general elimination rules
  5047 adequately; in this notation common elimination rules look like this:
  5048 
  5049   lemma exE:    -- "EX x. P x ==> (!!x. P x ==> thesis) ==> thesis"
  5050     assumes "EX x. P x"
  5051     obtains x where "P x"
  5052 
  5053   lemma conjE:  -- "A & B ==> (A ==> B ==> thesis) ==> thesis"
  5054     assumes "A & B"
  5055     obtains A and B
  5056 
  5057   lemma disjE:  -- "A | B ==> (A ==> thesis) ==> (B ==> thesis) ==> thesis"
  5058     assumes "A | B"
  5059     obtains
  5060       A
  5061     | B
  5062 
  5063 The subsequent classical rules even refer to the formal "thesis"
  5064 explicitly:
  5065 
  5066   lemma classical:     -- "(~ thesis ==> thesis) ==> thesis"
  5067     obtains "~ thesis"
  5068 
  5069   lemma Peirce's_Law:  -- "((thesis ==> something) ==> thesis) ==> thesis"
  5070     obtains "thesis ==> something"
  5071 
  5072 The actual proof of an 'obtains' statement is analogous to that of the
  5073 Isar proof element 'obtain', only that there may be several cases.
  5074 Optional case names may be specified in parentheses; these will be
  5075 available both in the present proof and as annotations in the
  5076 resulting rule, for later use with the 'cases' method (cf. attribute
  5077 case_names).
  5078 
  5079 * Isar: the assumptions of a long theorem statement are available as
  5080 "assms" fact in the proof context.  This is more appropriate than the
  5081 (historical) "prems", which refers to all assumptions of the current
  5082 context, including those from the target locale, proof body etc.
  5083 
  5084 * Isar: 'print_statement' prints theorems from the current theory or
  5085 proof context in long statement form, according to the syntax of a
  5086 top-level lemma.
  5087 
  5088 * Isar: 'obtain' takes an optional case name for the local context
  5089 introduction rule (default "that").
  5090 
  5091 * Isar: removed obsolete 'concl is' patterns.  INCOMPATIBILITY, use
  5092 explicit (is "_ ==> ?foo") in the rare cases where this still happens
  5093 to occur.
  5094 
  5095 * Pure: syntax "CONST name" produces a fully internalized constant
  5096 according to the current context.  This is particularly useful for
  5097 syntax translations that should refer to internal constant
  5098 representations independently of name spaces.
  5099 
  5100 * Pure: syntax constant for foo (binder "FOO ") is called "foo_binder"
  5101 instead of "FOO ". This allows multiple binder declarations to coexist
  5102 in the same context.  INCOMPATIBILITY.
  5103 
  5104 * Isar/locales: 'notation' provides a robust interface to the 'syntax'
  5105 primitive that also works in a locale context (both for constants and
  5106 fixed variables). Type declaration and internal syntactic representation
  5107 of given constants retrieved from the context. Likewise, the
  5108 'no_notation' command allows to remove given syntax annotations from the
  5109 current context.
  5110 
  5111 * Isar/locales: new derived specification elements 'axiomatization',
  5112 'definition', 'abbreviation', which support type-inference, admit
  5113 object-level specifications (equality, equivalence).  See also the
  5114 isar-ref manual.  Examples:
  5115 
  5116   axiomatization
  5117     eq  (infix "===" 50) where
  5118     eq_refl: "x === x" and eq_subst: "x === y ==> P x ==> P y"
  5119 
  5120   definition "f x y = x + y + 1"
  5121   definition g where "g x = f x x"
  5122 
  5123   abbreviation
  5124     neq  (infix "=!=" 50) where
  5125     "x =!= y == ~ (x === y)"
  5126 
  5127 These specifications may be also used in a locale context.  Then the
  5128 constants being introduced depend on certain fixed parameters, and the
  5129 constant name is qualified by the locale base name.  An internal
  5130 abbreviation takes care for convenient input and output, making the
  5131 parameters implicit and using the original short name.  See also
  5132 src/HOL/ex/Abstract_NAT.thy for an example of deriving polymorphic
  5133 entities from a monomorphic theory.
  5134 
  5135 Presently, abbreviations are only available 'in' a target locale, but
  5136 not inherited by general import expressions.  Also note that
  5137 'abbreviation' may be used as a type-safe replacement for 'syntax' +
  5138 'translations' in common applications.  The "no_abbrevs" print mode
  5139 prevents folding of abbreviations in term output.
  5140 
  5141 Concrete syntax is attached to specified constants in internal form,
  5142 independently of name spaces.  The parse tree representation is
  5143 slightly different -- use 'notation' instead of raw 'syntax', and
  5144 'translations' with explicit "CONST" markup to accommodate this.
  5145 
  5146 * Pure/Isar: unified syntax for new-style specification mechanisms
  5147 (e.g.  'definition', 'abbreviation', or 'inductive' in HOL) admits
  5148 full type inference and dummy patterns ("_").  For example:
  5149 
  5150   definition "K x _ = x"
  5151 
  5152   inductive conj for A B
  5153   where "A ==> B ==> conj A B"
  5154 
  5155 * Pure: command 'print_abbrevs' prints all constant abbreviations of
  5156 the current context.  Print mode "no_abbrevs" prevents inversion of
  5157 abbreviations on output.
  5158 
  5159 * Isar/locales: improved parameter handling: use of locales "var" and
  5160 "struct" no longer necessary; - parameter renamings are no longer
  5161 required to be injective.  For example, this allows to define
  5162 endomorphisms as locale endom = homom mult mult h.
  5163 
  5164 * Isar/locales: changed the way locales with predicates are defined.
  5165 Instead of accumulating the specification, the imported expression is
  5166 now an interpretation.  INCOMPATIBILITY: different normal form of
  5167 locale expressions.  In particular, in interpretations of locales with
  5168 predicates, goals repesenting already interpreted fragments are not
  5169 removed automatically.  Use methods `intro_locales' and
  5170 `unfold_locales'; see below.
  5171 
  5172 * Isar/locales: new methods `intro_locales' and `unfold_locales'
  5173 provide backward reasoning on locales predicates.  The methods are
  5174 aware of interpretations and discharge corresponding goals.
  5175 `intro_locales' is less aggressive then `unfold_locales' and does not
  5176 unfold predicates to assumptions.
  5177 
  5178 * Isar/locales: the order in which locale fragments are accumulated
  5179 has changed.  This enables to override declarations from fragments due
  5180 to interpretations -- for example, unwanted simp rules.
  5181 
  5182 * Isar/locales: interpretation in theories and proof contexts has been
  5183 extended.  One may now specify (and prove) equations, which are
  5184 unfolded in interpreted theorems.  This is useful for replacing
  5185 defined concepts (constants depending on locale parameters) by
  5186 concepts already existing in the target context.  Example:
  5187 
  5188   interpretation partial_order ["op <= :: [int, int] => bool"]
  5189     where "partial_order.less (op <=) (x::int) y = (x < y)"
  5190 
  5191 Typically, the constant `partial_order.less' is created by a
  5192 definition specification element in the context of locale
  5193 partial_order.
  5194 
  5195 * Method "induct": improved internal context management to support
  5196 local fixes and defines on-the-fly. Thus explicit meta-level
  5197 connectives !!  and ==> are rarely required anymore in inductive goals
  5198 (using object-logic connectives for this purpose has been long
  5199 obsolete anyway). Common proof patterns are explained in
  5200 src/HOL/Induct/Common_Patterns.thy, see also
  5201 src/HOL/Isar_examples/Puzzle.thy and src/HOL/Lambda for realistic
  5202 examples.
  5203 
  5204 * Method "induct": improved handling of simultaneous goals. Instead of
  5205 introducing object-level conjunction, the statement is now split into
  5206 several conclusions, while the corresponding symbolic cases are nested
  5207 accordingly. INCOMPATIBILITY, proofs need to be structured explicitly,
  5208 see src/HOL/Induct/Common_Patterns.thy, for example.
  5209 
  5210 * Method "induct": mutual induction rules are now specified as a list
  5211 of rule sharing the same induction cases. HOL packages usually provide
  5212 foo_bar.inducts for mutually defined items foo and bar (e.g. inductive
  5213 predicates/sets or datatypes). INCOMPATIBILITY, users need to specify
  5214 mutual induction rules differently, i.e. like this:
  5215 
  5216   (induct rule: foo_bar.inducts)
  5217   (induct set: foo bar)
  5218   (induct pred: foo bar)
  5219   (induct type: foo bar)
  5220 
  5221 The ML function ProjectRule.projections turns old-style rules into the
  5222 new format.
  5223 
  5224 * Method "coinduct": dual of induction, see
  5225 src/HOL/Library/Coinductive_List.thy for various examples.
  5226 
  5227 * Method "cases", "induct", "coinduct": the ``(open)'' option is
  5228 considered a legacy feature.
  5229 
  5230 * Attribute "symmetric" produces result with standardized schematic
  5231 variables (index 0).  Potential INCOMPATIBILITY.
  5232 
  5233 * Simplifier: by default the simplifier trace only shows top level
  5234 rewrites now. That is, trace_simp_depth_limit is set to 1 by
  5235 default. Thus there is less danger of being flooded by the trace. The
  5236 trace indicates where parts have been suppressed.
  5237   
  5238 * Provers/classical: removed obsolete classical version of elim_format
  5239 attribute; classical elim/dest rules are now treated uniformly when
  5240 manipulating the claset.
  5241 
  5242 * Provers/classical: stricter checks to ensure that supplied intro,
  5243 dest and elim rules are well-formed; dest and elim rules must have at
  5244 least one premise.
  5245 
  5246 * Provers/classical: attributes dest/elim/intro take an optional
  5247 weight argument for the rule (just as the Pure versions).  Weights are
  5248 ignored by automated tools, but determine the search order of single
  5249 rule steps.
  5250 
  5251 * Syntax: input syntax now supports dummy variable binding "%_. b",
  5252 where the body does not mention the bound variable.  Note that dummy
  5253 patterns implicitly depend on their context of bounds, which makes
  5254 "{_. _}" match any set comprehension as expected.  Potential
  5255 INCOMPATIBILITY -- parse translations need to cope with syntactic
  5256 constant "_idtdummy" in the binding position.
  5257 
  5258 * Syntax: removed obsolete syntactic constant "_K" and its associated
  5259 parse translation.  INCOMPATIBILITY -- use dummy abstraction instead,
  5260 for example "A -> B" => "Pi A (%_. B)".
  5261 
  5262 * Pure: 'class_deps' command visualizes the subclass relation, using
  5263 the graph browser tool.
  5264 
  5265 * Pure: 'print_theory' now suppresses certain internal declarations by
  5266 default; use '!' option for full details.
  5267 
  5268 
  5269 *** HOL ***
  5270 
  5271 * Method "metis" proves goals by applying the Metis general-purpose
  5272 resolution prover (see also http://gilith.com/software/metis/).
  5273 Examples are in the directory MetisExamples.  WARNING: the
  5274 Isabelle/HOL-Metis integration does not yet work properly with
  5275 multi-threading.
  5276   
  5277 * Command 'sledgehammer' invokes external automatic theorem provers as
  5278 background processes.  It generates calls to the "metis" method if
  5279 successful. These can be pasted into the proof.  Users do not have to
  5280 wait for the automatic provers to return.  WARNING: does not really
  5281 work with multi-threading.
  5282 
  5283 * New "auto_quickcheck" feature tests outermost goal statements for
  5284 potential counter-examples.  Controlled by ML references
  5285 auto_quickcheck (default true) and auto_quickcheck_time_limit (default
  5286 5000 milliseconds).  Fails silently if statements is outside of
  5287 executable fragment, or any other codgenerator problem occurs.
  5288 
  5289 * New constant "undefined" with axiom "undefined x = undefined".
  5290 
  5291 * Added class "HOL.eq", allowing for code generation with polymorphic
  5292 equality.
  5293 
  5294 * Some renaming of class constants due to canonical name prefixing in
  5295 the new 'class' package:
  5296 
  5297     HOL.abs ~> HOL.abs_class.abs
  5298     HOL.divide ~> HOL.divide_class.divide
  5299     0 ~> HOL.zero_class.zero
  5300     1 ~> HOL.one_class.one
  5301     op + ~> HOL.plus_class.plus
  5302     op - ~> HOL.minus_class.minus
  5303     uminus ~> HOL.minus_class.uminus
  5304     op * ~> HOL.times_class.times
  5305     op < ~> HOL.ord_class.less
  5306     op <= > HOL.ord_class.less_eq
  5307     Nat.power ~> Power.power_class.power
  5308     Nat.size ~> Nat.size_class.size
  5309     Numeral.number_of ~> Numeral.number_class.number_of
  5310     FixedPoint.Inf ~> Lattices.complete_lattice_class.Inf
  5311     FixedPoint.Sup ~> Lattices.complete_lattice_class.Sup
  5312     Orderings.min ~> Orderings.ord_class.min
  5313     Orderings.max ~> Orderings.ord_class.max
  5314     Divides.op div ~> Divides.div_class.div
  5315     Divides.op mod ~> Divides.div_class.mod
  5316     Divides.op dvd ~> Divides.div_class.dvd
  5317 
  5318 INCOMPATIBILITY.  Adaptions may be required in the following cases:
  5319 
  5320 a) User-defined constants using any of the names "plus", "minus",
  5321 "times", "less" or "less_eq". The standard syntax translations for
  5322 "+", "-" and "*" may go wrong.  INCOMPATIBILITY: use more specific
  5323 names.
  5324 
  5325 b) Variables named "plus", "minus", "times", "less", "less_eq"
  5326 INCOMPATIBILITY: use more specific names.
  5327 
  5328 c) Permutative equations (e.g. "a + b = b + a")
  5329 Since the change of names also changes the order of terms, permutative
  5330 rewrite rules may get applied in a different order. Experience shows
  5331 that this is rarely the case (only two adaptions in the whole Isabelle
  5332 distribution).  INCOMPATIBILITY: rewrite proofs
  5333 
  5334 d) ML code directly refering to constant names
  5335 This in general only affects hand-written proof tactics, simprocs and
  5336 so on.  INCOMPATIBILITY: grep your sourcecode and replace names.
  5337 Consider using @{const_name} antiquotation.
  5338 
  5339 * New class "default" with associated constant "default".
  5340 
  5341 * Function "sgn" is now overloaded and available on int, real, complex
  5342 (and other numeric types), using class "sgn".  Two possible defs of
  5343 sgn are given as equational assumptions in the classes sgn_if and
  5344 sgn_div_norm; ordered_idom now also inherits from sgn_if.
  5345 INCOMPATIBILITY.
  5346 
  5347 * Locale "partial_order" now unified with class "order" (cf. theory
  5348 Orderings), added parameter "less".  INCOMPATIBILITY.
  5349 
  5350 * Renamings in classes "order" and "linorder": facts "refl", "trans" and
  5351 "cases" to "order_refl", "order_trans" and "linorder_cases", to avoid
  5352 clashes with HOL "refl" and "trans".  INCOMPATIBILITY.
  5353 
  5354 * Classes "order" and "linorder": potential INCOMPATIBILITY due to
  5355 changed order of proof goals in instance proofs.
  5356 
  5357 * The transitivity reasoner for partial and linear orders is set up
  5358 for classes "order" and "linorder".  Instances of the reasoner are available
  5359 in all contexts importing or interpreting the corresponding locales.
  5360 Method "order" invokes the reasoner separately; the reasoner
  5361 is also integrated with the Simplifier as a solver.  Diagnostic
  5362 command 'print_orders' shows the available instances of the reasoner
  5363 in the current context.
  5364 
  5365 * Localized monotonicity predicate in theory "Orderings"; integrated
  5366 lemmas max_of_mono and min_of_mono with this predicate.
  5367 INCOMPATIBILITY.
  5368 
  5369 * Formulation of theorem "dense" changed slightly due to integration
  5370 with new class dense_linear_order.
  5371 
  5372 * Uniform lattice theory development in HOL.
  5373 
  5374     constants "meet" and "join" now named "inf" and "sup"
  5375     constant "Meet" now named "Inf"
  5376 
  5377     classes "meet_semilorder" and "join_semilorder" now named
  5378       "lower_semilattice" and "upper_semilattice"
  5379     class "lorder" now named "lattice"
  5380     class "comp_lat" now named "complete_lattice"
  5381 
  5382     Instantiation of lattice classes allows explicit definitions
  5383     for "inf" and "sup" operations (or "Inf" and "Sup" for complete lattices).
  5384 
  5385   INCOMPATIBILITY.  Theorem renames:
  5386 
  5387     meet_left_le            ~> inf_le1
  5388     meet_right_le           ~> inf_le2
  5389     join_left_le            ~> sup_ge1
  5390     join_right_le           ~> sup_ge2
  5391     meet_join_le            ~> inf_sup_ord
  5392     le_meetI                ~> le_infI
  5393     join_leI                ~> le_supI
  5394     le_meet                 ~> le_inf_iff
  5395     le_join                 ~> ge_sup_conv
  5396     meet_idempotent         ~> inf_idem
  5397     join_idempotent         ~> sup_idem
  5398     meet_comm               ~> inf_commute
  5399     join_comm               ~> sup_commute
  5400     meet_leI1               ~> le_infI1
  5401     meet_leI2               ~> le_infI2
  5402     le_joinI1               ~> le_supI1
  5403     le_joinI2               ~> le_supI2
  5404     meet_assoc              ~> inf_assoc
  5405     join_assoc              ~> sup_assoc
  5406     meet_left_comm          ~> inf_left_commute
  5407     meet_left_idempotent    ~> inf_left_idem
  5408     join_left_comm          ~> sup_left_commute
  5409     join_left_idempotent    ~> sup_left_idem
  5410     meet_aci                ~> inf_aci
  5411     join_aci                ~> sup_aci
  5412     le_def_meet             ~> le_iff_inf
  5413     le_def_join             ~> le_iff_sup
  5414     join_absorp2            ~> sup_absorb2
  5415     join_absorp1            ~> sup_absorb1
  5416     meet_absorp1            ~> inf_absorb1
  5417     meet_absorp2            ~> inf_absorb2
  5418     meet_join_absorp        ~> inf_sup_absorb
  5419     join_meet_absorp        ~> sup_inf_absorb
  5420     distrib_join_le         ~> distrib_sup_le
  5421     distrib_meet_le         ~> distrib_inf_le
  5422 
  5423     add_meet_distrib_left   ~> add_inf_distrib_left
  5424     add_join_distrib_left   ~> add_sup_distrib_left
  5425     is_join_neg_meet        ~> is_join_neg_inf
  5426     is_meet_neg_join        ~> is_meet_neg_sup
  5427     add_meet_distrib_right  ~> add_inf_distrib_right
  5428     add_join_distrib_right  ~> add_sup_distrib_right
  5429     add_meet_join_distribs  ~> add_sup_inf_distribs
  5430     join_eq_neg_meet        ~> sup_eq_neg_inf
  5431     meet_eq_neg_join        ~> inf_eq_neg_sup
  5432     add_eq_meet_join        ~> add_eq_inf_sup
  5433     meet_0_imp_0            ~> inf_0_imp_0
  5434     join_0_imp_0            ~> sup_0_imp_0
  5435     meet_0_eq_0             ~> inf_0_eq_0
  5436     join_0_eq_0             ~> sup_0_eq_0
  5437     neg_meet_eq_join        ~> neg_inf_eq_sup
  5438     neg_join_eq_meet        ~> neg_sup_eq_inf
  5439     join_eq_if              ~> sup_eq_if
  5440 
  5441     mono_meet               ~> mono_inf
  5442     mono_join               ~> mono_sup
  5443     meet_bool_eq            ~> inf_bool_eq
  5444     join_bool_eq            ~> sup_bool_eq
  5445     meet_fun_eq             ~> inf_fun_eq
  5446     join_fun_eq             ~> sup_fun_eq
  5447     meet_set_eq             ~> inf_set_eq
  5448     join_set_eq             ~> sup_set_eq
  5449     meet1_iff               ~> inf1_iff
  5450     meet2_iff               ~> inf2_iff
  5451     meet1I                  ~> inf1I
  5452     meet2I                  ~> inf2I
  5453     meet1D1                 ~> inf1D1
  5454     meet2D1                 ~> inf2D1
  5455     meet1D2                 ~> inf1D2
  5456     meet2D2                 ~> inf2D2
  5457     meet1E                  ~> inf1E
  5458     meet2E                  ~> inf2E
  5459     join1_iff               ~> sup1_iff
  5460     join2_iff               ~> sup2_iff
  5461     join1I1                 ~> sup1I1
  5462     join2I1                 ~> sup2I1
  5463     join1I1                 ~> sup1I1
  5464     join2I2                 ~> sup1I2
  5465     join1CI                 ~> sup1CI
  5466     join2CI                 ~> sup2CI
  5467     join1E                  ~> sup1E
  5468     join2E                  ~> sup2E
  5469 
  5470     is_meet_Meet            ~> is_meet_Inf
  5471     Meet_bool_def           ~> Inf_bool_def
  5472     Meet_fun_def            ~> Inf_fun_def
  5473     Meet_greatest           ~> Inf_greatest
  5474     Meet_lower              ~> Inf_lower
  5475     Meet_set_def            ~> Inf_set_def
  5476 
  5477     Sup_def                 ~> Sup_Inf
  5478     Sup_bool_eq             ~> Sup_bool_def
  5479     Sup_fun_eq              ~> Sup_fun_def
  5480     Sup_set_eq              ~> Sup_set_def
  5481 
  5482     listsp_meetI            ~> listsp_infI
  5483     listsp_meet_eq          ~> listsp_inf_eq
  5484 
  5485     meet_min                ~> inf_min
  5486     join_max                ~> sup_max
  5487 
  5488 * Added syntactic class "size"; overloaded constant "size" now has
  5489 type "'a::size ==> bool"
  5490 
  5491 * Internal reorganisation of `size' of datatypes: size theorems
  5492 "foo.size" are no longer subsumed by "foo.simps" (but are still
  5493 simplification rules by default!); theorems "prod.size" now named
  5494 "*.size".
  5495 
  5496 * Class "div" now inherits from class "times" rather than "type".
  5497 INCOMPATIBILITY.
  5498 
  5499 * HOL/Finite_Set: "name-space" locales Lattice, Distrib_lattice,
  5500 Linorder etc.  have disappeared; operations defined in terms of
  5501 fold_set now are named Inf_fin, Sup_fin.  INCOMPATIBILITY.
  5502 
  5503 * HOL/Nat: neq0_conv no longer declared as iff.  INCOMPATIBILITY.
  5504 
  5505 * HOL-Word: New extensive library and type for generic, fixed size
  5506 machine words, with arithmetic, bit-wise, shifting and rotating
  5507 operations, reflection into int, nat, and bool lists, automation for
  5508 linear arithmetic (by automatic reflection into nat or int), including
  5509 lemmas on overflow and monotonicity.  Instantiated to all appropriate
  5510 arithmetic type classes, supporting automatic simplification of
  5511 numerals on all operations.
  5512 
  5513 * Library/Boolean_Algebra: locales for abstract boolean algebras.
  5514 
  5515 * Library/Numeral_Type: numbers as types, e.g. TYPE(32).
  5516 
  5517 * Code generator library theories:
  5518   - Code_Integer represents HOL integers by big integer literals in target
  5519     languages.
  5520   - Code_Char represents HOL characters by character literals in target
  5521     languages.
  5522   - Code_Char_chr like Code_Char, but also offers treatment of character
  5523     codes; includes Code_Integer.
  5524   - Executable_Set allows to generate code for finite sets using lists.
  5525   - Executable_Rat implements rational numbers as triples (sign, enumerator,
  5526     denominator).
  5527   - Executable_Real implements a subset of real numbers, namly those
  5528     representable by rational numbers.
  5529   - Efficient_Nat implements natural numbers by integers, which in general will
  5530     result in higher efficency; pattern matching with 0/Suc is eliminated;
  5531     includes Code_Integer.
  5532   - Code_Index provides an additional datatype index which is mapped to
  5533     target-language built-in integers.
  5534   - Code_Message provides an additional datatype message_string which is isomorphic to
  5535     strings; messages are mapped to target-language strings.
  5536 
  5537 * New package for inductive predicates
  5538 
  5539   An n-ary predicate p with m parameters z_1, ..., z_m can now be defined via
  5540 
  5541     inductive
  5542       p :: "U_1 => ... => U_m => T_1 => ... => T_n => bool"
  5543       for z_1 :: U_1 and ... and z_n :: U_m
  5544     where
  5545       rule_1: "... ==> p z_1 ... z_m t_1_1 ... t_1_n"
  5546     | ...
  5547 
  5548   with full support for type-inference, rather than
  5549 
  5550     consts s :: "U_1 => ... => U_m => (T_1 * ... * T_n) set"
  5551 
  5552     abbreviation p :: "U_1 => ... => U_m => T_1 => ... => T_n => bool"
  5553     where "p z_1 ... z_m x_1 ... x_n == (x_1, ..., x_n) : s z_1 ... z_m"
  5554 
  5555     inductive "s z_1 ... z_m"
  5556     intros
  5557       rule_1: "... ==> (t_1_1, ..., t_1_n) : s z_1 ... z_m"
  5558       ...
  5559 
  5560   For backward compatibility, there is a wrapper allowing inductive
  5561   sets to be defined with the new package via
  5562 
  5563     inductive_set
  5564       s :: "U_1 => ... => U_m => (T_1 * ... * T_n) set"
  5565       for z_1 :: U_1 and ... and z_n :: U_m
  5566     where
  5567       rule_1: "... ==> (t_1_1, ..., t_1_n) : s z_1 ... z_m"
  5568     | ...
  5569 
  5570   or
  5571 
  5572     inductive_set
  5573       s :: "U_1 => ... => U_m => (T_1 * ... * T_n) set"
  5574       and p :: "U_1 => ... => U_m => T_1 => ... => T_n => bool"
  5575       for z_1 :: U_1 and ... and z_n :: U_m
  5576     where
  5577       "p z_1 ... z_m x_1 ... x_n == (x_1, ..., x_n) : s z_1 ... z_m"
  5578     | rule_1: "... ==> p z_1 ... z_m t_1_1 ... t_1_n"
  5579     | ...
  5580 
  5581   if the additional syntax "p ..." is required.
  5582 
  5583   Numerous examples can be found in the subdirectories src/HOL/Auth,
  5584   src/HOL/Bali, src/HOL/Induct, and src/HOL/MicroJava.
  5585 
  5586   INCOMPATIBILITIES:
  5587 
  5588   - Since declaration and definition of inductive sets or predicates
  5589     is no longer separated, abbreviations involving the newly
  5590     introduced sets or predicates must be specified together with the
  5591     introduction rules after the 'where' keyword (see above), rather
  5592     than before the actual inductive definition.
  5593 
  5594   - The variables in induction and elimination rules are now
  5595     quantified in the order of their occurrence in the introduction
  5596     rules, rather than in alphabetical order. Since this may break
  5597     some proofs, these proofs either have to be repaired, e.g. by
  5598     reordering the variables a_i_1 ... a_i_{k_i} in Isar 'case'
  5599     statements of the form
  5600 
  5601       case (rule_i a_i_1 ... a_i_{k_i})
  5602 
  5603     or the old order of quantification has to be restored by explicitly adding
  5604     meta-level quantifiers in the introduction rules, i.e.
  5605 
  5606       | rule_i: "!!a_i_1 ... a_i_{k_i}. ... ==> p z_1 ... z_m t_i_1 ... t_i_n"
  5607 
  5608   - The format of the elimination rules is now
  5609 
  5610       p z_1 ... z_m x_1 ... x_n ==>
  5611         (!!a_1_1 ... a_1_{k_1}. x_1 = t_1_1 ==> ... ==> x_n = t_1_n ==> ... ==> P)
  5612         ==> ... ==> P
  5613 
  5614     for predicates and
  5615 
  5616       (x_1, ..., x_n) : s z_1 ... z_m ==>
  5617         (!!a_1_1 ... a_1_{k_1}. x_1 = t_1_1 ==> ... ==> x_n = t_1_n ==> ... ==> P)
  5618         ==> ... ==> P
  5619 
  5620     for sets rather than
  5621 
  5622       x : s z_1 ... z_m ==>
  5623         (!!a_1_1 ... a_1_{k_1}. x = (t_1_1, ..., t_1_n) ==> ... ==> P)
  5624         ==> ... ==> P
  5625 
  5626     This may require terms in goals to be expanded to n-tuples
  5627     (e.g. using case_tac or simplification with the split_paired_all
  5628     rule) before the above elimination rule is applicable.
  5629 
  5630   - The elimination or case analysis rules for (mutually) inductive
  5631     sets or predicates are now called "p_1.cases" ... "p_k.cases". The
  5632     list of rules "p_1_..._p_k.elims" is no longer available.
  5633 
  5634 * New package "function"/"fun" for general recursive functions,
  5635 supporting mutual and nested recursion, definitions in local contexts,
  5636 more general pattern matching and partiality. See HOL/ex/Fundefs.thy
  5637 for small examples, and the separate tutorial on the function
  5638 package. The old recdef "package" is still available as before, but
  5639 users are encouraged to use the new package.
  5640 
  5641 * Method "lexicographic_order" automatically synthesizes termination
  5642 relations as lexicographic combinations of size measures. 
  5643 
  5644 * Case-expressions allow arbitrary constructor-patterns (including
  5645 "_") and take their order into account, like in functional
  5646 programming.  Internally, this is translated into nested
  5647 case-expressions; missing cases are added and mapped to the predefined
  5648 constant "undefined". In complicated cases printing may no longer show
  5649 the original input but the internal form. Lambda-abstractions allow
  5650 the same form of pattern matching: "% pat1 => e1 | ..." is an
  5651 abbreviation for "%x. case x of pat1 => e1 | ..." where x is a new
  5652 variable.
  5653 
  5654 * IntDef: The constant "int :: nat => int" has been removed; now "int"
  5655 is an abbreviation for "of_nat :: nat => int". The simplification
  5656 rules for "of_nat" have been changed to work like "int" did
  5657 previously.  Potential INCOMPATIBILITY:
  5658   - "of_nat (Suc m)" simplifies to "1 + of_nat m" instead of "of_nat m + 1"
  5659   - of_nat_diff and of_nat_mult are no longer default simp rules
  5660 
  5661 * Method "algebra" solves polynomial equations over (semi)rings using
  5662 Groebner bases. The (semi)ring structure is defined by locales and the
  5663 tool setup depends on that generic context. Installing the method for
  5664 a specific type involves instantiating the locale and possibly adding
  5665 declarations for computation on the coefficients.  The method is
  5666 already instantiated for natural numbers and for the axiomatic class
  5667 of idoms with numerals.  See also the paper by Chaieb and Wenzel at
  5668 CALCULEMUS 2007 for the general principles underlying this
  5669 architecture of context-aware proof-tools.
  5670 
  5671 * Method "ferrack" implements quantifier elimination over
  5672 special-purpose dense linear orders using locales (analogous to
  5673 "algebra"). The method is already installed for class
  5674 {ordered_field,recpower,number_ring} which subsumes real, hyperreal,
  5675 rat, etc.
  5676 
  5677 * Former constant "List.op @" now named "List.append".  Use ML
  5678 antiquotations @{const_name List.append} or @{term " ... @ ... "} to
  5679 circumvent possible incompatibilities when working on ML level.
  5680 
  5681 * primrec: missing cases mapped to "undefined" instead of "arbitrary".
  5682 
  5683 * New function listsum :: 'a list => 'a for arbitrary monoids.
  5684 Special syntax: "SUM x <- xs. f x" (and latex variants)
  5685 
  5686 * New syntax for Haskell-like list comprehension (input only), eg.
  5687 [(x,y). x <- xs, y <- ys, x ~= y], see also src/HOL/List.thy.
  5688 
  5689 * The special syntax for function "filter" has changed from [x :
  5690 xs. P] to [x <- xs. P] to avoid an ambiguity caused by list
  5691 comprehension syntax, and for uniformity.  INCOMPATIBILITY.
  5692 
  5693 * [a..b] is now defined for arbitrary linear orders.  It used to be
  5694 defined on nat only, as an abbreviation for [a..<Suc b]
  5695 INCOMPATIBILITY.
  5696 
  5697 * Renamed lemma "set_take_whileD"  to "set_takeWhileD".
  5698 
  5699 * New functions "sorted" and "sort" in src/HOL/List.thy.
  5700 
  5701 * New lemma collection field_simps (an extension of ring_simps) for
  5702 manipulating (in)equations involving division. Multiplies with all
  5703 denominators that can be proved to be non-zero (in equations) or
  5704 positive/negative (in inequations).
  5705 
  5706 * Lemma collections ring_eq_simps, group_eq_simps and ring_distrib
  5707 have been improved and renamed to ring_simps, group_simps and
  5708 ring_distribs.  Removed lemmas field_xyz in theory Ring_and_Field
  5709 because they were subsumed by lemmas xyz.  INCOMPATIBILITY.
  5710 
  5711 * Theory Library/Commutative_Ring: switched from recdef to function
  5712 package; constants add, mul, pow now curried.  Infix syntax for
  5713 algebraic operations.
  5714 
  5715 * Dropped redundant lemma def_imp_eq in favor of meta_eq_to_obj_eq.
  5716 INCOMPATIBILITY.
  5717 
  5718 * Dropped redundant lemma if_def2 in favor of if_bool_eq_conj.
  5719 INCOMPATIBILITY.
  5720 
  5721 * HOL/records: generalised field-update to take a function on the
  5722 field rather than the new value: r(|A := x|) is translated to A_update
  5723 (K x) r The K-combinator that is internally used is called K_record.
  5724 INCOMPATIBILITY: Usage of the plain update functions has to be
  5725 adapted.
  5726  
  5727 * Class "semiring_0" now contains annihilation axioms x * 0 = 0 and 0
  5728 * x = 0, which are required for a semiring.  Richer structures do not
  5729 inherit from semiring_0 anymore, because this property is a theorem
  5730 there, not an axiom.  INCOMPATIBILITY: In instances of semiring_0,
  5731 there is more to prove, but this is mostly trivial.
  5732 
  5733 * Class "recpower" is generalized to arbitrary monoids, not just
  5734 commutative semirings.  INCOMPATIBILITY: may need to incorporate
  5735 commutativity or semiring properties additionally.
  5736 
  5737 * Constant "List.list_all2" in List.thy now uses authentic syntax.
  5738 INCOMPATIBILITY: translations containing list_all2 may go wrong,
  5739 better use 'abbreviation'.
  5740 
  5741 * Renamed constant "List.op mem" to "List.member".  INCOMPATIBILITY.
  5742 
  5743 * Numeral syntax: type 'bin' which was a mere type copy of 'int' has
  5744 been abandoned in favour of plain 'int'.  INCOMPATIBILITY --
  5745 significant changes for setting up numeral syntax for types:
  5746   - New constants Numeral.pred and Numeral.succ instead
  5747       of former Numeral.bin_pred and Numeral.bin_succ.
  5748   - Use integer operations instead of bin_add, bin_mult and so on.
  5749   - Numeral simplification theorems named Numeral.numeral_simps instead of Bin_simps.
  5750   - ML structure Bin_Simprocs now named Int_Numeral_Base_Simprocs.
  5751 
  5752 See src/HOL/Integ/IntArith.thy for an example setup.
  5753 
  5754 * Command 'normal_form' computes the normal form of a term that may
  5755 contain free variables.  For example ``normal_form "rev [a, b, c]"''
  5756 produces ``[b, c, a]'' (without proof).  This command is suitable for
  5757 heavy-duty computations because the functions are compiled to ML
  5758 first.  Correspondingly, a method "normalization" is provided.  See
  5759 further src/HOL/ex/NormalForm.thy and src/Tools/nbe.ML.
  5760 
  5761 * Alternative iff syntax "A <-> B" for equality on bool (with priority
  5762 25 like -->); output depends on the "iff" print_mode, the default is
  5763 "A = B" (with priority 50).
  5764 
  5765 * Relations less (<) and less_eq (<=) are also available on type bool.
  5766 Modified syntax to disallow nesting without explicit parentheses,
  5767 e.g. "(x < y) < z" or "x < (y < z)", but NOT "x < y < z".  Potential
  5768 INCOMPATIBILITY.
  5769 
  5770 * "LEAST x:A. P" expands to "LEAST x. x:A & P" (input only).
  5771 
  5772 * Relation composition operator "op O" now has precedence 75 and binds
  5773 stronger than union and intersection. INCOMPATIBILITY.
  5774 
  5775 * The old set interval syntax "{m..n(}" (and relatives) has been
  5776 removed.  Use "{m..<n}" (and relatives) instead.
  5777 
  5778 * In the context of the assumption "~(s = t)" the Simplifier rewrites
  5779 "t = s" to False (by simproc "neq").  INCOMPATIBILITY, consider using
  5780 ``declare [[simproc del: neq]]''.
  5781 
  5782 * Simplifier: "m dvd n" where m and n are numbers is evaluated to
  5783 True/False.
  5784 
  5785 * Theorem Cons_eq_map_conv no longer declared as "simp".
  5786 
  5787 * Theorem setsum_mult renamed to setsum_right_distrib.
  5788 
  5789 * Prefer ex1I over ex_ex1I in single-step reasoning, e.g. by the
  5790 ``rule'' method.
  5791 
  5792 * Reimplemented methods "sat" and "satx", with several improvements:
  5793 goals no longer need to be stated as "<prems> ==> False", equivalences
  5794 (i.e. "=" on type bool) are handled, variable names of the form
  5795 "lit_<n>" are no longer reserved, significant speedup.
  5796 
  5797 * Methods "sat" and "satx" can now replay MiniSat proof traces.
  5798 zChaff is still supported as well.
  5799 
  5800 * 'inductive' and 'datatype': provide projections of mutual rules,
  5801 bundled as foo_bar.inducts;
  5802 
  5803 * Library: moved theories Parity, GCD, Binomial, Infinite_Set to
  5804 Library.
  5805 
  5806 * Library: moved theory Accessible_Part to main HOL.
  5807 
  5808 * Library: added theory Coinductive_List of potentially infinite lists
  5809 as greatest fixed-point.
  5810 
  5811 * Library: added theory AssocList which implements (finite) maps as
  5812 association lists.
  5813 
  5814 * Method "evaluation" solves goals (i.e. a boolean expression)
  5815 efficiently by compiling it to ML.  The goal is "proved" (via an
  5816 oracle) if it evaluates to True.
  5817 
  5818 * Linear arithmetic now splits certain operators (e.g. min, max, abs)
  5819 also when invoked by the simplifier.  This results in the Simplifier
  5820 being more powerful on arithmetic goals.  INCOMPATIBILITY.
  5821 Configuration option fast_arith_split_limit=0 recovers the old
  5822 behavior.
  5823 
  5824 * Support for hex (0x20) and binary (0b1001) numerals.
  5825 
  5826 * New method: reify eqs (t), where eqs are equations for an
  5827 interpretation I :: 'a list => 'b => 'c and t::'c is an optional
  5828 parameter, computes a term s::'b and a list xs::'a list and proves the
  5829 theorem I xs s = t. This is also known as reification or quoting. The
  5830 resulting theorem is applied to the subgoal to substitute t with I xs
  5831 s.  If t is omitted, the subgoal itself is reified.
  5832 
  5833 * New method: reflection corr_thm eqs (t). The parameters eqs and (t)
  5834 are as explained above. corr_thm is a theorem for I vs (f t) = I vs t,
  5835 where f is supposed to be a computable function (in the sense of code
  5836 generattion). The method uses reify to compute s and xs as above then
  5837 applies corr_thm and uses normalization by evaluation to "prove" f s =
  5838 r and finally gets the theorem t = r, which is again applied to the
  5839 subgoal. An Example is available in src/HOL/ex/ReflectionEx.thy.
  5840 
  5841 * Reflection: Automatic reification now handels binding, an example is
  5842 available in src/HOL/ex/ReflectionEx.thy
  5843 
  5844 * HOL-Statespace: ``State Spaces: The Locale Way'' introduces a
  5845 command 'statespace' that is similar to 'record', but introduces an
  5846 abstract specification based on the locale infrastructure instead of
  5847 HOL types.  This leads to extra flexibility in composing state spaces,
  5848 in particular multiple inheritance and renaming of components.
  5849 
  5850 
  5851 *** HOL-Complex ***
  5852 
  5853 * Hyperreal: Functions root and sqrt are now defined on negative real
  5854 inputs so that root n (- x) = - root n x and sqrt (- x) = - sqrt x.
  5855 Nonnegativity side conditions have been removed from many lemmas, so
  5856 that more subgoals may now be solved by simplification; potential
  5857 INCOMPATIBILITY.
  5858 
  5859 * Real: new type classes formalize real normed vector spaces and
  5860 algebras, using new overloaded constants scaleR :: real => 'a => 'a
  5861 and norm :: 'a => real.
  5862 
  5863 * Real: constant of_real :: real => 'a::real_algebra_1 injects from
  5864 reals into other types. The overloaded constant Reals :: 'a set is now
  5865 defined as range of_real; potential INCOMPATIBILITY.
  5866 
  5867 * Real: proper support for ML code generation, including 'quickcheck'.
  5868 Reals are implemented as arbitrary precision rationals.
  5869 
  5870 * Hyperreal: Several constants that previously worked only for the
  5871 reals have been generalized, so they now work over arbitrary vector
  5872 spaces. Type annotations may need to be added in some cases; potential
  5873 INCOMPATIBILITY.
  5874 
  5875   Infinitesimal  :: ('a::real_normed_vector) star set
  5876   HFinite        :: ('a::real_normed_vector) star set
  5877   HInfinite      :: ('a::real_normed_vector) star set
  5878   approx         :: ('a::real_normed_vector) star => 'a star => bool
  5879   monad          :: ('a::real_normed_vector) star => 'a star set
  5880   galaxy         :: ('a::real_normed_vector) star => 'a star set
  5881   (NS)LIMSEQ     :: [nat => 'a::real_normed_vector, 'a] => bool
  5882   (NS)convergent :: (nat => 'a::real_normed_vector) => bool
  5883   (NS)Bseq       :: (nat => 'a::real_normed_vector) => bool
  5884   (NS)Cauchy     :: (nat => 'a::real_normed_vector) => bool
  5885   (NS)LIM        :: ['a::real_normed_vector => 'b::real_normed_vector, 'a, 'b] => bool
  5886   is(NS)Cont     :: ['a::real_normed_vector => 'b::real_normed_vector, 'a] => bool
  5887   deriv          :: ['a::real_normed_field => 'a, 'a, 'a] => bool
  5888   sgn            :: 'a::real_normed_vector => 'a
  5889   exp            :: 'a::{recpower,real_normed_field,banach} => 'a
  5890 
  5891 * Complex: Some complex-specific constants are now abbreviations for
  5892 overloaded ones: complex_of_real = of_real, cmod = norm, hcmod =
  5893 hnorm.  Other constants have been entirely removed in favor of the
  5894 polymorphic versions (INCOMPATIBILITY):
  5895 
  5896   approx        <-- capprox
  5897   HFinite       <-- CFinite
  5898   HInfinite     <-- CInfinite
  5899   Infinitesimal <-- CInfinitesimal
  5900   monad         <-- cmonad
  5901   galaxy        <-- cgalaxy
  5902   (NS)LIM       <-- (NS)CLIM, (NS)CRLIM
  5903   is(NS)Cont    <-- is(NS)Contc, is(NS)contCR
  5904   (ns)deriv     <-- (ns)cderiv
  5905 
  5906 
  5907 *** HOL-Algebra ***
  5908 
  5909 * Formalisation of ideals and the quotient construction over rings.
  5910 
  5911 * Order and lattice theory no longer based on records.
  5912 INCOMPATIBILITY.
  5913 
  5914 * Renamed lemmas least_carrier -> least_closed and greatest_carrier ->
  5915 greatest_closed.  INCOMPATIBILITY.
  5916 
  5917 * Method algebra is now set up via an attribute.  For examples see
  5918 Ring.thy.  INCOMPATIBILITY: the method is now weaker on combinations
  5919 of algebraic structures.
  5920 
  5921 * Renamed theory CRing to Ring.
  5922 
  5923 
  5924 *** HOL-Nominal ***
  5925 
  5926 * Substantial, yet incomplete support for nominal datatypes (binding
  5927 structures) based on HOL-Nominal logic.  See src/HOL/Nominal and
  5928 src/HOL/Nominal/Examples.  Prospective users should consult
  5929 http://isabelle.in.tum.de/nominal/
  5930 
  5931 
  5932 *** ML ***
  5933 
  5934 * ML basics: just one true type int, which coincides with IntInf.int
  5935 (even on SML/NJ).
  5936 
  5937 * ML within Isar: antiquotations allow to embed statically-checked
  5938 formal entities in the source, referring to the context available at
  5939 compile-time.  For example:
  5940 
  5941 ML {* @{sort "{zero,one}"} *}
  5942 ML {* @{typ "'a => 'b"} *}
  5943 ML {* @{term "%x. x"} *}
  5944 ML {* @{prop "x == y"} *}
  5945 ML {* @{ctyp "'a => 'b"} *}
  5946 ML {* @{cterm "%x. x"} *}
  5947 ML {* @{cprop "x == y"} *}
  5948 ML {* @{thm asm_rl} *}
  5949 ML {* @{thms asm_rl} *}
  5950 ML {* @{type_name c} *}
  5951 ML {* @{type_syntax c} *}
  5952 ML {* @{const_name c} *}
  5953 ML {* @{const_syntax c} *}
  5954 ML {* @{context} *}
  5955 ML {* @{theory} *}
  5956 ML {* @{theory Pure} *}
  5957 ML {* @{theory_ref} *}
  5958 ML {* @{theory_ref Pure} *}
  5959 ML {* @{simpset} *}
  5960 ML {* @{claset} *}
  5961 ML {* @{clasimpset} *}
  5962 
  5963 The same works for sources being ``used'' within an Isar context.
  5964 
  5965 * ML in Isar: improved error reporting; extra verbosity with
  5966 ML_Context.trace enabled.
  5967 
  5968 * Pure/General/table.ML: the join operations now works via exceptions
  5969 DUP/SAME instead of type option. This is simpler in simple cases, and
  5970 admits slightly more efficient complex applications.
  5971 
  5972 * Pure: 'advanced' translation functions (parse_translation etc.) now
  5973 use Context.generic instead of just theory.
  5974 
  5975 * Pure: datatype Context.generic joins theory/Proof.context and
  5976 provides some facilities for code that works in either kind of
  5977 context, notably GenericDataFun for uniform theory and proof data.
  5978 
  5979 * Pure: simplified internal attribute type, which is now always
  5980 Context.generic * thm -> Context.generic * thm. Global (theory) vs.
  5981 local (Proof.context) attributes have been discontinued, while
  5982 minimizing code duplication. Thm.rule_attribute and
  5983 Thm.declaration_attribute build canonical attributes; see also structure
  5984 Context for further operations on Context.generic, notably
  5985 GenericDataFun. INCOMPATIBILITY, need to adapt attribute type
  5986 declarations and definitions.
  5987 
  5988 * Context data interfaces (Theory/Proof/GenericDataFun): removed
  5989 name/print, uninitialized data defaults to ad-hoc copy of empty value,
  5990 init only required for impure data. INCOMPATIBILITY: empty really need
  5991 to be empty (no dependencies on theory content!)
  5992 
  5993 * Pure/kernel: consts certification ignores sort constraints given in
  5994 signature declarations. (This information is not relevant to the
  5995 logic, but only for type inference.) SIGNIFICANT INTERNAL CHANGE,
  5996 potential INCOMPATIBILITY.
  5997 
  5998 * Pure: axiomatic type classes are now purely definitional, with
  5999 explicit proofs of class axioms and super class relations performed
  6000 internally. See Pure/axclass.ML for the main internal interfaces --
  6001 notably AxClass.define_class supercedes AxClass.add_axclass, and
  6002 AxClass.axiomatize_class/classrel/arity supersede
  6003 Sign.add_classes/classrel/arities.
  6004 
  6005 * Pure/Isar: Args/Attrib parsers operate on Context.generic --
  6006 global/local versions on theory vs. Proof.context have been
  6007 discontinued; Attrib.syntax and Method.syntax have been adapted
  6008 accordingly.  INCOMPATIBILITY, need to adapt parser expressions for
  6009 attributes, methods, etc.
  6010 
  6011 * Pure: several functions of signature "... -> theory -> theory * ..."
  6012 have been reoriented to "... -> theory -> ... * theory" in order to
  6013 allow natural usage in combination with the ||>, ||>>, |-> and
  6014 fold_map combinators.
  6015 
  6016 * Pure: official theorem names (closed derivations) and additional
  6017 comments (tags) are now strictly separate.  Name hints -- which are
  6018 maintained as tags -- may be attached any time without affecting the
  6019 derivation.
  6020 
  6021 * Pure: primitive rule lift_rule now takes goal cterm instead of an
  6022 actual goal state (thm).  Use Thm.lift_rule (Thm.cprem_of st i) to
  6023 achieve the old behaviour.
  6024 
  6025 * Pure: the "Goal" constant is now called "prop", supporting a
  6026 slightly more general idea of ``protecting'' meta-level rule
  6027 statements.
  6028 
  6029 * Pure: Logic.(un)varify only works in a global context, which is now
  6030 enforced instead of silently assumed.  INCOMPATIBILITY, may use
  6031 Logic.legacy_(un)varify as temporary workaround.
  6032 
  6033 * Pure: structure Name provides scalable operations for generating
  6034 internal variable names, notably Name.variants etc.  This replaces
  6035 some popular functions from term.ML:
  6036 
  6037   Term.variant		->  Name.variant
  6038   Term.variantlist	->  Name.variant_list
  6039   Term.invent_names	->  Name.invent_list
  6040 
  6041 Note that low-level renaming rarely occurs in new code -- operations
  6042 from structure Variable are used instead (see below).
  6043 
  6044 * Pure: structure Variable provides fundamental operations for proper
  6045 treatment of fixed/schematic variables in a context.  For example,
  6046 Variable.import introduces fixes for schematics of given facts and
  6047 Variable.export reverses the effect (up to renaming) -- this replaces
  6048 various freeze_thaw operations.
  6049 
  6050 * Pure: structure Goal provides simple interfaces for
  6051 init/conclude/finish and tactical prove operations (replacing former
  6052 Tactic.prove).  Goal.prove is the canonical way to prove results
  6053 within a given context; Goal.prove_global is a degraded version for
  6054 theory level goals, including a global Drule.standard.  Note that
  6055 OldGoals.prove_goalw_cterm has long been obsolete, since it is
  6056 ill-behaved in a local proof context (e.g. with local fixes/assumes or
  6057 in a locale context).
  6058 
  6059 * Pure/Syntax: generic interfaces for parsing (Syntax.parse_term etc.)
  6060 and type checking (Syntax.check_term etc.), with common combinations
  6061 (Syntax.read_term etc.). These supersede former Sign.read_term etc.
  6062 which are considered legacy and await removal.
  6063 
  6064 * Pure/Syntax: generic interfaces for type unchecking
  6065 (Syntax.uncheck_terms etc.) and unparsing (Syntax.unparse_term etc.),
  6066 with common combinations (Syntax.pretty_term, Syntax.string_of_term
  6067 etc.).  Former Sign.pretty_term, Sign.string_of_term etc. are still
  6068 available for convenience, but refer to the very same operations using
  6069 a mere theory instead of a full context.
  6070 
  6071 * Isar: simplified treatment of user-level errors, using exception
  6072 ERROR of string uniformly.  Function error now merely raises ERROR,
  6073 without any side effect on output channels.  The Isar toplevel takes
  6074 care of proper display of ERROR exceptions.  ML code may use plain
  6075 handle/can/try; cat_error may be used to concatenate errors like this:
  6076 
  6077   ... handle ERROR msg => cat_error msg "..."
  6078 
  6079 Toplevel ML code (run directly or through the Isar toplevel) may be
  6080 embedded into the Isar toplevel with exception display/debug like
  6081 this:
  6082 
  6083   Isar.toplevel (fn () => ...)
  6084 
  6085 INCOMPATIBILITY, removed special transform_error facilities, removed
  6086 obsolete variants of user-level exceptions (ERROR_MESSAGE,
  6087 Context.PROOF, ProofContext.CONTEXT, Proof.STATE, ProofHistory.FAIL)
  6088 -- use plain ERROR instead.
  6089 
  6090 * Isar: theory setup now has type (theory -> theory), instead of a
  6091 list.  INCOMPATIBILITY, may use #> to compose setup functions.
  6092 
  6093 * Isar: ML toplevel pretty printer for type Proof.context, subject to
  6094 ProofContext.debug/verbose flags.
  6095 
  6096 * Isar: Toplevel.theory_to_proof admits transactions that modify the
  6097 theory before entering a proof state.  Transactions now always see a
  6098 quasi-functional intermediate checkpoint, both in interactive and
  6099 batch mode.
  6100 
  6101 * Isar: simplified interfaces for outer syntax.  Renamed
  6102 OuterSyntax.add_keywords to OuterSyntax.keywords.  Removed
  6103 OuterSyntax.add_parsers -- this functionality is now included in
  6104 OuterSyntax.command etc.  INCOMPATIBILITY.
  6105 
  6106 * Simplifier: the simpset of a running simplification process now
  6107 contains a proof context (cf. Simplifier.the_context), which is the
  6108 very context that the initial simpset has been retrieved from (by
  6109 simpset_of/local_simpset_of).  Consequently, all plug-in components
  6110 (solver, looper etc.) may depend on arbitrary proof data.
  6111 
  6112 * Simplifier.inherit_context inherits the proof context (plus the
  6113 local bounds) of the current simplification process; any simproc
  6114 etc. that calls the Simplifier recursively should do this!  Removed
  6115 former Simplifier.inherit_bounds, which is already included here --
  6116 INCOMPATIBILITY.  Tools based on low-level rewriting may even have to
  6117 specify an explicit context using Simplifier.context/theory_context.
  6118 
  6119 * Simplifier/Classical Reasoner: more abstract interfaces
  6120 change_simpset/claset for modifying the simpset/claset reference of a
  6121 theory; raw versions simpset/claset_ref etc. have been discontinued --
  6122 INCOMPATIBILITY.
  6123 
  6124 * Provers: more generic wrt. syntax of object-logics, avoid hardwired
  6125 "Trueprop" etc.
  6126 
  6127 
  6128 *** System ***
  6129 
  6130 * settings: the default heap location within ISABELLE_HOME_USER now
  6131 includes ISABELLE_IDENTIFIER.  This simplifies use of multiple
  6132 Isabelle installations.
  6133 
  6134 * isabelle-process: option -S (secure mode) disables some critical
  6135 operations, notably runtime compilation and evaluation of ML source
  6136 code.
  6137 
  6138 * Basic Isabelle mode for jEdit, see Isabelle/lib/jedit/.
  6139 
  6140 * Support for parallel execution, using native multicore support of
  6141 Poly/ML 5.1.  The theory loader exploits parallelism when processing
  6142 independent theories, according to the given theory header
  6143 specifications. The maximum number of worker threads is specified via
  6144 usedir option -M or the "max-threads" setting in Proof General. A
  6145 speedup factor of 1.5--3.5 can be expected on a 4-core machine, and up
  6146 to 6 on a 8-core machine.  User-code needs to observe certain
  6147 guidelines for thread-safe programming, see appendix A in the Isar
  6148 Implementation manual.
  6149 
  6150 
  6151 
  6152 New in Isabelle2005 (October 2005)
  6153 ----------------------------------
  6154 
  6155 *** General ***
  6156 
  6157 * Theory headers: the new header syntax for Isar theories is
  6158 
  6159   theory <name>
  6160   imports <theory1> ... <theoryN>
  6161   uses <file1> ... <fileM>
  6162   begin
  6163 
  6164 where the 'uses' part is optional.  The previous syntax
  6165 
  6166   theory <name> = <theory1> + ... + <theoryN>:
  6167 
  6168 will disappear in the next release.  Use isatool fixheaders to convert
  6169 existing theory files.  Note that there is no change in ancient
  6170 non-Isar theories now, but these will disappear soon.
  6171 
  6172 * Theory loader: parent theories can now also be referred to via
  6173 relative and absolute paths.
  6174 
  6175 * Command 'find_theorems' searches for a list of criteria instead of a
  6176 list of constants. Known criteria are: intro, elim, dest, name:string,
  6177 simp:term, and any term. Criteria can be preceded by '-' to select
  6178 theorems that do not match. Intro, elim, dest select theorems that
  6179 match the current goal, name:s selects theorems whose fully qualified
  6180 name contain s, and simp:term selects all simplification rules whose
  6181 lhs match term.  Any other term is interpreted as pattern and selects
  6182 all theorems matching the pattern. Available in ProofGeneral under
  6183 'ProofGeneral -> Find Theorems' or C-c C-f.  Example:
  6184 
  6185   C-c C-f (100) "(_::nat) + _ + _" intro -name: "HOL."
  6186 
  6187 prints the last 100 theorems matching the pattern "(_::nat) + _ + _",
  6188 matching the current goal as introduction rule and not having "HOL."
  6189 in their name (i.e. not being defined in theory HOL).
  6190 
  6191 * Command 'thms_containing' has been discontinued in favour of
  6192 'find_theorems'; INCOMPATIBILITY.
  6193 
  6194 * Communication with Proof General is now 8bit clean, which means that
  6195 Unicode text in UTF-8 encoding may be used within theory texts (both
  6196 formal and informal parts).  Cf. option -U of the Isabelle Proof
  6197 General interface.  Here are some simple examples (cf. src/HOL/ex):
  6198 
  6199   http://isabelle.in.tum.de/library/HOL/ex/Hebrew.html
  6200   http://isabelle.in.tum.de/library/HOL/ex/Chinese.html
  6201 
  6202 * Improved efficiency of the Simplifier and, to a lesser degree, the
  6203 Classical Reasoner.  Typical big applications run around 2 times
  6204 faster.
  6205 
  6206 
  6207 *** Document preparation ***
  6208 
  6209 * Commands 'display_drafts' and 'print_drafts' perform simple output
  6210 of raw sources.  Only those symbols that do not require additional
  6211 LaTeX packages (depending on comments in isabellesym.sty) are
  6212 displayed properly, everything else is left verbatim.  isatool display
  6213 and isatool print are used as front ends (these are subject to the
  6214 DVI/PDF_VIEWER and PRINT_COMMAND settings, respectively).
  6215 
  6216 * Command tags control specific markup of certain regions of text,
  6217 notably folding and hiding.  Predefined tags include "theory" (for
  6218 theory begin and end), "proof" for proof commands, and "ML" for
  6219 commands involving ML code; the additional tags "visible" and
  6220 "invisible" are unused by default.  Users may give explicit tag
  6221 specifications in the text, e.g. ''by %invisible (auto)''.  The
  6222 interpretation of tags is determined by the LaTeX job during document
  6223 preparation: see option -V of isatool usedir, or options -n and -t of
  6224 isatool document, or even the LaTeX macros \isakeeptag, \isafoldtag,
  6225 \isadroptag.
  6226 
  6227 Several document versions may be produced at the same time via isatool
  6228 usedir (the generated index.html will link all of them).  Typical
  6229 specifications include ''-V document=theory,proof,ML'' to present
  6230 theory/proof/ML parts faithfully, ''-V outline=/proof,/ML'' to fold
  6231 proof and ML commands, and ''-V mutilated=-theory,-proof,-ML'' to omit
  6232 these parts without any formal replacement text.  The Isabelle site
  6233 default settings produce ''document'' and ''outline'' versions as
  6234 specified above.