NEWS
author wenzelm
Wed May 23 12:02:27 2012 +0200 (2012-05-23)
changeset 47958 c5f7be4a1734
parent 47887 4e9c06c194d9
parent 47866 2cc26ddd8298
child 47967 c422128d3889
permissions -rw-r--r--
merged, abandoning change of src/HOL/Tools/ATP/atp_problem_generate.ML from 6ea205a4d7fd;
     1 Isabelle NEWS -- history user-relevant changes
     2 ==============================================
     3 
     4 New in this Isabelle version
     5 ----------------------------
     6 
     7 
     8 New in Isabelle2012 (May 2012)
     9 ------------------------------
    10 
    11 *** General ***
    12 
    13 * Prover IDE (PIDE) improvements:
    14 
    15   - more robust Sledgehammer integration (as before the sledgehammer
    16     command-line needs to be typed into the source buffer)
    17   - markup for bound variables
    18   - markup for types of term variables (displayed as tooltips)
    19   - support for user-defined Isar commands within the running session
    20   - improved support for Unicode outside original 16bit range
    21     e.g. glyph for \<A> (thanks to jEdit 4.5.1)
    22 
    23 * Forward declaration of outer syntax keywords within the theory
    24 header -- minor INCOMPATIBILITY for user-defined commands.  Allow new
    25 commands to be used in the same theory where defined.
    26 
    27 * Auxiliary contexts indicate block structure for specifications with
    28 additional parameters and assumptions.  Such unnamed contexts may be
    29 nested within other targets, like 'theory', 'locale', 'class',
    30 'instantiation' etc.  Results from the local context are generalized
    31 accordingly and applied to the enclosing target context.  Example:
    32 
    33   context
    34     fixes x y z :: 'a
    35     assumes xy: "x = y" and yz: "y = z"
    36   begin
    37 
    38   lemma my_trans: "x = z" using xy yz by simp
    39 
    40   end
    41 
    42   thm my_trans
    43 
    44 The most basic application is to factor-out context elements of
    45 several fixes/assumes/shows theorem statements, e.g. see
    46 ~~/src/HOL/Isar_Examples/Group_Context.thy
    47 
    48 Any other local theory specification element works within the "context
    49 ... begin ... end" block as well.
    50 
    51 * Bundled declarations associate attributed fact expressions with a
    52 given name in the context.  These may be later included in other
    53 contexts.  This allows to manage context extensions casually, without
    54 the logical dependencies of locales and locale interpretation.  See
    55 commands 'bundle', 'include', 'including' etc. in the isar-ref manual.
    56 
    57 * Commands 'lemmas' and 'theorems' allow local variables using 'for'
    58 declaration, and results are standardized before being stored.  Thus
    59 old-style "standard" after instantiation or composition of facts
    60 becomes obsolete.  Minor INCOMPATIBILITY, due to potential change of
    61 indices of schematic variables.
    62 
    63 * Rule attributes in local theory declarations (e.g. locale or class)
    64 are now statically evaluated: the resulting theorem is stored instead
    65 of the original expression.  INCOMPATIBILITY in rare situations, where
    66 the historic accident of dynamic re-evaluation in interpretations
    67 etc. was exploited.
    68 
    69 * New tutorial "Programming and Proving in Isabelle/HOL"
    70 ("prog-prove").  It completely supersedes "A Tutorial Introduction to
    71 Structured Isar Proofs" ("isar-overview"), which has been removed.  It
    72 also supersedes "Isabelle/HOL, A Proof Assistant for Higher-Order
    73 Logic" as the recommended beginners tutorial, but does not cover all
    74 of the material of that old tutorial.
    75 
    76 * Updated and extended reference manuals: "isar-ref",
    77 "implementation", "system"; reduced remaining material in old "ref"
    78 manual.
    79 
    80 
    81 *** Pure ***
    82 
    83 * Command 'definition' no longer exports the foundational "raw_def"
    84 into the user context.  Minor INCOMPATIBILITY, may use the regular
    85 "def" result with attribute "abs_def" to imitate the old version.
    86 
    87 * Attribute "abs_def" turns an equation of the form "f x y == t" into
    88 "f == %x y. t", which ensures that "simp" or "unfold" steps always
    89 expand it.  This also works for object-logic equality.  (Formerly
    90 undocumented feature.)
    91 
    92 * Sort constraints are now propagated in simultaneous statements, just
    93 like type constraints.  INCOMPATIBILITY in rare situations, where
    94 distinct sorts used to be assigned accidentally.  For example:
    95 
    96   lemma "P (x::'a::foo)" and "Q (y::'a::bar)"  -- "now illegal"
    97 
    98   lemma "P (x::'a)" and "Q (y::'a::bar)"
    99     -- "now uniform 'a::bar instead of default sort for first occurrence (!)"
   100 
   101 * Rule composition via attribute "OF" (or ML functions OF/MRS) is more
   102 tolerant against multiple unifiers, as long as the final result is
   103 unique.  (As before, rules are composed in canonical right-to-left
   104 order to accommodate newly introduced premises.)
   105 
   106 * Renamed some inner syntax categories:
   107 
   108     num ~> num_token
   109     xnum ~> xnum_token
   110     xstr ~> str_token
   111 
   112 Minor INCOMPATIBILITY.  Note that in practice "num_const" or
   113 "num_position" etc. are mainly used instead (which also include
   114 position information via constraints).
   115 
   116 * Simplified configuration options for syntax ambiguity: see
   117 "syntax_ambiguity_warning" and "syntax_ambiguity_limit" in isar-ref
   118 manual.  Minor INCOMPATIBILITY.
   119 
   120 * Discontinued configuration option "syntax_positions": atomic terms
   121 in parse trees are always annotated by position constraints.
   122 
   123 * Old code generator for SML and its commands 'code_module',
   124 'code_library', 'consts_code', 'types_code' have been discontinued.
   125 Use commands of the generic code generator instead.  INCOMPATIBILITY.
   126 
   127 * Redundant attribute "code_inline" has been discontinued. Use
   128 "code_unfold" instead.  INCOMPATIBILITY.
   129 
   130 * Dropped attribute "code_unfold_post" in favor of the its dual
   131 "code_abbrev", which yields a common pattern in definitions like
   132 
   133   definition [code_abbrev]: "f = t"
   134 
   135 INCOMPATIBILITY.
   136 
   137 * Obsolete 'types' command has been discontinued.  Use 'type_synonym'
   138 instead.  INCOMPATIBILITY.
   139 
   140 * Discontinued old "prems" fact, which used to refer to the accidental
   141 collection of foundational premises in the context (already marked as
   142 legacy since Isabelle2011).
   143 
   144 
   145 *** HOL ***
   146 
   147 * Type 'a set is now a proper type constructor (just as before
   148 Isabelle2008).  Definitions mem_def and Collect_def have disappeared.
   149 Non-trivial INCOMPATIBILITY.  For developments keeping predicates and
   150 sets separate, it is often sufficient to rephrase some set S that has
   151 been accidentally used as predicates by "%x. x : S", and some
   152 predicate P that has been accidentally used as set by "{x. P x}".
   153 Corresponding proofs in a first step should be pruned from any
   154 tinkering with former theorems mem_def and Collect_def as far as
   155 possible.
   156 
   157 For developments which deliberately mix predicates and sets, a
   158 planning step is necessary to determine what should become a predicate
   159 and what a set.  It can be helpful to carry out that step in
   160 Isabelle2011-1 before jumping right into the current release.
   161 
   162 * Code generation by default implements sets as container type rather
   163 than predicates.  INCOMPATIBILITY.
   164 
   165 * New type synonym 'a rel = ('a * 'a) set
   166 
   167 * The representation of numerals has changed.  Datatype "num"
   168 represents strictly positive binary numerals, along with functions
   169 "numeral :: num => 'a" and "neg_numeral :: num => 'a" to represent
   170 positive and negated numeric literals, respectively.  See also
   171 definitions in ~~/src/HOL/Num.thy.  Potential INCOMPATIBILITY, some
   172 user theories may require adaptations as follows:
   173 
   174   - Theorems with number_ring or number_semiring constraints: These
   175     classes are gone; use comm_ring_1 or comm_semiring_1 instead.
   176 
   177   - Theories defining numeric types: Remove number, number_semiring,
   178     and number_ring instances. Defer all theorems about numerals until
   179     after classes one and semigroup_add have been instantiated.
   180 
   181   - Numeral-only simp rules: Replace each rule having a "number_of v"
   182     pattern with two copies, one for numeral and one for neg_numeral.
   183 
   184   - Theorems about subclasses of semiring_1 or ring_1: These classes
   185     automatically support numerals now, so more simp rules and
   186     simprocs may now apply within the proof.
   187 
   188   - Definitions and theorems using old constructors Pls/Min/Bit0/Bit1:
   189     Redefine using other integer operations.
   190 
   191 * Transfer: New package intended to generalize the existing
   192 "descending" method and related theorem attributes from the Quotient
   193 package.  (Not all functionality is implemented yet, but future
   194 development will focus on Transfer as an eventual replacement for the
   195 corresponding parts of the Quotient package.)
   196 
   197   - transfer_rule attribute: Maintains a collection of transfer rules,
   198     which relate constants at two different types. Transfer rules may
   199     relate different type instances of the same polymorphic constant,
   200     or they may relate an operation on a raw type to a corresponding
   201     operation on an abstract type (quotient or subtype). For example:
   202 
   203     ((A ===> B) ===> list_all2 A ===> list_all2 B) map map
   204     (cr_int ===> cr_int ===> cr_int) (%(x,y) (u,v). (x+u, y+v)) plus_int
   205 
   206   - transfer method: Replaces a subgoal on abstract types with an
   207     equivalent subgoal on the corresponding raw types. Constants are
   208     replaced with corresponding ones according to the transfer rules.
   209     Goals are generalized over all free variables by default; this is
   210     necessary for variables whose types change, but can be overridden
   211     for specific variables with e.g. "transfer fixing: x y z".  The
   212     variant transfer' method allows replacing a subgoal with one that
   213     is logically stronger (rather than equivalent).
   214 
   215   - relator_eq attribute: Collects identity laws for relators of
   216     various type constructors, e.g. "list_all2 (op =) = (op =)".  The
   217     transfer method uses these lemmas to infer transfer rules for
   218     non-polymorphic constants on the fly.
   219 
   220   - transfer_prover method: Assists with proving a transfer rule for a
   221     new constant, provided the constant is defined in terms of other
   222     constants that already have transfer rules. It should be applied
   223     after unfolding the constant definitions.
   224 
   225   - HOL/ex/Transfer_Int_Nat.thy: Example theory demonstrating transfer
   226     from type nat to type int.
   227 
   228 * Lifting: New package intended to generalize the quotient_definition
   229 facility of the Quotient package; designed to work with Transfer.
   230 
   231   - lift_definition command: Defines operations on an abstract type in
   232     terms of a corresponding operation on a representation
   233     type.  Example syntax:
   234 
   235     lift_definition dlist_insert :: "'a => 'a dlist => 'a dlist"
   236       is List.insert
   237 
   238     Users must discharge a respectfulness proof obligation when each
   239     constant is defined. (For a type copy, i.e. a typedef with UNIV,
   240     the proof is discharged automatically.) The obligation is
   241     presented in a user-friendly, readable form; a respectfulness
   242     theorem in the standard format and a transfer rule are generated
   243     by the package.
   244 
   245   - Integration with code_abstype: For typedefs (e.g. subtypes
   246     corresponding to a datatype invariant, such as dlist),
   247     lift_definition generates a code certificate theorem and sets up
   248     code generation for each constant.
   249 
   250   - setup_lifting command: Sets up the Lifting package to work with a
   251     user-defined type. The user must provide either a quotient theorem
   252     or a type_definition theorem.  The package configures transfer
   253     rules for equality and quantifiers on the type, and sets up the
   254     lift_definition command to work with the type.
   255 
   256   - Usage examples: See Quotient_Examples/Lift_DList.thy,
   257     Quotient_Examples/Lift_RBT.thy, Quotient_Examples/Lift_FSet.thy,
   258     Word/Word.thy and Library/Float.thy.
   259 
   260 * Quotient package:
   261 
   262   - The 'quotient_type' command now supports a 'morphisms' option with
   263     rep and abs functions, similar to typedef.
   264 
   265   - 'quotient_type' sets up new types to work with the Lifting and
   266     Transfer packages, as with 'setup_lifting'.
   267 
   268   - The 'quotient_definition' command now requires the user to prove a
   269     respectfulness property at the point where the constant is
   270     defined, similar to lift_definition; INCOMPATIBILITY.
   271 
   272   - Renamed predicate 'Quotient' to 'Quotient3', and renamed theorems
   273     accordingly, INCOMPATIBILITY.
   274 
   275 * New diagnostic command 'find_unused_assms' to find potentially
   276 superfluous assumptions in theorems using Quickcheck.
   277 
   278 * Quickcheck:
   279 
   280   - Quickcheck returns variable assignments as counterexamples, which
   281     allows to reveal the underspecification of functions under test.
   282     For example, refuting "hd xs = x", it presents the variable
   283     assignment xs = [] and x = a1 as a counterexample, assuming that
   284     any property is false whenever "hd []" occurs in it.
   285 
   286     These counterexample are marked as potentially spurious, as
   287     Quickcheck also returns "xs = []" as a counterexample to the
   288     obvious theorem "hd xs = hd xs".
   289 
   290     After finding a potentially spurious counterexample, Quickcheck
   291     continues searching for genuine ones.
   292 
   293     By default, Quickcheck shows potentially spurious and genuine
   294     counterexamples. The option "genuine_only" sets quickcheck to only
   295     show genuine counterexamples.
   296 
   297   - The command 'quickcheck_generator' creates random and exhaustive
   298     value generators for a given type and operations.
   299 
   300     It generates values by using the operations as if they were
   301     constructors of that type.
   302 
   303   - Support for multisets.
   304 
   305   - Added "use_subtype" options.
   306 
   307   - Added "quickcheck_locale" configuration to specify how to process
   308     conjectures in a locale context.
   309 
   310 * Nitpick: Fixed infinite loop caused by the 'peephole_optim' option
   311 and affecting 'rat' and 'real'.
   312 
   313 * Sledgehammer:
   314   - Integrated more tightly with SPASS, as described in the ITP 2012
   315     paper "More SPASS with Isabelle".
   316   - Made it try "smt" as a fallback if "metis" fails or times out.
   317   - Added support for the following provers: Alt-Ergo (via Why3 and
   318     TFF1), iProver, iProver-Eq.
   319   - Replaced remote E-SInE with remote Satallax in the default setup.
   320   - Sped up the minimizer.
   321   - Added "lam_trans", "uncurry_aliases", and "minimize" options.
   322   - Renamed "slicing" ("no_slicing") option to "slice" ("dont_slice").
   323   - Renamed "sound" option to "strict".
   324 
   325 * Metis: Added possibility to specify lambda translations scheme as a
   326 parenthesized argument (e.g., "by (metis (lifting) ...)").
   327 
   328 * SMT: Renamed "smt_fixed" option to "smt_read_only_certificates".
   329 
   330 * Command 'try0': Renamed from 'try_methods'. INCOMPATIBILITY.
   331 
   332 * New "case_product" attribute to generate a case rule doing multiple
   333 case distinctions at the same time.  E.g.
   334 
   335   list.exhaust [case_product nat.exhaust]
   336 
   337 produces a rule which can be used to perform case distinction on both
   338 a list and a nat.
   339 
   340 * New "eventually_elim" method as a generalized variant of the
   341 eventually_elim* rules.  Supports structured proofs.
   342 
   343 * Typedef with implicit set definition is considered legacy.  Use
   344 "typedef (open)" form instead, which will eventually become the
   345 default.
   346 
   347 * Record: code generation can be switched off manually with
   348 
   349   declare [[record_coden = false]]  -- "default true"
   350 
   351 * Datatype: type parameters allow explicit sort constraints.
   352 
   353 * Concrete syntax for case expressions includes constraints for source
   354 positions, and thus produces Prover IDE markup for its bindings.
   355 INCOMPATIBILITY for old-style syntax translations that augment the
   356 pattern notation; e.g. see src/HOL/HOLCF/One.thy for translations of
   357 one_case.
   358 
   359 * Clarified attribute "mono_set": pure declaration without modifying
   360 the result of the fact expression.
   361 
   362 * More default pred/set conversions on a couple of relation operations
   363 and predicates.  Added powers of predicate relations.  Consolidation
   364 of some relation theorems:
   365 
   366   converse_def ~> converse_unfold
   367   rel_comp_def ~> relcomp_unfold
   368   symp_def ~> (modified, use symp_def and sym_def instead)
   369   transp_def ~> transp_trans
   370   Domain_def ~> Domain_unfold
   371   Range_def ~> Domain_converse [symmetric]
   372 
   373 Generalized theorems INF_INT_eq, INF_INT_eq2, SUP_UN_eq, SUP_UN_eq2.
   374 
   375 See theory "Relation" for examples for making use of pred/set
   376 conversions by means of attributes "to_set" and "to_pred".
   377 
   378 INCOMPATIBILITY.
   379 
   380 * Renamed facts about the power operation on relations, i.e., relpow
   381 to match the constant's name:
   382 
   383   rel_pow_1 ~> relpow_1
   384   rel_pow_0_I ~> relpow_0_I
   385   rel_pow_Suc_I ~> relpow_Suc_I
   386   rel_pow_Suc_I2 ~> relpow_Suc_I2
   387   rel_pow_0_E ~> relpow_0_E
   388   rel_pow_Suc_E ~> relpow_Suc_E
   389   rel_pow_E ~> relpow_E
   390   rel_pow_Suc_D2 ~> relpow_Suc_D2
   391   rel_pow_Suc_E2 ~> relpow_Suc_E2
   392   rel_pow_Suc_D2' ~> relpow_Suc_D2'
   393   rel_pow_E2 ~> relpow_E2
   394   rel_pow_add ~> relpow_add
   395   rel_pow_commute ~> relpow
   396   rel_pow_empty ~> relpow_empty:
   397   rtrancl_imp_UN_rel_pow ~> rtrancl_imp_UN_relpow
   398   rel_pow_imp_rtrancl ~> relpow_imp_rtrancl
   399   rtrancl_is_UN_rel_pow ~> rtrancl_is_UN_relpow
   400   rtrancl_imp_rel_pow ~> rtrancl_imp_relpow
   401   rel_pow_fun_conv ~> relpow_fun_conv
   402   rel_pow_finite_bounded1 ~> relpow_finite_bounded1
   403   rel_pow_finite_bounded ~> relpow_finite_bounded
   404   rtrancl_finite_eq_rel_pow ~> rtrancl_finite_eq_relpow
   405   trancl_finite_eq_rel_pow ~> trancl_finite_eq_relpow
   406   single_valued_rel_pow ~> single_valued_relpow
   407 
   408 INCOMPATIBILITY.
   409 
   410 * Theory Relation: Consolidated constant name for relation composition
   411 and corresponding theorem names:
   412 
   413   - Renamed constant rel_comp to relcomp.
   414 
   415   - Dropped abbreviation pred_comp. Use relcompp instead.
   416 
   417   - Renamed theorems:
   418 
   419     rel_compI ~> relcompI
   420     rel_compEpair ~> relcompEpair
   421     rel_compE ~> relcompE
   422     pred_comp_rel_comp_eq ~> relcompp_relcomp_eq
   423     rel_comp_empty1 ~> relcomp_empty1
   424     rel_comp_mono ~> relcomp_mono
   425     rel_comp_subset_Sigma ~> relcomp_subset_Sigma
   426     rel_comp_distrib ~> relcomp_distrib
   427     rel_comp_distrib2 ~> relcomp_distrib2
   428     rel_comp_UNION_distrib ~> relcomp_UNION_distrib
   429     rel_comp_UNION_distrib2 ~> relcomp_UNION_distrib2
   430     single_valued_rel_comp ~> single_valued_relcomp
   431     rel_comp_def ~> relcomp_unfold
   432     converse_rel_comp ~> converse_relcomp
   433     pred_compI ~> relcomppI
   434     pred_compE ~> relcomppE
   435     pred_comp_bot1 ~> relcompp_bot1
   436     pred_comp_bot2 ~> relcompp_bot2
   437     transp_pred_comp_less_eq ~> transp_relcompp_less_eq
   438     pred_comp_mono ~> relcompp_mono
   439     pred_comp_distrib ~> relcompp_distrib
   440     pred_comp_distrib2 ~> relcompp_distrib2
   441     converse_pred_comp ~> converse_relcompp
   442 
   443     finite_rel_comp ~> finite_relcomp
   444 
   445     set_rel_comp ~> set_relcomp
   446 
   447 INCOMPATIBILITY.
   448 
   449 * Theory Divides: Discontinued redundant theorems about div and mod.
   450 INCOMPATIBILITY, use the corresponding generic theorems instead.
   451 
   452   DIVISION_BY_ZERO ~> div_by_0, mod_by_0
   453   zdiv_self ~> div_self
   454   zmod_self ~> mod_self
   455   zdiv_zero ~> div_0
   456   zmod_zero ~> mod_0
   457   zdiv_zmod_equality ~> div_mod_equality2
   458   zdiv_zmod_equality2 ~> div_mod_equality
   459   zmod_zdiv_trivial ~> mod_div_trivial
   460   zdiv_zminus_zminus ~> div_minus_minus
   461   zmod_zminus_zminus ~> mod_minus_minus
   462   zdiv_zminus2 ~> div_minus_right
   463   zmod_zminus2 ~> mod_minus_right
   464   zdiv_minus1_right ~> div_minus1_right
   465   zmod_minus1_right ~> mod_minus1_right
   466   zdvd_mult_div_cancel ~> dvd_mult_div_cancel
   467   zmod_zmult1_eq ~> mod_mult_right_eq
   468   zpower_zmod ~> power_mod
   469   zdvd_zmod ~> dvd_mod
   470   zdvd_zmod_imp_zdvd ~> dvd_mod_imp_dvd
   471   mod_mult_distrib ~> mult_mod_left
   472   mod_mult_distrib2 ~> mult_mod_right
   473 
   474 * Removed redundant theorems nat_mult_2 and nat_mult_2_right; use
   475 generic mult_2 and mult_2_right instead. INCOMPATIBILITY.
   476 
   477 * Finite_Set.fold now qualified.  INCOMPATIBILITY.
   478 
   479 * Consolidated theorem names concerning fold combinators:
   480 
   481   inf_INFI_fold_inf ~> inf_INF_fold_inf
   482   sup_SUPR_fold_sup ~> sup_SUP_fold_sup
   483   INFI_fold_inf ~> INF_fold_inf
   484   SUPR_fold_sup ~> SUP_fold_sup
   485   union_set ~> union_set_fold
   486   minus_set ~> minus_set_fold
   487   INFI_set_fold ~> INF_set_fold
   488   SUPR_set_fold ~> SUP_set_fold
   489   INF_code ~> INF_set_foldr
   490   SUP_code ~> SUP_set_foldr
   491   foldr.simps ~> foldr.simps (in point-free formulation)
   492   foldr_fold_rev ~> foldr_conv_fold
   493   foldl_fold ~> foldl_conv_fold
   494   foldr_foldr ~> foldr_conv_foldl
   495   foldl_foldr ~> foldl_conv_foldr
   496   fold_set_remdups ~> fold_set_fold_remdups
   497   fold_set ~> fold_set_fold
   498   fold1_set ~> fold1_set_fold
   499 
   500 INCOMPATIBILITY.
   501 
   502 * Dropped rarely useful theorems concerning fold combinators:
   503 foldl_apply, foldl_fun_comm, foldl_rev, fold_weak_invariant,
   504 rev_foldl_cons, fold_set_remdups, fold_set, fold_set1,
   505 concat_conv_foldl, foldl_weak_invariant, foldl_invariant,
   506 foldr_invariant, foldl_absorb0, foldl_foldr1_lemma, foldl_foldr1,
   507 listsum_conv_fold, listsum_foldl, sort_foldl_insort, foldl_assoc,
   508 foldr_conv_foldl, start_le_sum, elem_le_sum, sum_eq_0_conv.
   509 INCOMPATIBILITY.  For the common phrases "%xs. List.foldr plus xs 0"
   510 and "List.foldl plus 0", prefer "List.listsum".  Otherwise it can be
   511 useful to boil down "List.foldr" and "List.foldl" to "List.fold" by
   512 unfolding "foldr_conv_fold" and "foldl_conv_fold".
   513 
   514 * Dropped lemmas minus_set_foldr, union_set_foldr, union_coset_foldr,
   515 inter_coset_foldr, Inf_fin_set_foldr, Sup_fin_set_foldr,
   516 Min_fin_set_foldr, Max_fin_set_foldr, Inf_set_foldr, Sup_set_foldr,
   517 INF_set_foldr, SUP_set_foldr.  INCOMPATIBILITY.  Prefer corresponding
   518 lemmas over fold rather than foldr, or make use of lemmas
   519 fold_conv_foldr and fold_rev.
   520 
   521 * Congruence rules Option.map_cong and Option.bind_cong for recursion
   522 through option types.
   523 
   524 * "Transitive_Closure.ntrancl": bounded transitive closure on
   525 relations.
   526 
   527 * Constant "Set.not_member" now qualified.  INCOMPATIBILITY.
   528 
   529 * Theory Int: Discontinued many legacy theorems specific to type int.
   530 INCOMPATIBILITY, use the corresponding generic theorems instead.
   531 
   532   zminus_zminus ~> minus_minus
   533   zminus_0 ~> minus_zero
   534   zminus_zadd_distrib ~> minus_add_distrib
   535   zadd_commute ~> add_commute
   536   zadd_assoc ~> add_assoc
   537   zadd_left_commute ~> add_left_commute
   538   zadd_ac ~> add_ac
   539   zmult_ac ~> mult_ac
   540   zadd_0 ~> add_0_left
   541   zadd_0_right ~> add_0_right
   542   zadd_zminus_inverse2 ~> left_minus
   543   zmult_zminus ~> mult_minus_left
   544   zmult_commute ~> mult_commute
   545   zmult_assoc ~> mult_assoc
   546   zadd_zmult_distrib ~> left_distrib
   547   zadd_zmult_distrib2 ~> right_distrib
   548   zdiff_zmult_distrib ~> left_diff_distrib
   549   zdiff_zmult_distrib2 ~> right_diff_distrib
   550   zmult_1 ~> mult_1_left
   551   zmult_1_right ~> mult_1_right
   552   zle_refl ~> order_refl
   553   zle_trans ~> order_trans
   554   zle_antisym ~> order_antisym
   555   zle_linear ~> linorder_linear
   556   zless_linear ~> linorder_less_linear
   557   zadd_left_mono ~> add_left_mono
   558   zadd_strict_right_mono ~> add_strict_right_mono
   559   zadd_zless_mono ~> add_less_le_mono
   560   int_0_less_1 ~> zero_less_one
   561   int_0_neq_1 ~> zero_neq_one
   562   zless_le ~> less_le
   563   zpower_zadd_distrib ~> power_add
   564   zero_less_zpower_abs_iff ~> zero_less_power_abs_iff
   565   zero_le_zpower_abs ~> zero_le_power_abs
   566 
   567 * Theory Deriv: Renamed
   568 
   569   DERIV_nonneg_imp_nonincreasing ~> DERIV_nonneg_imp_nondecreasing
   570 
   571 * Theory Library/Multiset: Improved code generation of multisets.
   572 
   573 * Theory HOL/Library/Set_Algebras: Addition and multiplication on sets
   574 are expressed via type classes again. The special syntax
   575 \<oplus>/\<otimes> has been replaced by plain +/*. Removed constant
   576 setsum_set, which is now subsumed by Big_Operators.setsum.
   577 INCOMPATIBILITY.
   578 
   579 * Theory HOL/Library/Diagonalize has been removed. INCOMPATIBILITY,
   580 use theory HOL/Library/Nat_Bijection instead.
   581 
   582 * Theory HOL/Library/RBT_Impl: Backing implementation of red-black
   583 trees is now inside a type class context.  Names of affected
   584 operations and lemmas have been prefixed by rbt_.  INCOMPATIBILITY for
   585 theories working directly with raw red-black trees, adapt the names as
   586 follows:
   587 
   588   Operations:
   589   bulkload -> rbt_bulkload
   590   del_from_left -> rbt_del_from_left
   591   del_from_right -> rbt_del_from_right
   592   del -> rbt_del
   593   delete -> rbt_delete
   594   ins -> rbt_ins
   595   insert -> rbt_insert
   596   insertw -> rbt_insert_with
   597   insert_with_key -> rbt_insert_with_key
   598   map_entry -> rbt_map_entry
   599   lookup -> rbt_lookup
   600   sorted -> rbt_sorted
   601   tree_greater -> rbt_greater
   602   tree_less -> rbt_less
   603   tree_less_symbol -> rbt_less_symbol
   604   union -> rbt_union
   605   union_with -> rbt_union_with
   606   union_with_key -> rbt_union_with_key
   607 
   608   Lemmas:
   609   balance_left_sorted -> balance_left_rbt_sorted
   610   balance_left_tree_greater -> balance_left_rbt_greater
   611   balance_left_tree_less -> balance_left_rbt_less
   612   balance_right_sorted -> balance_right_rbt_sorted
   613   balance_right_tree_greater -> balance_right_rbt_greater
   614   balance_right_tree_less -> balance_right_rbt_less
   615   balance_sorted -> balance_rbt_sorted
   616   balance_tree_greater -> balance_rbt_greater
   617   balance_tree_less -> balance_rbt_less
   618   bulkload_is_rbt -> rbt_bulkload_is_rbt
   619   combine_sorted -> combine_rbt_sorted
   620   combine_tree_greater -> combine_rbt_greater
   621   combine_tree_less -> combine_rbt_less
   622   delete_in_tree -> rbt_delete_in_tree
   623   delete_is_rbt -> rbt_delete_is_rbt
   624   del_from_left_tree_greater -> rbt_del_from_left_rbt_greater
   625   del_from_left_tree_less -> rbt_del_from_left_rbt_less
   626   del_from_right_tree_greater -> rbt_del_from_right_rbt_greater
   627   del_from_right_tree_less -> rbt_del_from_right_rbt_less
   628   del_in_tree -> rbt_del_in_tree
   629   del_inv1_inv2 -> rbt_del_inv1_inv2
   630   del_sorted -> rbt_del_rbt_sorted
   631   del_tree_greater -> rbt_del_rbt_greater
   632   del_tree_less -> rbt_del_rbt_less
   633   dom_lookup_Branch -> dom_rbt_lookup_Branch
   634   entries_lookup -> entries_rbt_lookup
   635   finite_dom_lookup -> finite_dom_rbt_lookup
   636   insert_sorted -> rbt_insert_rbt_sorted
   637   insertw_is_rbt -> rbt_insertw_is_rbt
   638   insertwk_is_rbt -> rbt_insertwk_is_rbt
   639   insertwk_sorted -> rbt_insertwk_rbt_sorted
   640   insertw_sorted -> rbt_insertw_rbt_sorted
   641   ins_sorted -> ins_rbt_sorted
   642   ins_tree_greater -> ins_rbt_greater
   643   ins_tree_less -> ins_rbt_less
   644   is_rbt_sorted -> is_rbt_rbt_sorted
   645   lookup_balance -> rbt_lookup_balance
   646   lookup_bulkload -> rbt_lookup_rbt_bulkload
   647   lookup_delete -> rbt_lookup_rbt_delete
   648   lookup_Empty -> rbt_lookup_Empty
   649   lookup_from_in_tree -> rbt_lookup_from_in_tree
   650   lookup_in_tree -> rbt_lookup_in_tree
   651   lookup_ins -> rbt_lookup_ins
   652   lookup_insert -> rbt_lookup_rbt_insert
   653   lookup_insertw -> rbt_lookup_rbt_insertw
   654   lookup_insertwk -> rbt_lookup_rbt_insertwk
   655   lookup_keys -> rbt_lookup_keys
   656   lookup_map -> rbt_lookup_map
   657   lookup_map_entry -> rbt_lookup_rbt_map_entry
   658   lookup_tree_greater -> rbt_lookup_rbt_greater
   659   lookup_tree_less -> rbt_lookup_rbt_less
   660   lookup_union -> rbt_lookup_rbt_union
   661   map_entry_color_of -> rbt_map_entry_color_of
   662   map_entry_inv1 -> rbt_map_entry_inv1
   663   map_entry_inv2 -> rbt_map_entry_inv2
   664   map_entry_is_rbt -> rbt_map_entry_is_rbt
   665   map_entry_sorted -> rbt_map_entry_rbt_sorted
   666   map_entry_tree_greater -> rbt_map_entry_rbt_greater
   667   map_entry_tree_less -> rbt_map_entry_rbt_less
   668   map_tree_greater -> map_rbt_greater
   669   map_tree_less -> map_rbt_less
   670   map_sorted -> map_rbt_sorted
   671   paint_sorted -> paint_rbt_sorted
   672   paint_lookup -> paint_rbt_lookup
   673   paint_tree_greater -> paint_rbt_greater
   674   paint_tree_less -> paint_rbt_less
   675   sorted_entries -> rbt_sorted_entries
   676   tree_greater_eq_trans -> rbt_greater_eq_trans
   677   tree_greater_nit -> rbt_greater_nit
   678   tree_greater_prop -> rbt_greater_prop
   679   tree_greater_simps -> rbt_greater_simps
   680   tree_greater_trans -> rbt_greater_trans
   681   tree_less_eq_trans -> rbt_less_eq_trans
   682   tree_less_nit -> rbt_less_nit
   683   tree_less_prop -> rbt_less_prop
   684   tree_less_simps -> rbt_less_simps
   685   tree_less_trans -> rbt_less_trans
   686   tree_ord_props -> rbt_ord_props
   687   union_Branch -> rbt_union_Branch
   688   union_is_rbt -> rbt_union_is_rbt
   689   unionw_is_rbt -> rbt_unionw_is_rbt
   690   unionwk_is_rbt -> rbt_unionwk_is_rbt
   691   unionwk_sorted -> rbt_unionwk_rbt_sorted
   692 
   693 * Theory HOL/Library/Float: Floating point numbers are now defined as
   694 a subset of the real numbers.  All operations are defined using the
   695 lifing-framework and proofs use the transfer method.  INCOMPATIBILITY.
   696 
   697   Changed Operations:
   698   float_abs -> abs
   699   float_nprt -> nprt
   700   float_pprt -> pprt
   701   pow2 -> use powr
   702   round_down -> float_round_down
   703   round_up -> float_round_up
   704   scale -> exponent
   705 
   706   Removed Operations:
   707   ceiling_fl, lb_mult, lb_mod, ub_mult, ub_mod
   708 
   709   Renamed Lemmas:
   710   abs_float_def -> Float.compute_float_abs
   711   bitlen_ge0 -> bitlen_nonneg
   712   bitlen.simps -> Float.compute_bitlen
   713   float_components -> Float_mantissa_exponent
   714   float_divl.simps -> Float.compute_float_divl
   715   float_divr.simps -> Float.compute_float_divr
   716   float_eq_odd -> mult_powr_eq_mult_powr_iff
   717   float_power -> real_of_float_power
   718   lapprox_posrat_def -> Float.compute_lapprox_posrat
   719   lapprox_rat.simps -> Float.compute_lapprox_rat
   720   le_float_def' -> Float.compute_float_le
   721   le_float_def -> less_eq_float.rep_eq
   722   less_float_def' -> Float.compute_float_less
   723   less_float_def -> less_float.rep_eq
   724   normfloat_def -> Float.compute_normfloat
   725   normfloat_imp_odd_or_zero -> mantissa_not_dvd and mantissa_noteq_0
   726   normfloat -> normfloat_def
   727   normfloat_unique -> use normfloat_def
   728   number_of_float_Float -> Float.compute_float_numeral, Float.compute_float_neg_numeral
   729   one_float_def -> Float.compute_float_one
   730   plus_float_def -> Float.compute_float_plus
   731   rapprox_posrat_def -> Float.compute_rapprox_posrat
   732   rapprox_rat.simps -> Float.compute_rapprox_rat
   733   real_of_float_0 -> zero_float.rep_eq
   734   real_of_float_1 -> one_float.rep_eq
   735   real_of_float_abs -> abs_float.rep_eq
   736   real_of_float_add -> plus_float.rep_eq
   737   real_of_float_minus -> uminus_float.rep_eq
   738   real_of_float_mult -> times_float.rep_eq
   739   real_of_float_simp -> Float.rep_eq
   740   real_of_float_sub -> minus_float.rep_eq
   741   round_down.simps -> Float.compute_float_round_down
   742   round_up.simps -> Float.compute_float_round_up
   743   times_float_def -> Float.compute_float_times
   744   uminus_float_def -> Float.compute_float_uminus
   745   zero_float_def -> Float.compute_float_zero
   746 
   747   Lemmas not necessary anymore, use the transfer method:
   748   bitlen_B0, bitlen_B1, bitlen_ge1, bitlen_Min, bitlen_Pls, float_divl,
   749   float_divr, float_le_simp, float_less1_mantissa_bound,
   750   float_less_simp, float_less_zero, float_le_zero,
   751   float_pos_less1_e_neg, float_pos_m_pos, float_split, float_split2,
   752   floor_pos_exp, lapprox_posrat, lapprox_posrat_bottom, lapprox_rat,
   753   lapprox_rat_bottom, normalized_float, rapprox_posrat,
   754   rapprox_posrat_le1, rapprox_rat, real_of_float_ge0_exp,
   755   real_of_float_neg_exp, real_of_float_nge0_exp, round_down floor_fl,
   756   round_up, zero_le_float, zero_less_float
   757 
   758 * New theory HOL/Library/DAList provides an abstract type for
   759 association lists with distinct keys.
   760 
   761 * Session HOL/IMP: Added new theory of abstract interpretation of
   762 annotated commands.
   763 
   764 * Session HOL-Import: Re-implementation from scratch is faster,
   765 simpler, and more scalable.  Requires a proof bundle, which is
   766 available as an external component.  Discontinued old (and mostly
   767 dead) Importer for HOL4 and HOL Light.  INCOMPATIBILITY.
   768 
   769 * Session HOL-Word: Discontinued many redundant theorems specific to
   770 type 'a word. INCOMPATIBILITY, use the corresponding generic theorems
   771 instead.
   772 
   773   word_sub_alt ~> word_sub_wi
   774   word_add_alt ~> word_add_def
   775   word_mult_alt ~> word_mult_def
   776   word_minus_alt ~> word_minus_def
   777   word_0_alt ~> word_0_wi
   778   word_1_alt ~> word_1_wi
   779   word_add_0 ~> add_0_left
   780   word_add_0_right ~> add_0_right
   781   word_mult_1 ~> mult_1_left
   782   word_mult_1_right ~> mult_1_right
   783   word_add_commute ~> add_commute
   784   word_add_assoc ~> add_assoc
   785   word_add_left_commute ~> add_left_commute
   786   word_mult_commute ~> mult_commute
   787   word_mult_assoc ~> mult_assoc
   788   word_mult_left_commute ~> mult_left_commute
   789   word_left_distrib ~> left_distrib
   790   word_right_distrib ~> right_distrib
   791   word_left_minus ~> left_minus
   792   word_diff_0_right ~> diff_0_right
   793   word_diff_self ~> diff_self
   794   word_sub_def ~> diff_minus
   795   word_diff_minus ~> diff_minus
   796   word_add_ac ~> add_ac
   797   word_mult_ac ~> mult_ac
   798   word_plus_ac0 ~> add_0_left add_0_right add_ac
   799   word_times_ac1 ~> mult_1_left mult_1_right mult_ac
   800   word_order_trans ~> order_trans
   801   word_order_refl ~> order_refl
   802   word_order_antisym ~> order_antisym
   803   word_order_linear ~> linorder_linear
   804   lenw1_zero_neq_one ~> zero_neq_one
   805   word_number_of_eq ~> number_of_eq
   806   word_of_int_add_hom ~> wi_hom_add
   807   word_of_int_sub_hom ~> wi_hom_sub
   808   word_of_int_mult_hom ~> wi_hom_mult
   809   word_of_int_minus_hom ~> wi_hom_neg
   810   word_of_int_succ_hom ~> wi_hom_succ
   811   word_of_int_pred_hom ~> wi_hom_pred
   812   word_of_int_0_hom ~> word_0_wi
   813   word_of_int_1_hom ~> word_1_wi
   814 
   815 * Session HOL-Word: New proof method "word_bitwise" for splitting
   816 machine word equalities and inequalities into logical circuits,
   817 defined in HOL/Word/WordBitwise.thy.  Supports addition, subtraction,
   818 multiplication, shifting by constants, bitwise operators and numeric
   819 constants.  Requires fixed-length word types, not 'a word.  Solves
   820 many standard word identities outright and converts more into first
   821 order problems amenable to blast or similar.  See also examples in
   822 HOL/Word/Examples/WordExamples.thy.
   823 
   824 * Session HOL-Probability: Introduced the type "'a measure" to
   825 represent measures, this replaces the records 'a algebra and 'a
   826 measure_space.  The locales based on subset_class now have two
   827 locale-parameters the space \<Omega> and the set of measurable sets M.
   828 The product of probability spaces uses now the same constant as the
   829 finite product of sigma-finite measure spaces "PiM :: ('i => 'a)
   830 measure".  Most constants are defined now outside of locales and gain
   831 an additional parameter, like null_sets, almost_eventually or \<mu>'.
   832 Measure space constructions for distributions and densities now got
   833 their own constants distr and density.  Instead of using locales to
   834 describe measure spaces with a finite space, the measure count_space
   835 and point_measure is introduced.  INCOMPATIBILITY.
   836 
   837   Renamed constants:
   838   measure -> emeasure
   839   finite_measure.\<mu>' -> measure
   840   product_algebra_generator -> prod_algebra
   841   product_prob_space.emb -> prod_emb
   842   product_prob_space.infprod_algebra -> PiM
   843 
   844   Removed locales:
   845   completeable_measure_space
   846   finite_measure_space
   847   finite_prob_space
   848   finite_product_finite_prob_space
   849   finite_product_sigma_algebra
   850   finite_sigma_algebra
   851   measure_space
   852   pair_finite_prob_space
   853   pair_finite_sigma_algebra
   854   pair_finite_space
   855   pair_sigma_algebra
   856   product_sigma_algebra
   857 
   858   Removed constants:
   859   conditional_space
   860   distribution -> use distr measure, or distributed predicate
   861   image_space
   862   joint_distribution -> use distr measure, or distributed predicate
   863   pair_measure_generator
   864   product_prob_space.infprod_algebra -> use PiM
   865   subvimage
   866 
   867   Replacement theorems:
   868   finite_additivity_sufficient -> ring_of_sets.countably_additiveI_finite
   869   finite_measure.empty_measure -> measure_empty
   870   finite_measure.finite_continuity_from_above -> finite_measure.finite_Lim_measure_decseq
   871   finite_measure.finite_continuity_from_below -> finite_measure.finite_Lim_measure_incseq
   872   finite_measure.finite_measure_countably_subadditive -> finite_measure.finite_measure_subadditive_countably
   873   finite_measure.finite_measure_eq -> finite_measure.emeasure_eq_measure
   874   finite_measure.finite_measure -> finite_measure.emeasure_finite
   875   finite_measure.finite_measure_finite_singleton -> finite_measure.finite_measure_eq_setsum_singleton
   876   finite_measure.positive_measure' -> measure_nonneg
   877   finite_measure.real_measure -> finite_measure.emeasure_real
   878   finite_product_prob_space.finite_measure_times -> finite_product_prob_space.finite_measure_PiM_emb
   879   finite_product_sigma_algebra.in_P -> sets_PiM_I_finite
   880   finite_product_sigma_algebra.P_empty -> space_PiM_empty, sets_PiM_empty
   881   information_space.conditional_entropy_eq -> information_space.conditional_entropy_simple_distributed
   882   information_space.conditional_entropy_positive -> information_space.conditional_entropy_nonneg_simple
   883   information_space.conditional_mutual_information_eq_mutual_information -> information_space.conditional_mutual_information_eq_mutual_information_simple
   884   information_space.conditional_mutual_information_generic_positive -> information_space.conditional_mutual_information_nonneg_simple
   885   information_space.conditional_mutual_information_positive -> information_space.conditional_mutual_information_nonneg_simple
   886   information_space.entropy_commute -> information_space.entropy_commute_simple
   887   information_space.entropy_eq -> information_space.entropy_simple_distributed
   888   information_space.entropy_generic_eq -> information_space.entropy_simple_distributed
   889   information_space.entropy_positive -> information_space.entropy_nonneg_simple
   890   information_space.entropy_uniform_max -> information_space.entropy_uniform
   891   information_space.KL_eq_0_imp -> information_space.KL_eq_0_iff_eq
   892   information_space.KL_eq_0 -> information_space.KL_same_eq_0
   893   information_space.KL_ge_0 -> information_space.KL_nonneg
   894   information_space.mutual_information_eq -> information_space.mutual_information_simple_distributed
   895   information_space.mutual_information_positive -> information_space.mutual_information_nonneg_simple
   896   Int_stable_cuboids -> Int_stable_atLeastAtMost
   897   Int_stable_product_algebra_generator -> positive_integral
   898   measure_preserving -> equality "distr M N f = N" "f : measurable M N"
   899   measure_space.additive -> emeasure_additive
   900   measure_space.AE_iff_null_set -> AE_iff_null
   901   measure_space.almost_everywhere_def -> eventually_ae_filter
   902   measure_space.almost_everywhere_vimage -> AE_distrD
   903   measure_space.continuity_from_above -> INF_emeasure_decseq
   904   measure_space.continuity_from_above_Lim -> Lim_emeasure_decseq
   905   measure_space.continuity_from_below_Lim -> Lim_emeasure_incseq
   906   measure_space.continuity_from_below -> SUP_emeasure_incseq
   907   measure_space_density -> emeasure_density
   908   measure_space.density_is_absolutely_continuous -> absolutely_continuousI_density
   909   measure_space.integrable_vimage -> integrable_distr
   910   measure_space.integral_translated_density -> integral_density
   911   measure_space.integral_vimage -> integral_distr
   912   measure_space.measure_additive -> plus_emeasure
   913   measure_space.measure_compl -> emeasure_compl
   914   measure_space.measure_countable_increasing -> emeasure_countable_increasing
   915   measure_space.measure_countably_subadditive -> emeasure_subadditive_countably
   916   measure_space.measure_decseq -> decseq_emeasure
   917   measure_space.measure_Diff -> emeasure_Diff
   918   measure_space.measure_Diff_null_set -> emeasure_Diff_null_set
   919   measure_space.measure_eq_0 -> emeasure_eq_0
   920   measure_space.measure_finitely_subadditive -> emeasure_subadditive_finite
   921   measure_space.measure_finite_singleton -> emeasure_eq_setsum_singleton
   922   measure_space.measure_incseq -> incseq_emeasure
   923   measure_space.measure_insert -> emeasure_insert
   924   measure_space.measure_mono -> emeasure_mono
   925   measure_space.measure_not_negative -> emeasure_not_MInf
   926   measure_space.measure_preserving_Int_stable -> measure_eqI_generator_eq
   927   measure_space.measure_setsum -> setsum_emeasure
   928   measure_space.measure_setsum_split -> setsum_emeasure_cover
   929   measure_space.measure_space_vimage -> emeasure_distr
   930   measure_space.measure_subadditive_finite -> emeasure_subadditive_finite
   931   measure_space.measure_subadditive -> subadditive
   932   measure_space.measure_top -> emeasure_space
   933   measure_space.measure_UN_eq_0 -> emeasure_UN_eq_0
   934   measure_space.measure_Un_null_set -> emeasure_Un_null_set
   935   measure_space.positive_integral_translated_density -> positive_integral_density
   936   measure_space.positive_integral_vimage -> positive_integral_distr
   937   measure_space.real_continuity_from_above -> Lim_measure_decseq
   938   measure_space.real_continuity_from_below -> Lim_measure_incseq
   939   measure_space.real_measure_countably_subadditive -> measure_subadditive_countably
   940   measure_space.real_measure_Diff -> measure_Diff
   941   measure_space.real_measure_finite_Union -> measure_finite_Union
   942   measure_space.real_measure_setsum_singleton -> measure_eq_setsum_singleton
   943   measure_space.real_measure_subadditive -> measure_subadditive
   944   measure_space.real_measure_Union -> measure_Union
   945   measure_space.real_measure_UNION -> measure_UNION
   946   measure_space.simple_function_vimage -> simple_function_comp
   947   measure_space.simple_integral_vimage -> simple_integral_distr
   948   measure_space.simple_integral_vimage -> simple_integral_distr
   949   measure_unique_Int_stable -> measure_eqI_generator_eq
   950   measure_unique_Int_stable_vimage -> measure_eqI_generator_eq
   951   pair_sigma_algebra.measurable_cut_fst -> sets_Pair1
   952   pair_sigma_algebra.measurable_cut_snd -> sets_Pair2
   953   pair_sigma_algebra.measurable_pair_image_fst -> measurable_Pair1
   954   pair_sigma_algebra.measurable_pair_image_snd -> measurable_Pair2
   955   pair_sigma_algebra.measurable_product_swap -> measurable_pair_swap_iff
   956   pair_sigma_algebra.pair_sigma_algebra_measurable -> measurable_pair_swap
   957   pair_sigma_algebra.pair_sigma_algebra_swap_measurable -> measurable_pair_swap'
   958   pair_sigma_algebra.sets_swap -> sets_pair_swap
   959   pair_sigma_finite.measure_cut_measurable_fst -> pair_sigma_finite.measurable_emeasure_Pair1
   960   pair_sigma_finite.measure_cut_measurable_snd -> pair_sigma_finite.measurable_emeasure_Pair2
   961   pair_sigma_finite.measure_preserving_swap -> pair_sigma_finite.distr_pair_swap
   962   pair_sigma_finite.pair_measure_alt2 -> pair_sigma_finite.emeasure_pair_measure_alt2
   963   pair_sigma_finite.pair_measure_alt -> pair_sigma_finite.emeasure_pair_measure_alt
   964   pair_sigma_finite.pair_measure_times -> pair_sigma_finite.emeasure_pair_measure_Times
   965   prob_space.indep_distribution_eq_measure -> prob_space.indep_vars_iff_distr_eq_PiM
   966   prob_space.indep_var_distributionD -> prob_space.indep_var_distribution_eq
   967   prob_space.measure_space_1 -> prob_space.emeasure_space_1
   968   prob_space.prob_space_vimage -> prob_space_distr
   969   prob_space.random_variable_restrict -> measurable_restrict
   970   prob_space_unique_Int_stable -> measure_eqI_prob_space
   971   product_algebraE -> prod_algebraE_all
   972   product_algebra_generator_der -> prod_algebra_eq_finite
   973   product_algebra_generator_into_space -> prod_algebra_sets_into_space
   974   product_algebraI -> sets_PiM_I_finite
   975   product_measure_exists -> product_sigma_finite.sigma_finite
   976   product_prob_space.finite_index_eq_finite_product -> product_prob_space.sets_PiM_generator
   977   product_prob_space.finite_measure_infprod_emb_Pi -> product_prob_space.measure_PiM_emb
   978   product_prob_space.infprod_spec -> product_prob_space.emeasure_PiM_emb_not_empty
   979   product_prob_space.measurable_component -> measurable_component_singleton
   980   product_prob_space.measurable_emb -> measurable_prod_emb
   981   product_prob_space.measurable_into_infprod_algebra -> measurable_PiM_single
   982   product_prob_space.measurable_singleton_infprod -> measurable_component_singleton
   983   product_prob_space.measure_emb -> emeasure_prod_emb
   984   product_prob_space.measure_preserving_restrict -> product_prob_space.distr_restrict
   985   product_sigma_algebra.product_algebra_into_space -> space_closed
   986   product_sigma_finite.measure_fold -> product_sigma_finite.distr_merge
   987   product_sigma_finite.measure_preserving_component_singelton -> product_sigma_finite.distr_singleton
   988   product_sigma_finite.measure_preserving_merge -> product_sigma_finite.distr_merge
   989   sequence_space.measure_infprod -> sequence_space.measure_PiM_countable
   990   sets_product_algebra -> sets_PiM
   991   sigma_algebra.measurable_sigma -> measurable_measure_of
   992   sigma_finite_measure.disjoint_sigma_finite -> sigma_finite_disjoint
   993   sigma_finite_measure.RN_deriv_vimage -> sigma_finite_measure.RN_deriv_distr
   994   sigma_product_algebra_sigma_eq -> sigma_prod_algebra_sigma_eq
   995   space_product_algebra -> space_PiM
   996 
   997 * Session HOL-TPTP: support to parse and import TPTP problems (all
   998 languages) into Isabelle/HOL.
   999 
  1000 
  1001 *** FOL ***
  1002 
  1003 * New "case_product" attribute (see HOL).
  1004 
  1005 
  1006 *** ZF ***
  1007 
  1008 * Greater support for structured proofs involving induction or case
  1009 analysis.
  1010 
  1011 * Much greater use of mathematical symbols.
  1012 
  1013 * Removal of many ML theorem bindings.  INCOMPATIBILITY.
  1014 
  1015 
  1016 *** ML ***
  1017 
  1018 * Antiquotation @{keyword "name"} produces a parser for outer syntax
  1019 from a minor keyword introduced via theory header declaration.
  1020 
  1021 * Antiquotation @{command_spec "name"} produces the
  1022 Outer_Syntax.command_spec from a major keyword introduced via theory
  1023 header declaration; it can be passed to Outer_Syntax.command etc.
  1024 
  1025 * Local_Theory.define no longer hard-wires default theorem name
  1026 "foo_def", but retains the binding as given.  If that is Binding.empty
  1027 / Attrib.empty_binding, the result is not registered as user-level
  1028 fact.  The Local_Theory.define_internal variant allows to specify a
  1029 non-empty name (used for the foundation in the background theory),
  1030 while omitting the fact binding in the user-context.  Potential
  1031 INCOMPATIBILITY for derived definitional packages: need to specify
  1032 naming policy for primitive definitions more explicitly.
  1033 
  1034 * Renamed Thm.capply to Thm.apply, and Thm.cabs to Thm.lambda in
  1035 conformance with similar operations in structure Term and Logic.
  1036 
  1037 * Antiquotation @{attributes [...]} embeds attribute source
  1038 representation into the ML text, which is particularly useful with
  1039 declarations like Local_Theory.note.
  1040 
  1041 * Structure Proof_Context follows standard naming scheme.  Old
  1042 ProofContext has been discontinued.  INCOMPATIBILITY.
  1043 
  1044 * Refined Local_Theory.declaration {syntax, pervasive}, with subtle
  1045 change of semantics: update is applied to auxiliary local theory
  1046 context as well.
  1047 
  1048 * Modernized some old-style infix operations:
  1049 
  1050   addeqcongs    ~> Simplifier.add_eqcong
  1051   deleqcongs    ~> Simplifier.del_eqcong
  1052   addcongs      ~> Simplifier.add_cong
  1053   delcongs      ~> Simplifier.del_cong
  1054   setmksimps    ~> Simplifier.set_mksimps
  1055   setmkcong     ~> Simplifier.set_mkcong
  1056   setmksym      ~> Simplifier.set_mksym
  1057   setmkeqTrue   ~> Simplifier.set_mkeqTrue
  1058   settermless   ~> Simplifier.set_termless
  1059   setsubgoaler  ~> Simplifier.set_subgoaler
  1060   addsplits     ~> Splitter.add_split
  1061   delsplits     ~> Splitter.del_split
  1062 
  1063 
  1064 *** System ***
  1065 
  1066 * USER_HOME settings variable points to cross-platform user home
  1067 directory, which coincides with HOME on POSIX systems only.  Likewise,
  1068 the Isabelle path specification "~" now expands to $USER_HOME, instead
  1069 of former $HOME.  A different default for USER_HOME may be set
  1070 explicitly in shell environment, before Isabelle settings are
  1071 evaluated.  Minor INCOMPATIBILITY: need to adapt Isabelle path where
  1072 the generic user home was intended.
  1073 
  1074 * ISABELLE_HOME_WINDOWS refers to ISABELLE_HOME in windows file name
  1075 notation, which is useful for the jEdit file browser, for example.
  1076 
  1077 * ISABELLE_JDK_HOME settings variable points to JDK with javac and jar
  1078 (not just JRE).
  1079 
  1080 
  1081 
  1082 New in Isabelle2011-1 (October 2011)
  1083 ------------------------------------
  1084 
  1085 *** General ***
  1086 
  1087 * Improved Isabelle/jEdit Prover IDE (PIDE), which can be invoked as
  1088 "isabelle jedit" or "ISABELLE_HOME/Isabelle" on the command line.
  1089 
  1090   - Management of multiple theory files directly from the editor
  1091     buffer store -- bypassing the file-system (no requirement to save
  1092     files for checking).
  1093 
  1094   - Markup of formal entities within the text buffer, with semantic
  1095     highlighting, tooltips and hyperlinks to jump to defining source
  1096     positions.
  1097 
  1098   - Improved text rendering, with sub/superscripts in the source
  1099     buffer (including support for copy/paste wrt. output panel, HTML
  1100     theory output and other non-Isabelle text boxes).
  1101 
  1102   - Refined scheduling of proof checking and printing of results,
  1103     based on interactive editor view.  (Note: jEdit folding and
  1104     narrowing allows to restrict buffer perspectives explicitly.)
  1105 
  1106   - Reduced CPU performance requirements, usable on machines with few
  1107     cores.
  1108 
  1109   - Reduced memory requirements due to pruning of unused document
  1110     versions (garbage collection).
  1111 
  1112 See also ~~/src/Tools/jEdit/README.html for further information,
  1113 including some remaining limitations.
  1114 
  1115 * Theory loader: source files are exclusively located via the master
  1116 directory of each theory node (where the .thy file itself resides).
  1117 The global load path (such as src/HOL/Library) has been discontinued.
  1118 Note that the path element ~~ may be used to reference theories in the
  1119 Isabelle home folder -- for instance, "~~/src/HOL/Library/FuncSet".
  1120 INCOMPATIBILITY.
  1121 
  1122 * Theory loader: source files are identified by content via SHA1
  1123 digests.  Discontinued former path/modtime identification and optional
  1124 ISABELLE_FILE_IDENT plugin scripts.
  1125 
  1126 * Parallelization of nested Isar proofs is subject to
  1127 Goal.parallel_proofs_threshold (default 100).  See also isabelle
  1128 usedir option -Q.
  1129 
  1130 * Name space: former unsynchronized references are now proper
  1131 configuration options, with more conventional names:
  1132 
  1133   long_names   ~> names_long
  1134   short_names  ~> names_short
  1135   unique_names ~> names_unique
  1136 
  1137 Minor INCOMPATIBILITY, need to declare options in context like this:
  1138 
  1139   declare [[names_unique = false]]
  1140 
  1141 * Literal facts `prop` may contain dummy patterns, e.g. `_ = _`.  Note
  1142 that the result needs to be unique, which means fact specifications
  1143 may have to be refined after enriching a proof context.
  1144 
  1145 * Attribute "case_names" has been refined: the assumptions in each case
  1146 can be named now by following the case name with [name1 name2 ...].
  1147 
  1148 * Isabelle/Isar reference manual has been updated and extended:
  1149   - "Synopsis" provides a catalog of main Isar language concepts.
  1150   - Formal references in syntax diagrams, via @{rail} antiquotation.
  1151   - Updated material from classic "ref" manual, notably about
  1152     "Classical Reasoner".
  1153 
  1154 
  1155 *** HOL ***
  1156 
  1157 * Class bot and top require underlying partial order rather than
  1158 preorder: uniqueness of bot and top is guaranteed.  INCOMPATIBILITY.
  1159 
  1160 * Class complete_lattice: generalized a couple of lemmas from sets;
  1161 generalized theorems INF_cong and SUP_cong.  New type classes for
  1162 complete boolean algebras and complete linear orders.  Lemmas
  1163 Inf_less_iff, less_Sup_iff, INF_less_iff, less_SUP_iff now reside in
  1164 class complete_linorder.
  1165 
  1166 Changed proposition of lemmas Inf_bool_def, Sup_bool_def, Inf_fun_def,
  1167 Sup_fun_def, Inf_apply, Sup_apply.
  1168 
  1169 Removed redundant lemmas (the right hand side gives hints how to
  1170 replace them for (metis ...), or (simp only: ...) proofs):
  1171 
  1172   Inf_singleton ~> Inf_insert [where A="{}", unfolded Inf_empty inf_top_right]
  1173   Sup_singleton ~> Sup_insert [where A="{}", unfolded Sup_empty sup_bot_right]
  1174   Inf_binary ~> Inf_insert, Inf_empty, and inf_top_right
  1175   Sup_binary ~> Sup_insert, Sup_empty, and sup_bot_right
  1176   Int_eq_Inter ~> Inf_insert, Inf_empty, and inf_top_right
  1177   Un_eq_Union ~> Sup_insert, Sup_empty, and sup_bot_right
  1178   Inter_def ~> INF_def, image_def
  1179   Union_def ~> SUP_def, image_def
  1180   INT_eq ~> INF_def, and image_def
  1181   UN_eq ~> SUP_def, and image_def
  1182   INF_subset ~> INF_superset_mono [OF _ order_refl]
  1183 
  1184 More consistent and comprehensive names:
  1185 
  1186   INTER_eq_Inter_image ~> INF_def
  1187   UNION_eq_Union_image ~> SUP_def
  1188   INFI_def ~> INF_def
  1189   SUPR_def ~> SUP_def
  1190   INF_leI ~> INF_lower
  1191   INF_leI2 ~> INF_lower2
  1192   le_INFI ~> INF_greatest
  1193   le_SUPI ~> SUP_upper
  1194   le_SUPI2 ~> SUP_upper2
  1195   SUP_leI ~> SUP_least
  1196   INFI_bool_eq ~> INF_bool_eq
  1197   SUPR_bool_eq ~> SUP_bool_eq
  1198   INFI_apply ~> INF_apply
  1199   SUPR_apply ~> SUP_apply
  1200   INTER_def ~> INTER_eq
  1201   UNION_def ~> UNION_eq
  1202 
  1203 INCOMPATIBILITY.
  1204 
  1205 * Renamed theory Complete_Lattice to Complete_Lattices.
  1206 INCOMPATIBILITY.
  1207 
  1208 * Theory Complete_Lattices: lemmas Inf_eq_top_iff, INF_eq_top_iff,
  1209 INF_image, Inf_insert, INF_top, Inf_top_conv, INF_top_conv, SUP_bot,
  1210 Sup_bot_conv, SUP_bot_conv, Sup_eq_top_iff, SUP_eq_top_iff, SUP_image,
  1211 Sup_insert are now declared as [simp].  INCOMPATIBILITY.
  1212 
  1213 * Theory Lattice: lemmas compl_inf_bot, compl_le_comp_iff,
  1214 compl_sup_top, inf_idem, inf_left_idem, inf_sup_absorb, sup_idem,
  1215 sup_inf_absob, sup_left_idem are now declared as [simp].  Minor
  1216 INCOMPATIBILITY.
  1217 
  1218 * Added syntactic classes "inf" and "sup" for the respective
  1219 constants.  INCOMPATIBILITY: Changes in the argument order of the
  1220 (mostly internal) locale predicates for some derived classes.
  1221 
  1222 * Theorem collections ball_simps and bex_simps do not contain theorems
  1223 referring to UNION any longer; these have been moved to collection
  1224 UN_ball_bex_simps.  INCOMPATIBILITY.
  1225 
  1226 * Theory Archimedean_Field: floor now is defined as parameter of a
  1227 separate type class floor_ceiling.
  1228 
  1229 * Theory Finite_Set: more coherent development of fold_set locales:
  1230 
  1231     locale fun_left_comm ~> locale comp_fun_commute
  1232     locale fun_left_comm_idem ~> locale comp_fun_idem
  1233 
  1234 Both use point-free characterization; interpretation proofs may need
  1235 adjustment.  INCOMPATIBILITY.
  1236 
  1237 * Theory Limits: Type "'a net" has been renamed to "'a filter", in
  1238 accordance with standard mathematical terminology. INCOMPATIBILITY.
  1239 
  1240 * Theory Complex_Main: The locale interpretations for the
  1241 bounded_linear and bounded_bilinear locales have been removed, in
  1242 order to reduce the number of duplicate lemmas. Users must use the
  1243 original names for distributivity theorems, potential INCOMPATIBILITY.
  1244 
  1245   divide.add ~> add_divide_distrib
  1246   divide.diff ~> diff_divide_distrib
  1247   divide.setsum ~> setsum_divide_distrib
  1248   mult.add_right ~> right_distrib
  1249   mult.diff_right ~> right_diff_distrib
  1250   mult_right.setsum ~> setsum_right_distrib
  1251   mult_left.diff ~> left_diff_distrib
  1252 
  1253 * Theory Complex_Main: Several redundant theorems have been removed or
  1254 replaced by more general versions. INCOMPATIBILITY.
  1255 
  1256   real_diff_def ~> minus_real_def
  1257   real_divide_def ~> divide_real_def
  1258   real_less_def ~> less_le
  1259   real_abs_def ~> abs_real_def
  1260   real_sgn_def ~> sgn_real_def
  1261   real_mult_commute ~> mult_commute
  1262   real_mult_assoc ~> mult_assoc
  1263   real_mult_1 ~> mult_1_left
  1264   real_add_mult_distrib ~> left_distrib
  1265   real_zero_not_eq_one ~> zero_neq_one
  1266   real_mult_inverse_left ~> left_inverse
  1267   INVERSE_ZERO ~> inverse_zero
  1268   real_le_refl ~> order_refl
  1269   real_le_antisym ~> order_antisym
  1270   real_le_trans ~> order_trans
  1271   real_le_linear ~> linear
  1272   real_le_eq_diff ~> le_iff_diff_le_0
  1273   real_add_left_mono ~> add_left_mono
  1274   real_mult_order ~> mult_pos_pos
  1275   real_mult_less_mono2 ~> mult_strict_left_mono
  1276   real_of_int_real_of_nat ~> real_of_int_of_nat_eq
  1277   real_0_le_divide_iff ~> zero_le_divide_iff
  1278   realpow_two_disj ~> power2_eq_iff
  1279   real_squared_diff_one_factored ~> square_diff_one_factored
  1280   realpow_two_diff ~> square_diff_square_factored
  1281   reals_complete2 ~> complete_real
  1282   real_sum_squared_expand ~> power2_sum
  1283   exp_ln_eq ~> ln_unique
  1284   expi_add ~> exp_add
  1285   expi_zero ~> exp_zero
  1286   lemma_DERIV_subst ~> DERIV_cong
  1287   LIMSEQ_Zfun_iff ~> tendsto_Zfun_iff
  1288   LIMSEQ_const ~> tendsto_const
  1289   LIMSEQ_norm ~> tendsto_norm
  1290   LIMSEQ_add ~> tendsto_add
  1291   LIMSEQ_minus ~> tendsto_minus
  1292   LIMSEQ_minus_cancel ~> tendsto_minus_cancel
  1293   LIMSEQ_diff ~> tendsto_diff
  1294   bounded_linear.LIMSEQ ~> bounded_linear.tendsto
  1295   bounded_bilinear.LIMSEQ ~> bounded_bilinear.tendsto
  1296   LIMSEQ_mult ~> tendsto_mult
  1297   LIMSEQ_inverse ~> tendsto_inverse
  1298   LIMSEQ_divide ~> tendsto_divide
  1299   LIMSEQ_pow ~> tendsto_power
  1300   LIMSEQ_setsum ~> tendsto_setsum
  1301   LIMSEQ_setprod ~> tendsto_setprod
  1302   LIMSEQ_norm_zero ~> tendsto_norm_zero_iff
  1303   LIMSEQ_rabs_zero ~> tendsto_rabs_zero_iff
  1304   LIMSEQ_imp_rabs ~> tendsto_rabs
  1305   LIMSEQ_add_minus ~> tendsto_add [OF _ tendsto_minus]
  1306   LIMSEQ_add_const ~> tendsto_add [OF _ tendsto_const]
  1307   LIMSEQ_diff_const ~> tendsto_diff [OF _ tendsto_const]
  1308   LIMSEQ_Complex ~> tendsto_Complex
  1309   LIM_ident ~> tendsto_ident_at
  1310   LIM_const ~> tendsto_const
  1311   LIM_add ~> tendsto_add
  1312   LIM_add_zero ~> tendsto_add_zero
  1313   LIM_minus ~> tendsto_minus
  1314   LIM_diff ~> tendsto_diff
  1315   LIM_norm ~> tendsto_norm
  1316   LIM_norm_zero ~> tendsto_norm_zero
  1317   LIM_norm_zero_cancel ~> tendsto_norm_zero_cancel
  1318   LIM_norm_zero_iff ~> tendsto_norm_zero_iff
  1319   LIM_rabs ~> tendsto_rabs
  1320   LIM_rabs_zero ~> tendsto_rabs_zero
  1321   LIM_rabs_zero_cancel ~> tendsto_rabs_zero_cancel
  1322   LIM_rabs_zero_iff ~> tendsto_rabs_zero_iff
  1323   LIM_compose ~> tendsto_compose
  1324   LIM_mult ~> tendsto_mult
  1325   LIM_scaleR ~> tendsto_scaleR
  1326   LIM_of_real ~> tendsto_of_real
  1327   LIM_power ~> tendsto_power
  1328   LIM_inverse ~> tendsto_inverse
  1329   LIM_sgn ~> tendsto_sgn
  1330   isCont_LIM_compose ~> isCont_tendsto_compose
  1331   bounded_linear.LIM ~> bounded_linear.tendsto
  1332   bounded_linear.LIM_zero ~> bounded_linear.tendsto_zero
  1333   bounded_bilinear.LIM ~> bounded_bilinear.tendsto
  1334   bounded_bilinear.LIM_prod_zero ~> bounded_bilinear.tendsto_zero
  1335   bounded_bilinear.LIM_left_zero ~> bounded_bilinear.tendsto_left_zero
  1336   bounded_bilinear.LIM_right_zero ~> bounded_bilinear.tendsto_right_zero
  1337   LIM_inverse_fun ~> tendsto_inverse [OF tendsto_ident_at]
  1338 
  1339 * Theory Complex_Main: The definition of infinite series was
  1340 generalized.  Now it is defined on the type class {topological_space,
  1341 comm_monoid_add}.  Hence it is useable also for extended real numbers.
  1342 
  1343 * Theory Complex_Main: The complex exponential function "expi" is now
  1344 a type-constrained abbreviation for "exp :: complex => complex"; thus
  1345 several polymorphic lemmas about "exp" are now applicable to "expi".
  1346 
  1347 * Code generation:
  1348 
  1349   - Theory Library/Code_Char_ord provides native ordering of
  1350     characters in the target language.
  1351 
  1352   - Commands code_module and code_library are legacy, use export_code
  1353     instead.
  1354 
  1355   - Method "evaluation" is legacy, use method "eval" instead.
  1356 
  1357   - Legacy evaluator "SML" is deactivated by default.  May be
  1358     reactivated by the following theory command:
  1359 
  1360       setup {* Value.add_evaluator ("SML", Codegen.eval_term) *}
  1361 
  1362 * Declare ext [intro] by default.  Rare INCOMPATIBILITY.
  1363 
  1364 * New proof method "induction" that gives induction hypotheses the
  1365 name "IH", thus distinguishing them from further hypotheses that come
  1366 from rule induction.  The latter are still called "hyps".  Method
  1367 "induction" is a thin wrapper around "induct" and follows the same
  1368 syntax.
  1369 
  1370 * Method "fastsimp" has been renamed to "fastforce", but "fastsimp" is
  1371 still available as a legacy feature for some time.
  1372 
  1373 * Nitpick:
  1374   - Added "need" and "total_consts" options.
  1375   - Reintroduced "show_skolems" option by popular demand.
  1376   - Renamed attribute: nitpick_def ~> nitpick_unfold.
  1377     INCOMPATIBILITY.
  1378 
  1379 * Sledgehammer:
  1380   - Use quasi-sound (and efficient) translations by default.
  1381   - Added support for the following provers: E-ToFoF, LEO-II,
  1382     Satallax, SNARK, Waldmeister, and Z3 with TPTP syntax.
  1383   - Automatically preplay and minimize proofs before showing them if
  1384     this can be done within reasonable time.
  1385   - sledgehammer available_provers ~> sledgehammer supported_provers.
  1386     INCOMPATIBILITY.
  1387   - Added "preplay_timeout", "slicing", "type_enc", "sound",
  1388     "max_mono_iters", and "max_new_mono_instances" options.
  1389   - Removed "explicit_apply" and "full_types" options as well as "Full
  1390     Types" Proof General menu item. INCOMPATIBILITY.
  1391 
  1392 * Metis:
  1393   - Removed "metisF" -- use "metis" instead. INCOMPATIBILITY.
  1394   - Obsoleted "metisFT" -- use "metis (full_types)" instead.
  1395     INCOMPATIBILITY.
  1396 
  1397 * Command 'try':
  1398   - Renamed 'try_methods' and added "simp:", "intro:", "dest:", and
  1399     "elim:" options. INCOMPATIBILITY.
  1400   - Introduced 'try' that not only runs 'try_methods' but also
  1401     'solve_direct', 'sledgehammer', 'quickcheck', and 'nitpick'.
  1402 
  1403 * Quickcheck:
  1404   - Added "eval" option to evaluate terms for the found counterexample
  1405     (currently only supported by the default (exhaustive) tester).
  1406   - Added post-processing of terms to obtain readable counterexamples
  1407     (currently only supported by the default (exhaustive) tester).
  1408   - New counterexample generator quickcheck[narrowing] enables
  1409     narrowing-based testing.  Requires the Glasgow Haskell compiler
  1410     with its installation location defined in the Isabelle settings
  1411     environment as ISABELLE_GHC.
  1412   - Removed quickcheck tester "SML" based on the SML code generator
  1413     (formly in HOL/Library).
  1414 
  1415 * Function package: discontinued option "tailrec".  INCOMPATIBILITY,
  1416 use 'partial_function' instead.
  1417 
  1418 * Theory Library/Extended_Reals replaces now the positive extended
  1419 reals found in probability theory. This file is extended by
  1420 Multivariate_Analysis/Extended_Real_Limits.
  1421 
  1422 * Theory Library/Old_Recdef: old 'recdef' package has been moved here,
  1423 from where it must be imported explicitly if it is really required.
  1424 INCOMPATIBILITY.
  1425 
  1426 * Theory Library/Wfrec: well-founded recursion combinator "wfrec" has
  1427 been moved here.  INCOMPATIBILITY.
  1428 
  1429 * Theory Library/Saturated provides type of numbers with saturated
  1430 arithmetic.
  1431 
  1432 * Theory Library/Product_Lattice defines a pointwise ordering for the
  1433 product type 'a * 'b, and provides instance proofs for various order
  1434 and lattice type classes.
  1435 
  1436 * Theory Library/Countable now provides the "countable_datatype" proof
  1437 method for proving "countable" class instances for datatypes.
  1438 
  1439 * Theory Library/Cset_Monad allows do notation for computable sets
  1440 (cset) via the generic monad ad-hoc overloading facility.
  1441 
  1442 * Library: Theories of common data structures are split into theories
  1443 for implementation, an invariant-ensuring type, and connection to an
  1444 abstract type. INCOMPATIBILITY.
  1445 
  1446   - RBT is split into RBT and RBT_Mapping.
  1447   - AssocList is split and renamed into AList and AList_Mapping.
  1448   - DList is split into DList_Impl, DList, and DList_Cset.
  1449   - Cset is split into Cset and List_Cset.
  1450 
  1451 * Theory Library/Nat_Infinity has been renamed to
  1452 Library/Extended_Nat, with name changes of the following types and
  1453 constants:
  1454 
  1455   type inat   ~> type enat
  1456   Fin         ~> enat
  1457   Infty       ~> infinity (overloaded)
  1458   iSuc        ~> eSuc
  1459   the_Fin     ~> the_enat
  1460 
  1461 Every theorem name containing "inat", "Fin", "Infty", or "iSuc" has
  1462 been renamed accordingly. INCOMPATIBILITY.
  1463 
  1464 * Session Multivariate_Analysis: The euclidean_space type class now
  1465 fixes a constant "Basis :: 'a set" consisting of the standard
  1466 orthonormal basis for the type. Users now have the option of
  1467 quantifying over this set instead of using the "basis" function, e.g.
  1468 "ALL x:Basis. P x" vs "ALL i<DIM('a). P (basis i)".
  1469 
  1470 * Session Multivariate_Analysis: Type "('a, 'b) cart" has been renamed
  1471 to "('a, 'b) vec" (the syntax "'a ^ 'b" remains unaffected). Constants
  1472 "Cart_nth" and "Cart_lambda" have been respectively renamed to
  1473 "vec_nth" and "vec_lambda"; theorems mentioning those names have
  1474 changed to match. Definition theorems for overloaded constants now use
  1475 the standard "foo_vec_def" naming scheme. A few other theorems have
  1476 been renamed as follows (INCOMPATIBILITY):
  1477 
  1478   Cart_eq          ~> vec_eq_iff
  1479   dist_nth_le_cart ~> dist_vec_nth_le
  1480   tendsto_vector   ~> vec_tendstoI
  1481   Cauchy_vector    ~> vec_CauchyI
  1482 
  1483 * Session Multivariate_Analysis: Several duplicate theorems have been
  1484 removed, and other theorems have been renamed or replaced with more
  1485 general versions. INCOMPATIBILITY.
  1486 
  1487   finite_choice ~> finite_set_choice
  1488   eventually_conjI ~> eventually_conj
  1489   eventually_and ~> eventually_conj_iff
  1490   eventually_false ~> eventually_False
  1491   setsum_norm ~> norm_setsum
  1492   Lim_sequentially ~> LIMSEQ_def
  1493   Lim_ident_at ~> LIM_ident
  1494   Lim_const ~> tendsto_const
  1495   Lim_cmul ~> tendsto_scaleR [OF tendsto_const]
  1496   Lim_neg ~> tendsto_minus
  1497   Lim_add ~> tendsto_add
  1498   Lim_sub ~> tendsto_diff
  1499   Lim_mul ~> tendsto_scaleR
  1500   Lim_vmul ~> tendsto_scaleR [OF _ tendsto_const]
  1501   Lim_null_norm ~> tendsto_norm_zero_iff [symmetric]
  1502   Lim_linear ~> bounded_linear.tendsto
  1503   Lim_component ~> tendsto_euclidean_component
  1504   Lim_component_cart ~> tendsto_vec_nth
  1505   Lim_inner ~> tendsto_inner [OF tendsto_const]
  1506   dot_lsum ~> inner_setsum_left
  1507   dot_rsum ~> inner_setsum_right
  1508   continuous_cmul ~> continuous_scaleR [OF continuous_const]
  1509   continuous_neg ~> continuous_minus
  1510   continuous_sub ~> continuous_diff
  1511   continuous_vmul ~> continuous_scaleR [OF _ continuous_const]
  1512   continuous_mul ~> continuous_scaleR
  1513   continuous_inv ~> continuous_inverse
  1514   continuous_at_within_inv ~> continuous_at_within_inverse
  1515   continuous_at_inv ~> continuous_at_inverse
  1516   continuous_at_norm ~> continuous_norm [OF continuous_at_id]
  1517   continuous_at_infnorm ~> continuous_infnorm [OF continuous_at_id]
  1518   continuous_at_component ~> continuous_component [OF continuous_at_id]
  1519   continuous_on_neg ~> continuous_on_minus
  1520   continuous_on_sub ~> continuous_on_diff
  1521   continuous_on_cmul ~> continuous_on_scaleR [OF continuous_on_const]
  1522   continuous_on_vmul ~> continuous_on_scaleR [OF _ continuous_on_const]
  1523   continuous_on_mul ~> continuous_on_scaleR
  1524   continuous_on_mul_real ~> continuous_on_mult
  1525   continuous_on_inner ~> continuous_on_inner [OF continuous_on_const]
  1526   continuous_on_norm ~> continuous_on_norm [OF continuous_on_id]
  1527   continuous_on_inverse ~> continuous_on_inv
  1528   uniformly_continuous_on_neg ~> uniformly_continuous_on_minus
  1529   uniformly_continuous_on_sub ~> uniformly_continuous_on_diff
  1530   subset_interior ~> interior_mono
  1531   subset_closure ~> closure_mono
  1532   closure_univ ~> closure_UNIV
  1533   real_arch_lt ~> reals_Archimedean2
  1534   real_arch ~> reals_Archimedean3
  1535   real_abs_norm ~> abs_norm_cancel
  1536   real_abs_sub_norm ~> norm_triangle_ineq3
  1537   norm_cauchy_schwarz_abs ~> Cauchy_Schwarz_ineq2
  1538 
  1539 * Session HOL-Probability:
  1540   - Caratheodory's extension lemma is now proved for ring_of_sets.
  1541   - Infinite products of probability measures are now available.
  1542   - Sigma closure is independent, if the generator is independent
  1543   - Use extended reals instead of positive extended
  1544     reals. INCOMPATIBILITY.
  1545 
  1546 * Session HOLCF: Discontinued legacy theorem names, INCOMPATIBILITY.
  1547 
  1548   expand_fun_below ~> fun_below_iff
  1549   below_fun_ext ~> fun_belowI
  1550   expand_cfun_eq ~> cfun_eq_iff
  1551   ext_cfun ~> cfun_eqI
  1552   expand_cfun_below ~> cfun_below_iff
  1553   below_cfun_ext ~> cfun_belowI
  1554   monofun_fun_fun ~> fun_belowD
  1555   monofun_fun_arg ~> monofunE
  1556   monofun_lub_fun ~> adm_monofun [THEN admD]
  1557   cont_lub_fun ~> adm_cont [THEN admD]
  1558   cont2cont_Rep_CFun ~> cont2cont_APP
  1559   cont_Rep_CFun_app ~> cont_APP_app
  1560   cont_Rep_CFun_app_app ~> cont_APP_app_app
  1561   cont_cfun_fun ~> cont_Rep_cfun1 [THEN contE]
  1562   cont_cfun_arg ~> cont_Rep_cfun2 [THEN contE]
  1563   contlub_cfun ~> lub_APP [symmetric]
  1564   contlub_LAM ~> lub_LAM [symmetric]
  1565   thelubI ~> lub_eqI
  1566   UU_I ~> bottomI
  1567   lift_distinct1 ~> lift.distinct(1)
  1568   lift_distinct2 ~> lift.distinct(2)
  1569   Def_not_UU ~> lift.distinct(2)
  1570   Def_inject ~> lift.inject
  1571   below_UU_iff ~> below_bottom_iff
  1572   eq_UU_iff ~> eq_bottom_iff
  1573 
  1574 
  1575 *** Document preparation ***
  1576 
  1577 * Antiquotation @{rail} layouts railroad syntax diagrams, see also
  1578 isar-ref manual, both for description and actual application of the
  1579 same.
  1580 
  1581 * Antiquotation @{value} evaluates the given term and presents its
  1582 result.
  1583 
  1584 * Antiquotations: term style "isub" provides ad-hoc conversion of
  1585 variables x1, y23 into subscripted form x\<^isub>1,
  1586 y\<^isub>2\<^isub>3.
  1587 
  1588 * Predefined LaTeX macros for Isabelle symbols \<bind> and \<then>
  1589 (e.g. see ~~/src/HOL/Library/Monad_Syntax.thy).
  1590 
  1591 * Localized \isabellestyle switch can be used within blocks or groups
  1592 like this:
  1593 
  1594   \isabellestyle{it}  %preferred default
  1595   {\isabellestylett @{text "typewriter stuff"}}
  1596 
  1597 * Discontinued special treatment of hard tabulators.  Implicit
  1598 tab-width is now defined as 1.  Potential INCOMPATIBILITY for visual
  1599 layouts.
  1600 
  1601 
  1602 *** ML ***
  1603 
  1604 * The inner syntax of sort/type/term/prop supports inlined YXML
  1605 representations within quoted string tokens.  By encoding logical
  1606 entities via Term_XML (in ML or Scala) concrete syntax can be
  1607 bypassed, which is particularly useful for producing bits of text
  1608 under external program control.
  1609 
  1610 * Antiquotations for ML and document preparation are managed as theory
  1611 data, which requires explicit setup.
  1612 
  1613 * Isabelle_Process.is_active allows tools to check if the official
  1614 process wrapper is running (Isabelle/Scala/jEdit) or the old TTY loop
  1615 (better known as Proof General).
  1616 
  1617 * Structure Proof_Context follows standard naming scheme.  Old
  1618 ProofContext is still available for some time as legacy alias.
  1619 
  1620 * Structure Timing provides various operations for timing; supersedes
  1621 former start_timing/end_timing etc.
  1622 
  1623 * Path.print is the official way to show file-system paths to users
  1624 (including quotes etc.).
  1625 
  1626 * Inner syntax: identifiers in parse trees of generic categories
  1627 "logic", "aprop", "idt" etc. carry position information (disguised as
  1628 type constraints).  Occasional INCOMPATIBILITY with non-compliant
  1629 translations that choke on unexpected type constraints.  Positions can
  1630 be stripped in ML translations via Syntax.strip_positions /
  1631 Syntax.strip_positions_ast, or via the syntax constant
  1632 "_strip_positions" within parse trees.  As last resort, positions can
  1633 be disabled via the configuration option Syntax.positions, which is
  1634 called "syntax_positions" in Isar attribute syntax.
  1635 
  1636 * Discontinued special status of various ML structures that contribute
  1637 to structure Syntax (Ast, Lexicon, Mixfix, Parser, Printer etc.): less
  1638 pervasive content, no inclusion in structure Syntax.  INCOMPATIBILITY,
  1639 refer directly to Ast.Constant, Lexicon.is_identifier,
  1640 Syntax_Trans.mk_binder_tr etc.
  1641 
  1642 * Typed print translation: discontinued show_sorts argument, which is
  1643 already available via context of "advanced" translation.
  1644 
  1645 * Refined PARALLEL_GOALS tactical: degrades gracefully for schematic
  1646 goal states; body tactic needs to address all subgoals uniformly.
  1647 
  1648 * Slightly more special eq_list/eq_set, with shortcut involving
  1649 pointer equality (assumes that eq relation is reflexive).
  1650 
  1651 * Classical tactics use proper Proof.context instead of historic types
  1652 claset/clasimpset.  Old-style declarations like addIs, addEs, addDs
  1653 operate directly on Proof.context.  Raw type claset retains its use as
  1654 snapshot of the classical context, which can be recovered via
  1655 (put_claset HOL_cs) etc.  Type clasimpset has been discontinued.
  1656 INCOMPATIBILITY, classical tactics and derived proof methods require
  1657 proper Proof.context.
  1658 
  1659 
  1660 *** System ***
  1661 
  1662 * Discontinued support for Poly/ML 5.2, which was the last version
  1663 without proper multithreading and TimeLimit implementation.
  1664 
  1665 * Discontinued old lib/scripts/polyml-platform, which has been
  1666 obsolete since Isabelle2009-2.
  1667 
  1668 * Various optional external tools are referenced more robustly and
  1669 uniformly by explicit Isabelle settings as follows:
  1670 
  1671   ISABELLE_CSDP   (formerly CSDP_EXE)
  1672   ISABELLE_GHC    (formerly EXEC_GHC or GHC_PATH)
  1673   ISABELLE_OCAML  (formerly EXEC_OCAML)
  1674   ISABELLE_SWIPL  (formerly EXEC_SWIPL)
  1675   ISABELLE_YAP    (formerly EXEC_YAP)
  1676 
  1677 Note that automated detection from the file-system or search path has
  1678 been discontinued.  INCOMPATIBILITY.
  1679 
  1680 * Scala layer provides JVM method invocation service for static
  1681 methods of type (String)String, see Invoke_Scala.method in ML.  For
  1682 example:
  1683 
  1684   Invoke_Scala.method "java.lang.System.getProperty" "java.home"
  1685 
  1686 Together with YXML.string_of_body/parse_body and XML.Encode/Decode
  1687 this allows to pass structured values between ML and Scala.
  1688 
  1689 * The IsabelleText fonts includes some further glyphs to support the
  1690 Prover IDE.  Potential INCOMPATIBILITY: users who happen to have
  1691 installed a local copy (which is normally *not* required) need to
  1692 delete or update it from ~~/lib/fonts/.
  1693 
  1694 
  1695 
  1696 New in Isabelle2011 (January 2011)
  1697 ----------------------------------
  1698 
  1699 *** General ***
  1700 
  1701 * Experimental Prover IDE based on Isabelle/Scala and jEdit (see
  1702 src/Tools/jEdit).  This also serves as IDE for Isabelle/ML, with
  1703 useful tooltips and hyperlinks produced from its static analysis.  The
  1704 bundled component provides an executable Isabelle tool that can be run
  1705 like this:
  1706 
  1707   Isabelle2011/bin/isabelle jedit
  1708 
  1709 * Significantly improved Isabelle/Isar implementation manual.
  1710 
  1711 * System settings: ISABELLE_HOME_USER now includes ISABELLE_IDENTIFIER
  1712 (and thus refers to something like $HOME/.isabelle/Isabelle2011),
  1713 while the default heap location within that directory lacks that extra
  1714 suffix.  This isolates multiple Isabelle installations from each
  1715 other, avoiding problems with old settings in new versions.
  1716 INCOMPATIBILITY, need to copy/upgrade old user settings manually.
  1717 
  1718 * Source files are always encoded as UTF-8, instead of old-fashioned
  1719 ISO-Latin-1.  INCOMPATIBILITY.  Isabelle LaTeX documents might require
  1720 the following package declarations:
  1721 
  1722   \usepackage[utf8]{inputenc}
  1723   \usepackage{textcomp}
  1724 
  1725 * Explicit treatment of UTF-8 sequences as Isabelle symbols, such that
  1726 a Unicode character is treated as a single symbol, not a sequence of
  1727 non-ASCII bytes as before.  Since Isabelle/ML string literals may
  1728 contain symbols without further backslash escapes, Unicode can now be
  1729 used here as well.  Recall that Symbol.explode in ML provides a
  1730 consistent view on symbols, while raw explode (or String.explode)
  1731 merely give a byte-oriented representation.
  1732 
  1733 * Theory loader: source files are primarily located via the master
  1734 directory of each theory node (where the .thy file itself resides).
  1735 The global load path is still partially available as legacy feature.
  1736 Minor INCOMPATIBILITY due to subtle change in file lookup: use
  1737 explicit paths, relatively to the theory.
  1738 
  1739 * Special treatment of ML file names has been discontinued.
  1740 Historically, optional extensions .ML or .sml were added on demand --
  1741 at the cost of clarity of file dependencies.  Recall that Isabelle/ML
  1742 files exclusively use the .ML extension.  Minor INCOMPATIBILTY.
  1743 
  1744 * Various options that affect pretty printing etc. are now properly
  1745 handled within the context via configuration options, instead of
  1746 unsynchronized references or print modes.  There are both ML Config.T
  1747 entities and Isar declaration attributes to access these.
  1748 
  1749   ML (Config.T)                 Isar (attribute)
  1750 
  1751   eta_contract                  eta_contract
  1752   show_brackets                 show_brackets
  1753   show_sorts                    show_sorts
  1754   show_types                    show_types
  1755   show_question_marks           show_question_marks
  1756   show_consts                   show_consts
  1757   show_abbrevs                  show_abbrevs
  1758 
  1759   Syntax.ast_trace              syntax_ast_trace
  1760   Syntax.ast_stat               syntax_ast_stat
  1761   Syntax.ambiguity_level        syntax_ambiguity_level
  1762 
  1763   Goal_Display.goals_limit      goals_limit
  1764   Goal_Display.show_main_goal   show_main_goal
  1765 
  1766   Method.rule_trace             rule_trace
  1767 
  1768   Thy_Output.display            thy_output_display
  1769   Thy_Output.quotes             thy_output_quotes
  1770   Thy_Output.indent             thy_output_indent
  1771   Thy_Output.source             thy_output_source
  1772   Thy_Output.break              thy_output_break
  1773 
  1774 Note that corresponding "..._default" references in ML may only be
  1775 changed globally at the ROOT session setup, but *not* within a theory.
  1776 The option "show_abbrevs" supersedes the former print mode
  1777 "no_abbrevs" with inverted meaning.
  1778 
  1779 * More systematic naming of some configuration options.
  1780 INCOMPATIBILITY.
  1781 
  1782   trace_simp  ~>  simp_trace
  1783   debug_simp  ~>  simp_debug
  1784 
  1785 * Support for real valued configuration options, using simplistic
  1786 floating-point notation that coincides with the inner syntax for
  1787 float_token.
  1788 
  1789 * Support for real valued preferences (with approximative PGIP type):
  1790 front-ends need to accept "pgint" values in float notation.
  1791 INCOMPATIBILITY.
  1792 
  1793 * The IsabelleText font now includes Cyrillic, Hebrew, Arabic from
  1794 DejaVu Sans.
  1795 
  1796 * Discontinued support for Poly/ML 5.0 and 5.1 versions.
  1797 
  1798 
  1799 *** Pure ***
  1800 
  1801 * Command 'type_synonym' (with single argument) replaces somewhat
  1802 outdated 'types', which is still available as legacy feature for some
  1803 time.
  1804 
  1805 * Command 'nonterminal' (with 'and' separated list of arguments)
  1806 replaces somewhat outdated 'nonterminals'.  INCOMPATIBILITY.
  1807 
  1808 * Command 'notepad' replaces former 'example_proof' for
  1809 experimentation in Isar without any result.  INCOMPATIBILITY.
  1810 
  1811 * Locale interpretation commands 'interpret' and 'sublocale' accept
  1812 lists of equations to map definitions in a locale to appropriate
  1813 entities in the context of the interpretation.  The 'interpretation'
  1814 command already provided this functionality.
  1815 
  1816 * Diagnostic command 'print_dependencies' prints the locale instances
  1817 that would be activated if the specified expression was interpreted in
  1818 the current context.  Variant "print_dependencies!" assumes a context
  1819 without interpretations.
  1820 
  1821 * Diagnostic command 'print_interps' prints interpretations in proofs
  1822 in addition to interpretations in theories.
  1823 
  1824 * Discontinued obsolete 'global' and 'local' commands to manipulate
  1825 the theory name space.  Rare INCOMPATIBILITY.  The ML functions
  1826 Sign.root_path and Sign.local_path may be applied directly where this
  1827 feature is still required for historical reasons.
  1828 
  1829 * Discontinued obsolete 'constdefs' command.  INCOMPATIBILITY, use
  1830 'definition' instead.
  1831 
  1832 * The "prems" fact, which refers to the accidental collection of
  1833 foundational premises in the context, is now explicitly marked as
  1834 legacy feature and will be discontinued soon.  Consider using "assms"
  1835 of the head statement or reference facts by explicit names.
  1836 
  1837 * Document antiquotations @{class} and @{type} print classes and type
  1838 constructors.
  1839 
  1840 * Document antiquotation @{file} checks file/directory entries within
  1841 the local file system.
  1842 
  1843 
  1844 *** HOL ***
  1845 
  1846 * Coercive subtyping: functions can be declared as coercions and type
  1847 inference will add them as necessary upon input of a term.  Theory
  1848 Complex_Main declares real :: nat => real and real :: int => real as
  1849 coercions. A coercion function f is declared like this:
  1850 
  1851   declare [[coercion f]]
  1852 
  1853 To lift coercions through type constructors (e.g. from nat => real to
  1854 nat list => real list), map functions can be declared, e.g.
  1855 
  1856   declare [[coercion_map map]]
  1857 
  1858 Currently coercion inference is activated only in theories including
  1859 real numbers, i.e. descendants of Complex_Main.  This is controlled by
  1860 the configuration option "coercion_enabled", e.g. it can be enabled in
  1861 other theories like this:
  1862 
  1863   declare [[coercion_enabled]]
  1864 
  1865 * Command 'partial_function' provides basic support for recursive
  1866 function definitions over complete partial orders.  Concrete instances
  1867 are provided for i) the option type, ii) tail recursion on arbitrary
  1868 types, and iii) the heap monad of Imperative_HOL.  See
  1869 src/HOL/ex/Fundefs.thy and src/HOL/Imperative_HOL/ex/Linked_Lists.thy
  1870 for examples.
  1871 
  1872 * Function package: f.psimps rules are no longer implicitly declared
  1873 as [simp].  INCOMPATIBILITY.
  1874 
  1875 * Datatype package: theorems generated for executable equality (class
  1876 "eq") carry proper names and are treated as default code equations.
  1877 
  1878 * Inductive package: now offers command 'inductive_simps' to
  1879 automatically derive instantiated and simplified equations for
  1880 inductive predicates, similar to 'inductive_cases'.
  1881 
  1882 * Command 'enriched_type' allows to register properties of the
  1883 functorial structure of types.
  1884 
  1885 * Improved infrastructure for term evaluation using code generator
  1886 techniques, in particular static evaluation conversions.
  1887 
  1888 * Code generator: Scala (2.8 or higher) has been added to the target
  1889 languages.
  1890 
  1891 * Code generator: globbing constant expressions "*" and "Theory.*"
  1892 have been replaced by the more idiomatic "_" and "Theory._".
  1893 INCOMPATIBILITY.
  1894 
  1895 * Code generator: export_code without explicit file declaration prints
  1896 to standard output.  INCOMPATIBILITY.
  1897 
  1898 * Code generator: do not print function definitions for case
  1899 combinators any longer.
  1900 
  1901 * Code generator: simplification with rules determined with
  1902 src/Tools/Code/code_simp.ML and method "code_simp".
  1903 
  1904 * Code generator for records: more idiomatic representation of record
  1905 types.  Warning: records are not covered by ancient SML code
  1906 generation any longer.  INCOMPATIBILITY.  In cases of need, a suitable
  1907 rep_datatype declaration helps to succeed then:
  1908 
  1909   record 'a foo = ...
  1910   ...
  1911   rep_datatype foo_ext ...
  1912 
  1913 * Records: logical foundation type for records does not carry a
  1914 '_type' suffix any longer (obsolete due to authentic syntax).
  1915 INCOMPATIBILITY.
  1916 
  1917 * Quickcheck now by default uses exhaustive testing instead of random
  1918 testing.  Random testing can be invoked by "quickcheck [random]",
  1919 exhaustive testing by "quickcheck [exhaustive]".
  1920 
  1921 * Quickcheck instantiates polymorphic types with small finite
  1922 datatypes by default. This enables a simple execution mechanism to
  1923 handle quantifiers and function equality over the finite datatypes.
  1924 
  1925 * Quickcheck random generator has been renamed from "code" to
  1926 "random".  INCOMPATIBILITY.
  1927 
  1928 * Quickcheck now has a configurable time limit which is set to 30
  1929 seconds by default. This can be changed by adding [timeout = n] to the
  1930 quickcheck command. The time limit for Auto Quickcheck is still set
  1931 independently.
  1932 
  1933 * Quickcheck in locales considers interpretations of that locale for
  1934 counter example search.
  1935 
  1936 * Sledgehammer:
  1937   - Added "smt" and "remote_smt" provers based on the "smt" proof
  1938     method. See the Sledgehammer manual for details ("isabelle doc
  1939     sledgehammer").
  1940   - Renamed commands:
  1941     sledgehammer atp_info ~> sledgehammer running_provers
  1942     sledgehammer atp_kill ~> sledgehammer kill_provers
  1943     sledgehammer available_atps ~> sledgehammer available_provers
  1944     INCOMPATIBILITY.
  1945   - Renamed options:
  1946     sledgehammer [atps = ...] ~> sledgehammer [provers = ...]
  1947     sledgehammer [atp = ...] ~> sledgehammer [prover = ...]
  1948     sledgehammer [timeout = 77 s] ~> sledgehammer [timeout = 77]
  1949     (and "ms" and "min" are no longer supported)
  1950     INCOMPATIBILITY.
  1951 
  1952 * Nitpick:
  1953   - Renamed options:
  1954     nitpick [timeout = 77 s] ~> nitpick [timeout = 77]
  1955     nitpick [tac_timeout = 777 ms] ~> nitpick [tac_timeout = 0.777]
  1956     INCOMPATIBILITY.
  1957   - Added support for partial quotient types.
  1958   - Added local versions of the "Nitpick.register_xxx" functions.
  1959   - Added "whack" option.
  1960   - Allow registration of quotient types as codatatypes.
  1961   - Improved "merge_type_vars" option to merge more types.
  1962   - Removed unsound "fast_descrs" option.
  1963   - Added custom symmetry breaking for datatypes, making it possible to reach
  1964     higher cardinalities.
  1965   - Prevent the expansion of too large definitions.
  1966 
  1967 * Proof methods "metis" and "meson" now have configuration options
  1968 "meson_trace", "metis_trace", and "metis_verbose" that can be enabled
  1969 to diagnose these tools. E.g.
  1970 
  1971     using [[metis_trace = true]]
  1972 
  1973 * Auto Solve: Renamed "Auto Solve Direct".  The tool is now available
  1974 manually as command 'solve_direct'.
  1975 
  1976 * The default SMT solver Z3 must be enabled explicitly (due to
  1977 licensing issues) by setting the environment variable
  1978 Z3_NON_COMMERCIAL in etc/settings of the component, for example.  For
  1979 commercial applications, the SMT solver CVC3 is provided as fall-back;
  1980 changing the SMT solver is done via the configuration option
  1981 "smt_solver".
  1982 
  1983 * Remote SMT solvers need to be referred to by the "remote_" prefix,
  1984 i.e. "remote_cvc3" and "remote_z3".
  1985 
  1986 * Added basic SMT support for datatypes, records, and typedefs using
  1987 the oracle mode (no proofs).  Direct support of pairs has been dropped
  1988 in exchange (pass theorems fst_conv snd_conv pair_collapse to the SMT
  1989 support for a similar behavior).  Minor INCOMPATIBILITY.
  1990 
  1991 * Changed SMT configuration options:
  1992   - Renamed:
  1993     z3_proofs ~> smt_oracle (with inverted meaning)
  1994     z3_trace_assms ~> smt_trace_used_facts
  1995     INCOMPATIBILITY.
  1996   - Added:
  1997     smt_verbose
  1998     smt_random_seed
  1999     smt_datatypes
  2000     smt_infer_triggers
  2001     smt_monomorph_limit
  2002     cvc3_options
  2003     remote_cvc3_options
  2004     remote_z3_options
  2005     yices_options
  2006 
  2007 * Boogie output files (.b2i files) need to be declared in the theory
  2008 header.
  2009 
  2010 * Simplification procedure "list_to_set_comprehension" rewrites list
  2011 comprehensions applied to List.set to set comprehensions.  Occasional
  2012 INCOMPATIBILITY, may be deactivated like this:
  2013 
  2014   declare [[simproc del: list_to_set_comprehension]]
  2015 
  2016 * Removed old version of primrec package.  INCOMPATIBILITY.
  2017 
  2018 * Removed simplifier congruence rule of "prod_case", as has for long
  2019 been the case with "split".  INCOMPATIBILITY.
  2020 
  2021 * String.literal is a type, but not a datatype.  INCOMPATIBILITY.
  2022 
  2023 * Removed [split_format ... and ... and ...] version of
  2024 [split_format].  Potential INCOMPATIBILITY.
  2025 
  2026 * Predicate "sorted" now defined inductively, with nice induction
  2027 rules.  INCOMPATIBILITY: former sorted.simps now named sorted_simps.
  2028 
  2029 * Constant "contents" renamed to "the_elem", to free the generic name
  2030 contents for other uses.  INCOMPATIBILITY.
  2031 
  2032 * Renamed class eq and constant eq (for code generation) to class
  2033 equal and constant equal, plus renaming of related facts and various
  2034 tuning.  INCOMPATIBILITY.
  2035 
  2036 * Dropped type classes mult_mono and mult_mono1.  INCOMPATIBILITY.
  2037 
  2038 * Removed output syntax "'a ~=> 'b" for "'a => 'b option".
  2039 INCOMPATIBILITY.
  2040 
  2041 * Renamed theory Fset to Cset, type Fset.fset to Cset.set, in order to
  2042 avoid confusion with finite sets.  INCOMPATIBILITY.
  2043 
  2044 * Abandoned locales equiv, congruent and congruent2 for equivalence
  2045 relations.  INCOMPATIBILITY: use equivI rather than equiv_intro (same
  2046 for congruent(2)).
  2047 
  2048 * Some previously unqualified names have been qualified:
  2049 
  2050   types
  2051     bool ~> HOL.bool
  2052     nat ~> Nat.nat
  2053 
  2054   constants
  2055     Trueprop ~> HOL.Trueprop
  2056     True ~> HOL.True
  2057     False ~> HOL.False
  2058     op & ~> HOL.conj
  2059     op | ~> HOL.disj
  2060     op --> ~> HOL.implies
  2061     op = ~> HOL.eq
  2062     Not ~> HOL.Not
  2063     The ~> HOL.The
  2064     All ~> HOL.All
  2065     Ex ~> HOL.Ex
  2066     Ex1 ~> HOL.Ex1
  2067     Let ~> HOL.Let
  2068     If ~> HOL.If
  2069     Ball ~> Set.Ball
  2070     Bex ~> Set.Bex
  2071     Suc ~> Nat.Suc
  2072     Pair ~> Product_Type.Pair
  2073     fst ~> Product_Type.fst
  2074     snd ~> Product_Type.snd
  2075     curry ~> Product_Type.curry
  2076     op : ~> Set.member
  2077     Collect ~> Set.Collect
  2078 
  2079 INCOMPATIBILITY.
  2080 
  2081 * More canonical naming convention for some fundamental definitions:
  2082 
  2083     bot_bool_eq ~> bot_bool_def
  2084     top_bool_eq ~> top_bool_def
  2085     inf_bool_eq ~> inf_bool_def
  2086     sup_bool_eq ~> sup_bool_def
  2087     bot_fun_eq  ~> bot_fun_def
  2088     top_fun_eq  ~> top_fun_def
  2089     inf_fun_eq  ~> inf_fun_def
  2090     sup_fun_eq  ~> sup_fun_def
  2091 
  2092 INCOMPATIBILITY.
  2093 
  2094 * More stylized fact names:
  2095 
  2096   expand_fun_eq ~> fun_eq_iff
  2097   expand_set_eq ~> set_eq_iff
  2098   set_ext       ~> set_eqI
  2099   nat_number    ~> eval_nat_numeral
  2100 
  2101 INCOMPATIBILITY.
  2102 
  2103 * Refactoring of code-generation specific operations in theory List:
  2104 
  2105   constants
  2106     null ~> List.null
  2107 
  2108   facts
  2109     mem_iff ~> member_def
  2110     null_empty ~> null_def
  2111 
  2112 INCOMPATIBILITY.  Note that these were not supposed to be used
  2113 regularly unless for striking reasons; their main purpose was code
  2114 generation.
  2115 
  2116 Various operations from the Haskell prelude are used for generating
  2117 Haskell code.
  2118 
  2119 * Term "bij f" is now an abbreviation of "bij_betw f UNIV UNIV".  Term
  2120 "surj f" is now an abbreviation of "range f = UNIV".  The theorems
  2121 bij_def and surj_def are unchanged.  INCOMPATIBILITY.
  2122 
  2123 * Abolished some non-alphabetic type names: "prod" and "sum" replace
  2124 "*" and "+" respectively.  INCOMPATIBILITY.
  2125 
  2126 * Name "Plus" of disjoint sum operator "<+>" is now hidden.  Write
  2127 "Sum_Type.Plus" instead.
  2128 
  2129 * Constant "split" has been merged with constant "prod_case"; names of
  2130 ML functions, facts etc. involving split have been retained so far,
  2131 though.  INCOMPATIBILITY.
  2132 
  2133 * Dropped old infix syntax "_ mem _" for List.member; use "_ : set _"
  2134 instead.  INCOMPATIBILITY.
  2135 
  2136 * Removed lemma "Option.is_none_none" which duplicates "is_none_def".
  2137 INCOMPATIBILITY.
  2138 
  2139 * Former theory Library/Enum is now part of the HOL-Main image.
  2140 INCOMPATIBILITY: all constants of the Enum theory now have to be
  2141 referred to by its qualified name.
  2142 
  2143   enum    ~>  Enum.enum
  2144   nlists  ~>  Enum.nlists
  2145   product ~>  Enum.product
  2146 
  2147 * Theory Library/Monad_Syntax provides do-syntax for monad types.
  2148 Syntax in Library/State_Monad has been changed to avoid ambiguities.
  2149 INCOMPATIBILITY.
  2150 
  2151 * Theory Library/SetsAndFunctions has been split into
  2152 Library/Function_Algebras and Library/Set_Algebras; canonical names
  2153 for instance definitions for functions; various improvements.
  2154 INCOMPATIBILITY.
  2155 
  2156 * Theory Library/Multiset provides stable quicksort implementation of
  2157 sort_key.
  2158 
  2159 * Theory Library/Multiset: renamed empty_idemp ~> empty_neutral.
  2160 INCOMPATIBILITY.
  2161 
  2162 * Session Multivariate_Analysis: introduced a type class for euclidean
  2163 space.  Most theorems are now stated in terms of euclidean spaces
  2164 instead of finite cartesian products.
  2165 
  2166   types
  2167     real ^ 'n ~>  'a::real_vector
  2168               ~>  'a::euclidean_space
  2169               ~>  'a::ordered_euclidean_space
  2170         (depends on your needs)
  2171 
  2172   constants
  2173      _ $ _        ~> _ $$ _
  2174      \<chi> x. _  ~> \<chi>\<chi> x. _
  2175      CARD('n)     ~> DIM('a)
  2176 
  2177 Also note that the indices are now natural numbers and not from some
  2178 finite type. Finite cartesian products of euclidean spaces, products
  2179 of euclidean spaces the real and complex numbers are instantiated to
  2180 be euclidean_spaces.  INCOMPATIBILITY.
  2181 
  2182 * Session Probability: introduced pextreal as positive extended real
  2183 numbers.  Use pextreal as value for measures.  Introduce the
  2184 Radon-Nikodym derivative, product spaces and Fubini's theorem for
  2185 arbitrary sigma finite measures.  Introduces Lebesgue measure based on
  2186 the integral in Multivariate Analysis.  INCOMPATIBILITY.
  2187 
  2188 * Session Imperative_HOL: revamped, corrected dozens of inadequacies.
  2189 INCOMPATIBILITY.
  2190 
  2191 * Session SPARK (with image HOL-SPARK) provides commands to load and
  2192 prove verification conditions generated by the SPARK Ada program
  2193 verifier.  See also src/HOL/SPARK and src/HOL/SPARK/Examples.
  2194 
  2195 
  2196 *** HOL-Algebra ***
  2197 
  2198 * Theorems for additive ring operations (locale abelian_monoid and
  2199 descendants) are generated by interpretation from their multiplicative
  2200 counterparts.  Names (in particular theorem names) have the mandatory
  2201 qualifier 'add'.  Previous theorem names are redeclared for
  2202 compatibility.
  2203 
  2204 * Structure "int_ring" is now an abbreviation (previously a
  2205 definition).  This fits more natural with advanced interpretations.
  2206 
  2207 
  2208 *** HOLCF ***
  2209 
  2210 * The domain package now runs in definitional mode by default: The
  2211 former command 'new_domain' is now called 'domain'.  To use the domain
  2212 package in its original axiomatic mode, use 'domain (unsafe)'.
  2213 INCOMPATIBILITY.
  2214 
  2215 * The new class "domain" is now the default sort.  Class "predomain"
  2216 is an unpointed version of "domain". Theories can be updated by
  2217 replacing sort annotations as shown below.  INCOMPATIBILITY.
  2218 
  2219   'a::type ~> 'a::countable
  2220   'a::cpo  ~> 'a::predomain
  2221   'a::pcpo ~> 'a::domain
  2222 
  2223 * The old type class "rep" has been superseded by class "domain".
  2224 Accordingly, users of the definitional package must remove any
  2225 "default_sort rep" declarations.  INCOMPATIBILITY.
  2226 
  2227 * The domain package (definitional mode) now supports unpointed
  2228 predomain argument types, as long as they are marked 'lazy'. (Strict
  2229 arguments must be in class "domain".) For example, the following
  2230 domain definition now works:
  2231 
  2232   domain natlist = nil | cons (lazy "nat discr") (lazy "natlist")
  2233 
  2234 * Theory HOLCF/Library/HOL_Cpo provides cpo and predomain class
  2235 instances for types from main HOL: bool, nat, int, char, 'a + 'b,
  2236 'a option, and 'a list.  Additionally, it configures fixrec and the
  2237 domain package to work with these types.  For example:
  2238 
  2239   fixrec isInl :: "('a + 'b) u -> tr"
  2240     where "isInl$(up$(Inl x)) = TT" | "isInl$(up$(Inr y)) = FF"
  2241 
  2242   domain V = VFun (lazy "V -> V") | VCon (lazy "nat") (lazy "V list")
  2243 
  2244 * The "(permissive)" option of fixrec has been replaced with a
  2245 per-equation "(unchecked)" option. See
  2246 src/HOL/HOLCF/Tutorial/Fixrec_ex.thy for examples. INCOMPATIBILITY.
  2247 
  2248 * The "bifinite" class no longer fixes a constant "approx"; the class
  2249 now just asserts that such a function exists.  INCOMPATIBILITY.
  2250 
  2251 * Former type "alg_defl" has been renamed to "defl".  HOLCF no longer
  2252 defines an embedding of type 'a defl into udom by default; instances
  2253 of "bifinite" and "domain" classes are available in
  2254 src/HOL/HOLCF/Library/Defl_Bifinite.thy.
  2255 
  2256 * The syntax "REP('a)" has been replaced with "DEFL('a)".
  2257 
  2258 * The predicate "directed" has been removed.  INCOMPATIBILITY.
  2259 
  2260 * The type class "finite_po" has been removed.  INCOMPATIBILITY.
  2261 
  2262 * The function "cprod_map" has been renamed to "prod_map".
  2263 INCOMPATIBILITY.
  2264 
  2265 * The monadic bind operator on each powerdomain has new binder syntax
  2266 similar to sets, e.g. "\<Union>\<sharp>x\<in>xs. t" represents
  2267 "upper_bind\<cdot>xs\<cdot>(\<Lambda> x. t)".
  2268 
  2269 * The infix syntax for binary union on each powerdomain has changed
  2270 from e.g. "+\<sharp>" to "\<union>\<sharp>", for consistency with set
  2271 syntax.  INCOMPATIBILITY.
  2272 
  2273 * The constant "UU" has been renamed to "bottom".  The syntax "UU" is
  2274 still supported as an input translation.
  2275 
  2276 * Renamed some theorems (the original names are also still available).
  2277 
  2278   expand_fun_below   ~> fun_below_iff
  2279   below_fun_ext      ~> fun_belowI
  2280   expand_cfun_eq     ~> cfun_eq_iff
  2281   ext_cfun           ~> cfun_eqI
  2282   expand_cfun_below  ~> cfun_below_iff
  2283   below_cfun_ext     ~> cfun_belowI
  2284   cont2cont_Rep_CFun ~> cont2cont_APP
  2285 
  2286 * The Abs and Rep functions for various types have changed names.
  2287 Related theorem names have also changed to match. INCOMPATIBILITY.
  2288 
  2289   Rep_CFun  ~> Rep_cfun
  2290   Abs_CFun  ~> Abs_cfun
  2291   Rep_Sprod ~> Rep_sprod
  2292   Abs_Sprod ~> Abs_sprod
  2293   Rep_Ssum  ~> Rep_ssum
  2294   Abs_Ssum  ~> Abs_ssum
  2295 
  2296 * Lemmas with names of the form *_defined_iff or *_strict_iff have
  2297 been renamed to *_bottom_iff.  INCOMPATIBILITY.
  2298 
  2299 * Various changes to bisimulation/coinduction with domain package:
  2300 
  2301   - Definitions of "bisim" constants no longer mention definedness.
  2302   - With mutual recursion, "bisim" predicate is now curried.
  2303   - With mutual recursion, each type gets a separate coind theorem.
  2304   - Variable names in bisim_def and coinduct rules have changed.
  2305 
  2306 INCOMPATIBILITY.
  2307 
  2308 * Case combinators generated by the domain package for type "foo" are
  2309 now named "foo_case" instead of "foo_when".  INCOMPATIBILITY.
  2310 
  2311 * Several theorems have been renamed to more accurately reflect the
  2312 names of constants and types involved.  INCOMPATIBILITY.
  2313 
  2314   thelub_const    ~> lub_const
  2315   lub_const       ~> is_lub_const
  2316   thelubI         ~> lub_eqI
  2317   is_lub_lub      ~> is_lubD2
  2318   lubI            ~> is_lub_lub
  2319   unique_lub      ~> is_lub_unique
  2320   is_ub_lub       ~> is_lub_rangeD1
  2321   lub_bin_chain   ~> is_lub_bin_chain
  2322   lub_fun         ~> is_lub_fun
  2323   thelub_fun      ~> lub_fun
  2324   thelub_cfun     ~> lub_cfun
  2325   thelub_Pair     ~> lub_Pair
  2326   lub_cprod       ~> is_lub_prod
  2327   thelub_cprod    ~> lub_prod
  2328   minimal_cprod   ~> minimal_prod
  2329   inst_cprod_pcpo ~> inst_prod_pcpo
  2330   UU_I            ~> bottomI
  2331   compact_UU      ~> compact_bottom
  2332   deflation_UU    ~> deflation_bottom
  2333   finite_deflation_UU ~> finite_deflation_bottom
  2334 
  2335 * Many legacy theorem names have been discontinued.  INCOMPATIBILITY.
  2336 
  2337   sq_ord_less_eq_trans ~> below_eq_trans
  2338   sq_ord_eq_less_trans ~> eq_below_trans
  2339   refl_less            ~> below_refl
  2340   trans_less           ~> below_trans
  2341   antisym_less         ~> below_antisym
  2342   antisym_less_inverse ~> po_eq_conv [THEN iffD1]
  2343   box_less             ~> box_below
  2344   rev_trans_less       ~> rev_below_trans
  2345   not_less2not_eq      ~> not_below2not_eq
  2346   less_UU_iff          ~> below_UU_iff
  2347   flat_less_iff        ~> flat_below_iff
  2348   adm_less             ~> adm_below
  2349   adm_not_less         ~> adm_not_below
  2350   adm_compact_not_less ~> adm_compact_not_below
  2351   less_fun_def         ~> below_fun_def
  2352   expand_fun_less      ~> fun_below_iff
  2353   less_fun_ext         ~> fun_belowI
  2354   less_discr_def       ~> below_discr_def
  2355   discr_less_eq        ~> discr_below_eq
  2356   less_unit_def        ~> below_unit_def
  2357   less_cprod_def       ~> below_prod_def
  2358   prod_lessI           ~> prod_belowI
  2359   Pair_less_iff        ~> Pair_below_iff
  2360   fst_less_iff         ~> fst_below_iff
  2361   snd_less_iff         ~> snd_below_iff
  2362   expand_cfun_less     ~> cfun_below_iff
  2363   less_cfun_ext        ~> cfun_belowI
  2364   injection_less       ~> injection_below
  2365   less_up_def          ~> below_up_def
  2366   not_Iup_less         ~> not_Iup_below
  2367   Iup_less             ~> Iup_below
  2368   up_less              ~> up_below
  2369   Def_inject_less_eq   ~> Def_below_Def
  2370   Def_less_is_eq       ~> Def_below_iff
  2371   spair_less_iff       ~> spair_below_iff
  2372   less_sprod           ~> below_sprod
  2373   spair_less           ~> spair_below
  2374   sfst_less_iff        ~> sfst_below_iff
  2375   ssnd_less_iff        ~> ssnd_below_iff
  2376   fix_least_less       ~> fix_least_below
  2377   dist_less_one        ~> dist_below_one
  2378   less_ONE             ~> below_ONE
  2379   ONE_less_iff         ~> ONE_below_iff
  2380   less_sinlD           ~> below_sinlD
  2381   less_sinrD           ~> below_sinrD
  2382 
  2383 
  2384 *** FOL and ZF ***
  2385 
  2386 * All constant names are now qualified internally and use proper
  2387 identifiers, e.g. "IFOL.eq" instead of "op =".  INCOMPATIBILITY.
  2388 
  2389 
  2390 *** ML ***
  2391 
  2392 * Antiquotation @{assert} inlines a function bool -> unit that raises
  2393 Fail if the argument is false.  Due to inlining the source position of
  2394 failed assertions is included in the error output.
  2395 
  2396 * Discontinued antiquotation @{theory_ref}, which is obsolete since ML
  2397 text is in practice always evaluated with a stable theory checkpoint.
  2398 Minor INCOMPATIBILITY, use (Theory.check_thy @{theory}) instead.
  2399 
  2400 * Antiquotation @{theory A} refers to theory A from the ancestry of
  2401 the current context, not any accidental theory loader state as before.
  2402 Potential INCOMPATIBILITY, subtle change in semantics.
  2403 
  2404 * Syntax.pretty_priority (default 0) configures the required priority
  2405 of pretty-printed output and thus affects insertion of parentheses.
  2406 
  2407 * Syntax.default_root (default "any") configures the inner syntax
  2408 category (nonterminal symbol) for parsing of terms.
  2409 
  2410 * Former exception Library.UnequalLengths now coincides with
  2411 ListPair.UnequalLengths.
  2412 
  2413 * Renamed structure MetaSimplifier to Raw_Simplifier.  Note that the
  2414 main functionality is provided by structure Simplifier.
  2415 
  2416 * Renamed raw "explode" function to "raw_explode" to emphasize its
  2417 meaning.  Note that internally to Isabelle, Symbol.explode is used in
  2418 almost all situations.
  2419 
  2420 * Discontinued obsolete function sys_error and exception SYS_ERROR.
  2421 See implementation manual for further details on exceptions in
  2422 Isabelle/ML.
  2423 
  2424 * Renamed setmp_noncritical to Unsynchronized.setmp to emphasize its
  2425 meaning.
  2426 
  2427 * Renamed structure PureThy to Pure_Thy and moved most of its
  2428 operations to structure Global_Theory, to emphasize that this is
  2429 rarely-used global-only stuff.
  2430 
  2431 * Discontinued Output.debug.  Minor INCOMPATIBILITY, use plain writeln
  2432 instead (or tracing for high-volume output).
  2433 
  2434 * Configuration option show_question_marks only affects regular pretty
  2435 printing of types and terms, not raw Term.string_of_vname.
  2436 
  2437 * ML_Context.thm and ML_Context.thms are no longer pervasive.  Rare
  2438 INCOMPATIBILITY, superseded by static antiquotations @{thm} and
  2439 @{thms} for most purposes.
  2440 
  2441 * ML structure Unsynchronized is never opened, not even in Isar
  2442 interaction mode as before.  Old Unsynchronized.set etc. have been
  2443 discontinued -- use plain := instead.  This should be *rare* anyway,
  2444 since modern tools always work via official context data, notably
  2445 configuration options.
  2446 
  2447 * Parallel and asynchronous execution requires special care concerning
  2448 interrupts.  Structure Exn provides some convenience functions that
  2449 avoid working directly with raw Interrupt.  User code must not absorb
  2450 interrupts -- intermediate handling (for cleanup etc.) needs to be
  2451 followed by re-raising of the original exception.  Another common
  2452 source of mistakes are "handle _" patterns, which make the meaning of
  2453 the program subject to physical effects of the environment.
  2454 
  2455 
  2456 
  2457 New in Isabelle2009-2 (June 2010)
  2458 ---------------------------------
  2459 
  2460 *** General ***
  2461 
  2462 * Authentic syntax for *all* logical entities (type classes, type
  2463 constructors, term constants): provides simple and robust
  2464 correspondence between formal entities and concrete syntax.  Within
  2465 the parse tree / AST representations, "constants" are decorated by
  2466 their category (class, type, const) and spelled out explicitly with
  2467 their full internal name.
  2468 
  2469 Substantial INCOMPATIBILITY concerning low-level syntax declarations
  2470 and translations (translation rules and translation functions in ML).
  2471 Some hints on upgrading:
  2472 
  2473   - Many existing uses of 'syntax' and 'translations' can be replaced
  2474     by more modern 'type_notation', 'notation' and 'abbreviation',
  2475     which are independent of this issue.
  2476 
  2477   - 'translations' require markup within the AST; the term syntax
  2478     provides the following special forms:
  2479 
  2480       CONST c   -- produces syntax version of constant c from context
  2481       XCONST c  -- literally c, checked as constant from context
  2482       c         -- literally c, if declared by 'syntax'
  2483 
  2484     Plain identifiers are treated as AST variables -- occasionally the
  2485     system indicates accidental variables via the error "rhs contains
  2486     extra variables".
  2487 
  2488     Type classes and type constructors are marked according to their
  2489     concrete syntax.  Some old translations rules need to be written
  2490     for the "type" category, using type constructor application
  2491     instead of pseudo-term application of the default category
  2492     "logic".
  2493 
  2494   - 'parse_translation' etc. in ML may use the following
  2495     antiquotations:
  2496 
  2497       @{class_syntax c}   -- type class c within parse tree / AST
  2498       @{term_syntax c}    -- type constructor c within parse tree / AST
  2499       @{const_syntax c}   -- ML version of "CONST c" above
  2500       @{syntax_const c}   -- literally c (checked wrt. 'syntax' declarations)
  2501 
  2502   - Literal types within 'typed_print_translations', i.e. those *not*
  2503     represented as pseudo-terms are represented verbatim.  Use @{class
  2504     c} or @{type_name c} here instead of the above syntax
  2505     antiquotations.
  2506 
  2507 Note that old non-authentic syntax was based on unqualified base
  2508 names, so all of the above "constant" names would coincide.  Recall
  2509 that 'print_syntax' and ML_command "set Syntax.trace_ast" help to
  2510 diagnose syntax problems.
  2511 
  2512 * Type constructors admit general mixfix syntax, not just infix.
  2513 
  2514 * Concrete syntax may be attached to local entities without a proof
  2515 body, too.  This works via regular mixfix annotations for 'fix',
  2516 'def', 'obtain' etc. or via the explicit 'write' command, which is
  2517 similar to the 'notation' command in theory specifications.
  2518 
  2519 * Discontinued unnamed infix syntax (legacy feature for many years) --
  2520 need to specify constant name and syntax separately.  Internal ML
  2521 datatype constructors have been renamed from InfixName to Infix etc.
  2522 Minor INCOMPATIBILITY.
  2523 
  2524 * Schematic theorem statements need to be explicitly markup as such,
  2525 via commands 'schematic_lemma', 'schematic_theorem',
  2526 'schematic_corollary'.  Thus the relevance of the proof is made
  2527 syntactically clear, which impacts performance in a parallel or
  2528 asynchronous interactive environment.  Minor INCOMPATIBILITY.
  2529 
  2530 * Use of cumulative prems via "!" in some proof methods has been
  2531 discontinued (old legacy feature).
  2532 
  2533 * References 'trace_simp' and 'debug_simp' have been replaced by
  2534 configuration options stored in the context. Enabling tracing (the
  2535 case of debugging is similar) in proofs works via
  2536 
  2537   using [[trace_simp = true]]
  2538 
  2539 Tracing is then active for all invocations of the simplifier in
  2540 subsequent goal refinement steps. Tracing may also still be enabled or
  2541 disabled via the ProofGeneral settings menu.
  2542 
  2543 * Separate commands 'hide_class', 'hide_type', 'hide_const',
  2544 'hide_fact' replace the former 'hide' KIND command.  Minor
  2545 INCOMPATIBILITY.
  2546 
  2547 * Improved parallelism of proof term normalization: usedir -p2 -q0 is
  2548 more efficient than combinations with -q1 or -q2.
  2549 
  2550 
  2551 *** Pure ***
  2552 
  2553 * Proofterms record type-class reasoning explicitly, using the
  2554 "unconstrain" operation internally.  This eliminates all sort
  2555 constraints from a theorem and proof, introducing explicit
  2556 OFCLASS-premises.  On the proof term level, this operation is
  2557 automatically applied at theorem boundaries, such that closed proofs
  2558 are always free of sort constraints.  INCOMPATIBILITY for tools that
  2559 inspect proof terms.
  2560 
  2561 * Local theory specifications may depend on extra type variables that
  2562 are not present in the result type -- arguments TYPE('a) :: 'a itself
  2563 are added internally.  For example:
  2564 
  2565   definition unitary :: bool where "unitary = (ALL (x::'a) y. x = y)"
  2566 
  2567 * Predicates of locales introduced by classes carry a mandatory
  2568 "class" prefix.  INCOMPATIBILITY.
  2569 
  2570 * Vacuous class specifications observe default sort.  INCOMPATIBILITY.
  2571 
  2572 * Old 'axclass' command has been discontinued.  INCOMPATIBILITY, use
  2573 'class' instead.
  2574 
  2575 * Command 'code_reflect' allows to incorporate generated ML code into
  2576 runtime environment; replaces immature code_datatype antiquotation.
  2577 INCOMPATIBILITY.
  2578 
  2579 * Code generator: simple concept for abstract datatypes obeying
  2580 invariants.
  2581 
  2582 * Code generator: details of internal data cache have no impact on the
  2583 user space functionality any longer.
  2584 
  2585 * Methods "unfold_locales" and "intro_locales" ignore non-locale
  2586 subgoals.  This is more appropriate for interpretations with 'where'.
  2587 INCOMPATIBILITY.
  2588 
  2589 * Command 'example_proof' opens an empty proof body.  This allows to
  2590 experiment with Isar, without producing any persistent result.
  2591 
  2592 * Commands 'type_notation' and 'no_type_notation' declare type syntax
  2593 within a local theory context, with explicit checking of the
  2594 constructors involved (in contrast to the raw 'syntax' versions).
  2595 
  2596 * Commands 'types' and 'typedecl' now work within a local theory
  2597 context -- without introducing dependencies on parameters or
  2598 assumptions, which is not possible in Isabelle/Pure.
  2599 
  2600 * Command 'defaultsort' has been renamed to 'default_sort', it works
  2601 within a local theory context.  Minor INCOMPATIBILITY.
  2602 
  2603 
  2604 *** HOL ***
  2605 
  2606 * Command 'typedef' now works within a local theory context -- without
  2607 introducing dependencies on parameters or assumptions, which is not
  2608 possible in Isabelle/Pure/HOL.  Note that the logical environment may
  2609 contain multiple interpretations of local typedefs (with different
  2610 non-emptiness proofs), even in a global theory context.
  2611 
  2612 * New package for quotient types.  Commands 'quotient_type' and
  2613 'quotient_definition' may be used for defining types and constants by
  2614 quotient constructions.  An example is the type of integers created by
  2615 quotienting pairs of natural numbers:
  2616 
  2617   fun
  2618     intrel :: "(nat * nat) => (nat * nat) => bool"
  2619   where
  2620     "intrel (x, y) (u, v) = (x + v = u + y)"
  2621 
  2622   quotient_type int = "nat * nat" / intrel
  2623     by (auto simp add: equivp_def expand_fun_eq)
  2624 
  2625   quotient_definition
  2626     "0::int" is "(0::nat, 0::nat)"
  2627 
  2628 The method "lifting" can be used to lift of theorems from the
  2629 underlying "raw" type to the quotient type.  The example
  2630 src/HOL/Quotient_Examples/FSet.thy includes such a quotient
  2631 construction and provides a reasoning infrastructure for finite sets.
  2632 
  2633 * Renamed Library/Quotient.thy to Library/Quotient_Type.thy to avoid
  2634 clash with new theory Quotient in Main HOL.
  2635 
  2636 * Moved the SMT binding into the main HOL session, eliminating
  2637 separate HOL-SMT session.
  2638 
  2639 * List membership infix mem operation is only an input abbreviation.
  2640 INCOMPATIBILITY.
  2641 
  2642 * Theory Library/Word.thy has been removed.  Use library Word/Word.thy
  2643 for future developements; former Library/Word.thy is still present in
  2644 the AFP entry RSAPPS.
  2645 
  2646 * Theorem Int.int_induct renamed to Int.int_of_nat_induct and is no
  2647 longer shadowed.  INCOMPATIBILITY.
  2648 
  2649 * Dropped theorem duplicate comp_arith; use semiring_norm instead.
  2650 INCOMPATIBILITY.
  2651 
  2652 * Dropped theorem RealPow.real_sq_order; use power2_le_imp_le instead.
  2653 INCOMPATIBILITY.
  2654 
  2655 * Dropped normalizing_semiring etc; use the facts in semiring classes
  2656 instead.  INCOMPATIBILITY.
  2657 
  2658 * Dropped several real-specific versions of lemmas about floor and
  2659 ceiling; use the generic lemmas from theory "Archimedean_Field"
  2660 instead.  INCOMPATIBILITY.
  2661 
  2662   floor_number_of_eq         ~> floor_number_of
  2663   le_floor_eq_number_of      ~> number_of_le_floor
  2664   le_floor_eq_zero           ~> zero_le_floor
  2665   le_floor_eq_one            ~> one_le_floor
  2666   floor_less_eq_number_of    ~> floor_less_number_of
  2667   floor_less_eq_zero         ~> floor_less_zero
  2668   floor_less_eq_one          ~> floor_less_one
  2669   less_floor_eq_number_of    ~> number_of_less_floor
  2670   less_floor_eq_zero         ~> zero_less_floor
  2671   less_floor_eq_one          ~> one_less_floor
  2672   floor_le_eq_number_of      ~> floor_le_number_of
  2673   floor_le_eq_zero           ~> floor_le_zero
  2674   floor_le_eq_one            ~> floor_le_one
  2675   floor_subtract_number_of   ~> floor_diff_number_of
  2676   floor_subtract_one         ~> floor_diff_one
  2677   ceiling_number_of_eq       ~> ceiling_number_of
  2678   ceiling_le_eq_number_of    ~> ceiling_le_number_of
  2679   ceiling_le_zero_eq         ~> ceiling_le_zero
  2680   ceiling_le_eq_one          ~> ceiling_le_one
  2681   less_ceiling_eq_number_of  ~> number_of_less_ceiling
  2682   less_ceiling_eq_zero       ~> zero_less_ceiling
  2683   less_ceiling_eq_one        ~> one_less_ceiling
  2684   ceiling_less_eq_number_of  ~> ceiling_less_number_of
  2685   ceiling_less_eq_zero       ~> ceiling_less_zero
  2686   ceiling_less_eq_one        ~> ceiling_less_one
  2687   le_ceiling_eq_number_of    ~> number_of_le_ceiling
  2688   le_ceiling_eq_zero         ~> zero_le_ceiling
  2689   le_ceiling_eq_one          ~> one_le_ceiling
  2690   ceiling_subtract_number_of ~> ceiling_diff_number_of
  2691   ceiling_subtract_one       ~> ceiling_diff_one
  2692 
  2693 * Theory "Finite_Set": various folding_XXX locales facilitate the
  2694 application of the various fold combinators on finite sets.
  2695 
  2696 * Library theory "RBT" renamed to "RBT_Impl"; new library theory "RBT"
  2697 provides abstract red-black tree type which is backed by "RBT_Impl" as
  2698 implementation.  INCOMPATIBILTY.
  2699 
  2700 * Theory Library/Coinductive_List has been removed -- superseded by
  2701 AFP/thys/Coinductive.
  2702 
  2703 * Theory PReal, including the type "preal" and related operations, has
  2704 been removed.  INCOMPATIBILITY.
  2705 
  2706 * Real: new development using Cauchy Sequences.
  2707 
  2708 * Split off theory "Big_Operators" containing setsum, setprod,
  2709 Inf_fin, Sup_fin, Min, Max from theory Finite_Set.  INCOMPATIBILITY.
  2710 
  2711 * Theory "Rational" renamed to "Rat", for consistency with "Nat",
  2712 "Int" etc.  INCOMPATIBILITY.
  2713 
  2714 * Constant Rat.normalize needs to be qualified.  INCOMPATIBILITY.
  2715 
  2716 * New set of rules "ac_simps" provides combined assoc / commute
  2717 rewrites for all interpretations of the appropriate generic locales.
  2718 
  2719 * Renamed theory "OrderedGroup" to "Groups" and split theory
  2720 "Ring_and_Field" into theories "Rings" and "Fields"; for more
  2721 appropriate and more consistent names suitable for name prefixes
  2722 within the HOL theories.  INCOMPATIBILITY.
  2723 
  2724 * Some generic constants have been put to appropriate theories:
  2725   - less_eq, less: Orderings
  2726   - zero, one, plus, minus, uminus, times, abs, sgn: Groups
  2727   - inverse, divide: Rings
  2728 INCOMPATIBILITY.
  2729 
  2730 * More consistent naming of type classes involving orderings (and
  2731 lattices):
  2732 
  2733     lower_semilattice                   ~> semilattice_inf
  2734     upper_semilattice                   ~> semilattice_sup
  2735 
  2736     dense_linear_order                  ~> dense_linorder
  2737 
  2738     pordered_ab_group_add               ~> ordered_ab_group_add
  2739     pordered_ab_group_add_abs           ~> ordered_ab_group_add_abs
  2740     pordered_ab_semigroup_add           ~> ordered_ab_semigroup_add
  2741     pordered_ab_semigroup_add_imp_le    ~> ordered_ab_semigroup_add_imp_le
  2742     pordered_cancel_ab_semigroup_add    ~> ordered_cancel_ab_semigroup_add
  2743     pordered_cancel_comm_semiring       ~> ordered_cancel_comm_semiring
  2744     pordered_cancel_semiring            ~> ordered_cancel_semiring
  2745     pordered_comm_monoid_add            ~> ordered_comm_monoid_add
  2746     pordered_comm_ring                  ~> ordered_comm_ring
  2747     pordered_comm_semiring              ~> ordered_comm_semiring
  2748     pordered_ring                       ~> ordered_ring
  2749     pordered_ring_abs                   ~> ordered_ring_abs
  2750     pordered_semiring                   ~> ordered_semiring
  2751 
  2752     ordered_ab_group_add                ~> linordered_ab_group_add
  2753     ordered_ab_semigroup_add            ~> linordered_ab_semigroup_add
  2754     ordered_cancel_ab_semigroup_add     ~> linordered_cancel_ab_semigroup_add
  2755     ordered_comm_semiring_strict        ~> linordered_comm_semiring_strict
  2756     ordered_field                       ~> linordered_field
  2757     ordered_field_no_lb                 ~> linordered_field_no_lb
  2758     ordered_field_no_ub                 ~> linordered_field_no_ub
  2759     ordered_field_dense_linear_order    ~> dense_linordered_field
  2760     ordered_idom                        ~> linordered_idom
  2761     ordered_ring                        ~> linordered_ring
  2762     ordered_ring_le_cancel_factor       ~> linordered_ring_le_cancel_factor
  2763     ordered_ring_less_cancel_factor     ~> linordered_ring_less_cancel_factor
  2764     ordered_ring_strict                 ~> linordered_ring_strict
  2765     ordered_semidom                     ~> linordered_semidom
  2766     ordered_semiring                    ~> linordered_semiring
  2767     ordered_semiring_1                  ~> linordered_semiring_1
  2768     ordered_semiring_1_strict           ~> linordered_semiring_1_strict
  2769     ordered_semiring_strict             ~> linordered_semiring_strict
  2770 
  2771   The following slightly odd type classes have been moved to a
  2772   separate theory Library/Lattice_Algebras:
  2773 
  2774     lordered_ab_group_add               ~> lattice_ab_group_add
  2775     lordered_ab_group_add_abs           ~> lattice_ab_group_add_abs
  2776     lordered_ab_group_add_meet          ~> semilattice_inf_ab_group_add
  2777     lordered_ab_group_add_join          ~> semilattice_sup_ab_group_add
  2778     lordered_ring                       ~> lattice_ring
  2779 
  2780 INCOMPATIBILITY.
  2781 
  2782 * Refined field classes:
  2783   - classes division_ring_inverse_zero, field_inverse_zero,
  2784     linordered_field_inverse_zero include rule inverse 0 = 0 --
  2785     subsumes former division_by_zero class;
  2786   - numerous lemmas have been ported from field to division_ring.
  2787 INCOMPATIBILITY.
  2788 
  2789 * Refined algebra theorem collections:
  2790   - dropped theorem group group_simps, use algebra_simps instead;
  2791   - dropped theorem group ring_simps, use field_simps instead;
  2792   - proper theorem collection field_simps subsumes former theorem
  2793     groups field_eq_simps and field_simps;
  2794   - dropped lemma eq_minus_self_iff which is a duplicate for
  2795     equal_neg_zero.
  2796 INCOMPATIBILITY.
  2797 
  2798 * Theory Finite_Set and List: some lemmas have been generalized from
  2799 sets to lattices:
  2800 
  2801   fun_left_comm_idem_inter      ~> fun_left_comm_idem_inf
  2802   fun_left_comm_idem_union      ~> fun_left_comm_idem_sup
  2803   inter_Inter_fold_inter        ~> inf_Inf_fold_inf
  2804   union_Union_fold_union        ~> sup_Sup_fold_sup
  2805   Inter_fold_inter              ~> Inf_fold_inf
  2806   Union_fold_union              ~> Sup_fold_sup
  2807   inter_INTER_fold_inter        ~> inf_INFI_fold_inf
  2808   union_UNION_fold_union        ~> sup_SUPR_fold_sup
  2809   INTER_fold_inter              ~> INFI_fold_inf
  2810   UNION_fold_union              ~> SUPR_fold_sup
  2811 
  2812 * Theory "Complete_Lattice": lemmas top_def and bot_def have been
  2813 replaced by the more convenient lemmas Inf_empty and Sup_empty.
  2814 Dropped lemmas Inf_insert_simp and Sup_insert_simp, which are subsumed
  2815 by Inf_insert and Sup_insert.  Lemmas Inf_UNIV and Sup_UNIV replace
  2816 former Inf_Univ and Sup_Univ.  Lemmas inf_top_right and sup_bot_right
  2817 subsume inf_top and sup_bot respectively.  INCOMPATIBILITY.
  2818 
  2819 * Reorganized theory Multiset: swapped notation of pointwise and
  2820 multiset order:
  2821 
  2822   - pointwise ordering is instance of class order with standard syntax
  2823     <= and <;
  2824   - multiset ordering has syntax <=# and <#; partial order properties
  2825     are provided by means of interpretation with prefix
  2826     multiset_order;
  2827   - less duplication, less historical organization of sections,
  2828     conversion from associations lists to multisets, rudimentary code
  2829     generation;
  2830   - use insert_DiffM2 [symmetric] instead of elem_imp_eq_diff_union,
  2831     if needed.
  2832 
  2833 Renamed:
  2834 
  2835   multiset_eq_conv_count_eq  ~>  multiset_ext_iff
  2836   multi_count_ext  ~>  multiset_ext
  2837   diff_union_inverse2  ~>  diff_union_cancelR
  2838 
  2839 INCOMPATIBILITY.
  2840 
  2841 * Theory Permutation: replaced local "remove" by List.remove1.
  2842 
  2843 * Code generation: ML and OCaml code is decorated with signatures.
  2844 
  2845 * Theory List: added transpose.
  2846 
  2847 * Library/Nat_Bijection.thy is a collection of bijective functions
  2848 between nat and other types, which supersedes the older libraries
  2849 Library/Nat_Int_Bij.thy and HOLCF/NatIso.thy.  INCOMPATIBILITY.
  2850 
  2851   Constants:
  2852   Nat_Int_Bij.nat2_to_nat         ~> prod_encode
  2853   Nat_Int_Bij.nat_to_nat2         ~> prod_decode
  2854   Nat_Int_Bij.int_to_nat_bij      ~> int_encode
  2855   Nat_Int_Bij.nat_to_int_bij      ~> int_decode
  2856   Countable.pair_encode           ~> prod_encode
  2857   NatIso.prod2nat                 ~> prod_encode
  2858   NatIso.nat2prod                 ~> prod_decode
  2859   NatIso.sum2nat                  ~> sum_encode
  2860   NatIso.nat2sum                  ~> sum_decode
  2861   NatIso.list2nat                 ~> list_encode
  2862   NatIso.nat2list                 ~> list_decode
  2863   NatIso.set2nat                  ~> set_encode
  2864   NatIso.nat2set                  ~> set_decode
  2865 
  2866   Lemmas:
  2867   Nat_Int_Bij.bij_nat_to_int_bij  ~> bij_int_decode
  2868   Nat_Int_Bij.nat2_to_nat_inj     ~> inj_prod_encode
  2869   Nat_Int_Bij.nat2_to_nat_surj    ~> surj_prod_encode
  2870   Nat_Int_Bij.nat_to_nat2_inj     ~> inj_prod_decode
  2871   Nat_Int_Bij.nat_to_nat2_surj    ~> surj_prod_decode
  2872   Nat_Int_Bij.i2n_n2i_id          ~> int_encode_inverse
  2873   Nat_Int_Bij.n2i_i2n_id          ~> int_decode_inverse
  2874   Nat_Int_Bij.surj_nat_to_int_bij ~> surj_int_encode
  2875   Nat_Int_Bij.surj_int_to_nat_bij ~> surj_int_decode
  2876   Nat_Int_Bij.inj_nat_to_int_bij  ~> inj_int_encode
  2877   Nat_Int_Bij.inj_int_to_nat_bij  ~> inj_int_decode
  2878   Nat_Int_Bij.bij_nat_to_int_bij  ~> bij_int_encode
  2879   Nat_Int_Bij.bij_int_to_nat_bij  ~> bij_int_decode
  2880 
  2881 * Sledgehammer:
  2882   - Renamed ATP commands:
  2883     atp_info     ~> sledgehammer running_atps
  2884     atp_kill     ~> sledgehammer kill_atps
  2885     atp_messages ~> sledgehammer messages
  2886     atp_minimize ~> sledgehammer minimize
  2887     print_atps   ~> sledgehammer available_atps
  2888     INCOMPATIBILITY.
  2889   - Added user's manual ("isabelle doc sledgehammer").
  2890   - Added option syntax and "sledgehammer_params" to customize
  2891     Sledgehammer's behavior.  See the manual for details.
  2892   - Modified the Isar proof reconstruction code so that it produces
  2893     direct proofs rather than proofs by contradiction.  (This feature
  2894     is still experimental.)
  2895   - Made Isar proof reconstruction work for SPASS, remote ATPs, and in
  2896     full-typed mode.
  2897   - Added support for TPTP syntax for SPASS via the "spass_tptp" ATP.
  2898 
  2899 * Nitpick:
  2900   - Added and implemented "binary_ints" and "bits" options.
  2901   - Added "std" option and implemented support for nonstandard models.
  2902   - Added and implemented "finitize" option to improve the precision
  2903     of infinite datatypes based on a monotonicity analysis.
  2904   - Added support for quotient types.
  2905   - Added support for "specification" and "ax_specification"
  2906     constructs.
  2907   - Added support for local definitions (for "function" and
  2908     "termination" proofs).
  2909   - Added support for term postprocessors.
  2910   - Optimized "Multiset.multiset" and "FinFun.finfun".
  2911   - Improved efficiency of "destroy_constrs" optimization.
  2912   - Fixed soundness bugs related to "destroy_constrs" optimization and
  2913     record getters.
  2914   - Fixed soundness bug related to higher-order constructors.
  2915   - Fixed soundness bug when "full_descrs" is enabled.
  2916   - Improved precision of set constructs.
  2917   - Added "atoms" option.
  2918   - Added cache to speed up repeated Kodkod invocations on the same
  2919     problems.
  2920   - Renamed "MiniSatJNI", "zChaffJNI", "BerkMinAlloy", and
  2921     "SAT4JLight" to "MiniSat_JNI", "zChaff_JNI", "BerkMin_Alloy", and
  2922     "SAT4J_Light".  INCOMPATIBILITY.
  2923   - Removed "skolemize", "uncurry", "sym_break", "flatten_prop",
  2924     "sharing_depth", and "show_skolems" options.  INCOMPATIBILITY.
  2925   - Removed "nitpick_intro" attribute.  INCOMPATIBILITY.
  2926 
  2927 * Method "induct" now takes instantiations of the form t, where t is not
  2928   a variable, as a shorthand for "x == t", where x is a fresh variable.
  2929   If this is not intended, t has to be enclosed in parentheses.
  2930   By default, the equalities generated by definitional instantiations
  2931   are pre-simplified, which may cause parameters of inductive cases
  2932   to disappear, or may even delete some of the inductive cases.
  2933   Use "induct (no_simp)" instead of "induct" to restore the old
  2934   behaviour. The (no_simp) option is also understood by the "cases"
  2935   and "nominal_induct" methods, which now perform pre-simplification, too.
  2936   INCOMPATIBILITY.
  2937 
  2938 
  2939 *** HOLCF ***
  2940 
  2941 * Variable names in lemmas generated by the domain package have
  2942 changed; the naming scheme is now consistent with the HOL datatype
  2943 package.  Some proof scripts may be affected, INCOMPATIBILITY.
  2944 
  2945 * The domain package no longer defines the function "foo_copy" for
  2946 recursive domain "foo".  The reach lemma is now stated directly in
  2947 terms of "foo_take".  Lemmas and proofs that mention "foo_copy" must
  2948 be reformulated in terms of "foo_take", INCOMPATIBILITY.
  2949 
  2950 * Most definedness lemmas generated by the domain package (previously
  2951 of the form "x ~= UU ==> foo$x ~= UU") now have an if-and-only-if form
  2952 like "foo$x = UU <-> x = UU", which works better as a simp rule.
  2953 Proofs that used definedness lemmas as intro rules may break,
  2954 potential INCOMPATIBILITY.
  2955 
  2956 * Induction and casedist rules generated by the domain package now
  2957 declare proper case_names (one called "bottom", and one named for each
  2958 constructor).  INCOMPATIBILITY.
  2959 
  2960 * For mutually-recursive domains, separate "reach" and "take_lemma"
  2961 rules are generated for each domain, INCOMPATIBILITY.
  2962 
  2963   foo_bar.reach       ~> foo.reach  bar.reach
  2964   foo_bar.take_lemmas ~> foo.take_lemma  bar.take_lemma
  2965 
  2966 * Some lemmas generated by the domain package have been renamed for
  2967 consistency with the datatype package, INCOMPATIBILITY.
  2968 
  2969   foo.ind        ~> foo.induct
  2970   foo.finite_ind ~> foo.finite_induct
  2971   foo.coind      ~> foo.coinduct
  2972   foo.casedist   ~> foo.exhaust
  2973   foo.exhaust    ~> foo.nchotomy
  2974 
  2975 * For consistency with other definition packages, the fixrec package
  2976 now generates qualified theorem names, INCOMPATIBILITY.
  2977 
  2978   foo_simps  ~> foo.simps
  2979   foo_unfold ~> foo.unfold
  2980   foo_induct ~> foo.induct
  2981 
  2982 * The "fixrec_simp" attribute has been removed.  The "fixrec_simp"
  2983 method and internal fixrec proofs now use the default simpset instead.
  2984 INCOMPATIBILITY.
  2985 
  2986 * The "contlub" predicate has been removed.  Proof scripts should use
  2987 lemma contI2 in place of monocontlub2cont, INCOMPATIBILITY.
  2988 
  2989 * The "admw" predicate has been removed, INCOMPATIBILITY.
  2990 
  2991 * The constants cpair, cfst, and csnd have been removed in favor of
  2992 Pair, fst, and snd from Isabelle/HOL, INCOMPATIBILITY.
  2993 
  2994 
  2995 *** ML ***
  2996 
  2997 * Antiquotations for basic formal entities:
  2998 
  2999     @{class NAME}         -- type class
  3000     @{class_syntax NAME}  -- syntax representation of the above
  3001 
  3002     @{type_name NAME}     -- logical type
  3003     @{type_abbrev NAME}   -- type abbreviation
  3004     @{nonterminal NAME}   -- type of concrete syntactic category
  3005     @{type_syntax NAME}   -- syntax representation of any of the above
  3006 
  3007     @{const_name NAME}    -- logical constant (INCOMPATIBILITY)
  3008     @{const_abbrev NAME}  -- abbreviated constant
  3009     @{const_syntax NAME}  -- syntax representation of any of the above
  3010 
  3011 * Antiquotation @{syntax_const NAME} ensures that NAME refers to a raw
  3012 syntax constant (cf. 'syntax' command).
  3013 
  3014 * Antiquotation @{make_string} inlines a function to print arbitrary
  3015 values similar to the ML toplevel.  The result is compiler dependent
  3016 and may fall back on "?" in certain situations.
  3017 
  3018 * Diagnostic commands 'ML_val' and 'ML_command' may refer to
  3019 antiquotations @{Isar.state} and @{Isar.goal}.  This replaces impure
  3020 Isar.state() and Isar.goal(), which belong to the old TTY loop and do
  3021 not work with the asynchronous Isar document model.
  3022 
  3023 * Configuration options now admit dynamic default values, depending on
  3024 the context or even global references.
  3025 
  3026 * SHA1.digest digests strings according to SHA-1 (see RFC 3174).  It
  3027 uses an efficient external library if available (for Poly/ML).
  3028 
  3029 * Renamed some important ML structures, while keeping the old names
  3030 for some time as aliases within the structure Legacy:
  3031 
  3032   OuterKeyword  ~>  Keyword
  3033   OuterLex      ~>  Token
  3034   OuterParse    ~>  Parse
  3035   OuterSyntax   ~>  Outer_Syntax
  3036   PrintMode     ~>  Print_Mode
  3037   SpecParse     ~>  Parse_Spec
  3038   ThyInfo       ~>  Thy_Info
  3039   ThyLoad       ~>  Thy_Load
  3040   ThyOutput     ~>  Thy_Output
  3041   TypeInfer     ~>  Type_Infer
  3042 
  3043 Note that "open Legacy" simplifies porting of sources, but forgetting
  3044 to remove it again will complicate porting again in the future.
  3045 
  3046 * Most operations that refer to a global context are named
  3047 accordingly, e.g. Simplifier.global_context or
  3048 ProofContext.init_global.  There are some situations where a global
  3049 context actually works, but under normal circumstances one needs to
  3050 pass the proper local context through the code!
  3051 
  3052 * Discontinued old TheoryDataFun with its copy/init operation -- data
  3053 needs to be pure.  Functor Theory_Data_PP retains the traditional
  3054 Pretty.pp argument to merge, which is absent in the standard
  3055 Theory_Data version.
  3056 
  3057 * Sorts.certify_sort and derived "cert" operations for types and terms
  3058 no longer minimize sorts.  Thus certification at the boundary of the
  3059 inference kernel becomes invariant under addition of class relations,
  3060 which is an important monotonicity principle.  Sorts are now minimized
  3061 in the syntax layer only, at the boundary between the end-user and the
  3062 system.  Subtle INCOMPATIBILITY, may have to use Sign.minimize_sort
  3063 explicitly in rare situations.
  3064 
  3065 * Renamed old-style Drule.standard to Drule.export_without_context, to
  3066 emphasize that this is in no way a standard operation.
  3067 INCOMPATIBILITY.
  3068 
  3069 * Subgoal.FOCUS (and variants): resulting goal state is normalized as
  3070 usual for resolution.  Rare INCOMPATIBILITY.
  3071 
  3072 * Renamed varify/unvarify operations to varify_global/unvarify_global
  3073 to emphasize that these only work in a global situation (which is
  3074 quite rare).
  3075 
  3076 * Curried take and drop in library.ML; negative length is interpreted
  3077 as infinity (as in chop).  Subtle INCOMPATIBILITY.
  3078 
  3079 * Proof terms: type substitutions on proof constants now use canonical
  3080 order of type variables.  INCOMPATIBILITY for tools working with proof
  3081 terms.
  3082 
  3083 * Raw axioms/defs may no longer carry sort constraints, and raw defs
  3084 may no longer carry premises.  User-level specifications are
  3085 transformed accordingly by Thm.add_axiom/add_def.
  3086 
  3087 
  3088 *** System ***
  3089 
  3090 * Discontinued special HOL_USEDIR_OPTIONS for the main HOL image;
  3091 ISABELLE_USEDIR_OPTIONS applies uniformly to all sessions.  Note that
  3092 proof terms are enabled unconditionally in the new HOL-Proofs image.
  3093 
  3094 * Discontinued old ISABELLE and ISATOOL environment settings (legacy
  3095 feature since Isabelle2009).  Use ISABELLE_PROCESS and ISABELLE_TOOL,
  3096 respectively.
  3097 
  3098 * Old lib/scripts/polyml-platform is superseded by the
  3099 ISABELLE_PLATFORM setting variable, which defaults to the 32 bit
  3100 variant, even on a 64 bit machine.  The following example setting
  3101 prefers 64 bit if available:
  3102 
  3103   ML_PLATFORM="${ISABELLE_PLATFORM64:-$ISABELLE_PLATFORM}"
  3104 
  3105 * The preliminary Isabelle/jEdit application demonstrates the emerging
  3106 Isabelle/Scala layer for advanced prover interaction and integration.
  3107 See src/Tools/jEdit or "isabelle jedit" provided by the properly built
  3108 component.
  3109 
  3110 * "IsabelleText" is a Unicode font derived from Bitstream Vera Mono
  3111 and Bluesky TeX fonts.  It provides the usual Isabelle symbols,
  3112 similar to the default assignment of the document preparation system
  3113 (cf. isabellesym.sty).  The Isabelle/Scala class Isabelle_System
  3114 provides some operations for direct access to the font without asking
  3115 the user for manual installation.
  3116 
  3117 
  3118 
  3119 New in Isabelle2009-1 (December 2009)
  3120 -------------------------------------
  3121 
  3122 *** General ***
  3123 
  3124 * Discontinued old form of "escaped symbols" such as \\<forall>.  Only
  3125 one backslash should be used, even in ML sources.
  3126 
  3127 
  3128 *** Pure ***
  3129 
  3130 * Locale interpretation propagates mixins along the locale hierarchy.
  3131 The currently only available mixins are the equations used to map
  3132 local definitions to terms of the target domain of an interpretation.
  3133 
  3134 * Reactivated diagnostic command 'print_interps'.  Use "print_interps
  3135 loc" to print all interpretations of locale "loc" in the theory.
  3136 Interpretations in proofs are not shown.
  3137 
  3138 * Thoroughly revised locales tutorial.  New section on conditional
  3139 interpretation.
  3140 
  3141 * On instantiation of classes, remaining undefined class parameters
  3142 are formally declared.  INCOMPATIBILITY.
  3143 
  3144 
  3145 *** Document preparation ***
  3146 
  3147 * New generalized style concept for printing terms: @{foo (style) ...}
  3148 instead of @{foo_style style ...}  (old form is still retained for
  3149 backward compatibility).  Styles can be also applied for
  3150 antiquotations prop, term_type and typeof.
  3151 
  3152 
  3153 *** HOL ***
  3154 
  3155 * New proof method "smt" for a combination of first-order logic with
  3156 equality, linear and nonlinear (natural/integer/real) arithmetic, and
  3157 fixed-size bitvectors; there is also basic support for higher-order
  3158 features (esp. lambda abstractions).  It is an incomplete decision
  3159 procedure based on external SMT solvers using the oracle mechanism;
  3160 for the SMT solver Z3, this method is proof-producing.  Certificates
  3161 are provided to avoid calling the external solvers solely for
  3162 re-checking proofs.  Due to a remote SMT service there is no need for
  3163 installing SMT solvers locally.  See src/HOL/SMT.
  3164 
  3165 * New commands to load and prove verification conditions generated by
  3166 the Boogie program verifier or derived systems (e.g. the Verifying C
  3167 Compiler (VCC) or Spec#).  See src/HOL/Boogie.
  3168 
  3169 * New counterexample generator tool 'nitpick' based on the Kodkod
  3170 relational model finder.  See src/HOL/Tools/Nitpick and
  3171 src/HOL/Nitpick_Examples.
  3172 
  3173 * New commands 'code_pred' and 'values' to invoke the predicate
  3174 compiler and to enumerate values of inductive predicates.
  3175 
  3176 * A tabled implementation of the reflexive transitive closure.
  3177 
  3178 * New implementation of quickcheck uses generic code generator;
  3179 default generators are provided for all suitable HOL types, records
  3180 and datatypes.  Old quickcheck can be re-activated importing theory
  3181 Library/SML_Quickcheck.
  3182 
  3183 * New testing tool Mirabelle for automated proof tools.  Applies
  3184 several tools and tactics like sledgehammer, metis, or quickcheck, to
  3185 every proof step in a theory.  To be used in batch mode via the
  3186 "mirabelle" utility.
  3187 
  3188 * New proof method "sos" (sum of squares) for nonlinear real
  3189 arithmetic (originally due to John Harison). It requires theory
  3190 Library/Sum_Of_Squares.  It is not a complete decision procedure but
  3191 works well in practice on quantifier-free real arithmetic with +, -,
  3192 *, ^, =, <= and <, i.e. boolean combinations of equalities and
  3193 inequalities between polynomials.  It makes use of external
  3194 semidefinite programming solvers.  Method "sos" generates a
  3195 certificate that can be pasted into the proof thus avoiding the need
  3196 to call an external tool every time the proof is checked.  See
  3197 src/HOL/Library/Sum_Of_Squares.
  3198 
  3199 * New method "linarith" invokes existing linear arithmetic decision
  3200 procedure only.
  3201 
  3202 * New command 'atp_minimal' reduces result produced by Sledgehammer.
  3203 
  3204 * New Sledgehammer option "Full Types" in Proof General settings menu.
  3205 Causes full type information to be output to the ATPs.  This slows
  3206 ATPs down considerably but eliminates a source of unsound "proofs"
  3207 that fail later.
  3208 
  3209 * New method "metisFT": A version of metis that uses full type
  3210 information in order to avoid failures of proof reconstruction.
  3211 
  3212 * New evaluator "approximate" approximates an real valued term using
  3213 the same method as the approximation method.
  3214 
  3215 * Method "approximate" now supports arithmetic expressions as
  3216 boundaries of intervals and implements interval splitting and Taylor
  3217 series expansion.
  3218 
  3219 * ML antiquotation @{code_datatype} inserts definition of a datatype
  3220 generated by the code generator; e.g. see src/HOL/Predicate.thy.
  3221 
  3222 * New theory SupInf of the supremum and infimum operators for sets of
  3223 reals.
  3224 
  3225 * New theory Probability, which contains a development of measure
  3226 theory, eventually leading to Lebesgue integration and probability.
  3227 
  3228 * Extended Multivariate Analysis to include derivation and Brouwer's
  3229 fixpoint theorem.
  3230 
  3231 * Reorganization of number theory, INCOMPATIBILITY:
  3232   - new number theory development for nat and int, in theories Divides
  3233     and GCD as well as in new session Number_Theory
  3234   - some constants and facts now suffixed with _nat and _int
  3235     accordingly
  3236   - former session NumberTheory now named Old_Number_Theory, including
  3237     theories Legacy_GCD and Primes (prefer Number_Theory if possible)
  3238   - moved theory Pocklington from src/HOL/Library to
  3239     src/HOL/Old_Number_Theory
  3240 
  3241 * Theory GCD includes functions Gcd/GCD and Lcm/LCM for the gcd and
  3242 lcm of finite and infinite sets. It is shown that they form a complete
  3243 lattice.
  3244 
  3245 * Class semiring_div requires superclass no_zero_divisors and proof of
  3246 div_mult_mult1; theorems div_mult_mult1, div_mult_mult2,
  3247 div_mult_mult1_if, div_mult_mult1 and div_mult_mult2 have been
  3248 generalized to class semiring_div, subsuming former theorems
  3249 zdiv_zmult_zmult1, zdiv_zmult_zmult1_if, zdiv_zmult_zmult1 and
  3250 zdiv_zmult_zmult2.  div_mult_mult1 is now [simp] by default.
  3251 INCOMPATIBILITY.
  3252 
  3253 * Refinements to lattice classes and sets:
  3254   - less default intro/elim rules in locale variant, more default
  3255     intro/elim rules in class variant: more uniformity
  3256   - lemma ge_sup_conv renamed to le_sup_iff, in accordance with
  3257     le_inf_iff
  3258   - dropped lemma alias inf_ACI for inf_aci (same for sup_ACI and
  3259     sup_aci)
  3260   - renamed ACI to inf_sup_aci
  3261   - new class "boolean_algebra"
  3262   - class "complete_lattice" moved to separate theory
  3263     "Complete_Lattice"; corresponding constants (and abbreviations)
  3264     renamed and with authentic syntax:
  3265     Set.Inf ~>    Complete_Lattice.Inf
  3266     Set.Sup ~>    Complete_Lattice.Sup
  3267     Set.INFI ~>   Complete_Lattice.INFI
  3268     Set.SUPR ~>   Complete_Lattice.SUPR
  3269     Set.Inter ~>  Complete_Lattice.Inter
  3270     Set.Union ~>  Complete_Lattice.Union
  3271     Set.INTER ~>  Complete_Lattice.INTER
  3272     Set.UNION ~>  Complete_Lattice.UNION
  3273   - authentic syntax for
  3274     Set.Pow
  3275     Set.image
  3276   - mere abbreviations:
  3277     Set.empty               (for bot)
  3278     Set.UNIV                (for top)
  3279     Set.inter               (for inf, formerly Set.Int)
  3280     Set.union               (for sup, formerly Set.Un)
  3281     Complete_Lattice.Inter  (for Inf)
  3282     Complete_Lattice.Union  (for Sup)
  3283     Complete_Lattice.INTER  (for INFI)
  3284     Complete_Lattice.UNION  (for SUPR)
  3285   - object-logic definitions as far as appropriate
  3286 
  3287 INCOMPATIBILITY.  Care is required when theorems Int_subset_iff or
  3288 Un_subset_iff are explicitly deleted as default simp rules; then also
  3289 their lattice counterparts le_inf_iff and le_sup_iff have to be
  3290 deleted to achieve the desired effect.
  3291 
  3292 * Rules inf_absorb1, inf_absorb2, sup_absorb1, sup_absorb2 are no simp
  3293 rules by default any longer; the same applies to min_max.inf_absorb1
  3294 etc.  INCOMPATIBILITY.
  3295 
  3296 * Rules sup_Int_eq and sup_Un_eq are no longer declared as
  3297 pred_set_conv by default.  INCOMPATIBILITY.
  3298 
  3299 * Power operations on relations and functions are now one dedicated
  3300 constant "compow" with infix syntax "^^".  Power operation on
  3301 multiplicative monoids retains syntax "^" and is now defined generic
  3302 in class power.  INCOMPATIBILITY.
  3303 
  3304 * Relation composition "R O S" now has a more standard argument order:
  3305 "R O S = {(x, z). EX y. (x, y) : R & (y, z) : S}".  INCOMPATIBILITY,
  3306 rewrite propositions with "S O R" --> "R O S". Proofs may occasionally
  3307 break, since the O_assoc rule was not rewritten like this.  Fix using
  3308 O_assoc[symmetric].  The same applies to the curried version "R OO S".
  3309 
  3310 * Function "Inv" is renamed to "inv_into" and function "inv" is now an
  3311 abbreviation for "inv_into UNIV".  Lemmas are renamed accordingly.
  3312 INCOMPATIBILITY.
  3313 
  3314 * Most rules produced by inductive and datatype package have mandatory
  3315 prefixes.  INCOMPATIBILITY.
  3316 
  3317 * Changed "DERIV_intros" to a dynamic fact, which can be augmented by
  3318 the attribute of the same name.  Each of the theorems in the list
  3319 DERIV_intros assumes composition with an additional function and
  3320 matches a variable to the derivative, which has to be solved by the
  3321 Simplifier.  Hence (auto intro!: DERIV_intros) computes the derivative
  3322 of most elementary terms.  Former Maclauren.DERIV_tac and
  3323 Maclauren.deriv_tac should be replaced by (auto intro!: DERIV_intros).
  3324 INCOMPATIBILITY.
  3325 
  3326 * Code generator attributes follow the usual underscore convention:
  3327     code_unfold     replaces    code unfold
  3328     code_post       replaces    code post
  3329     etc.
  3330   INCOMPATIBILITY.
  3331 
  3332 * Renamed methods:
  3333     sizechange -> size_change
  3334     induct_scheme -> induction_schema
  3335   INCOMPATIBILITY.
  3336 
  3337 * Discontinued abbreviation "arbitrary" of constant "undefined".
  3338 INCOMPATIBILITY, use "undefined" directly.
  3339 
  3340 * Renamed theorems:
  3341     Suc_eq_add_numeral_1 -> Suc_eq_plus1
  3342     Suc_eq_add_numeral_1_left -> Suc_eq_plus1_left
  3343     Suc_plus1 -> Suc_eq_plus1
  3344     *anti_sym -> *antisym*
  3345     vector_less_eq_def -> vector_le_def
  3346   INCOMPATIBILITY.
  3347 
  3348 * Added theorem List.map_map as [simp].  Removed List.map_compose.
  3349 INCOMPATIBILITY.
  3350 
  3351 * Removed predicate "M hassize n" (<--> card M = n & finite M).
  3352 INCOMPATIBILITY.
  3353 
  3354 
  3355 *** HOLCF ***
  3356 
  3357 * Theory Representable defines a class "rep" of domains that are
  3358 representable (via an ep-pair) in the universal domain type "udom".
  3359 Instances are provided for all type constructors defined in HOLCF.
  3360 
  3361 * The 'new_domain' command is a purely definitional version of the
  3362 domain package, for representable domains.  Syntax is identical to the
  3363 old domain package.  The 'new_domain' package also supports indirect
  3364 recursion using previously-defined type constructors.  See
  3365 src/HOLCF/ex/New_Domain.thy for examples.
  3366 
  3367 * Method "fixrec_simp" unfolds one step of a fixrec-defined constant
  3368 on the left-hand side of an equation, and then performs
  3369 simplification.  Rewriting is done using rules declared with the
  3370 "fixrec_simp" attribute.  The "fixrec_simp" method is intended as a
  3371 replacement for "fixpat"; see src/HOLCF/ex/Fixrec_ex.thy for examples.
  3372 
  3373 * The pattern-match compiler in 'fixrec' can now handle constructors
  3374 with HOL function types.  Pattern-match combinators for the Pair
  3375 constructor are pre-configured.
  3376 
  3377 * The 'fixrec' package now produces better fixed-point induction rules
  3378 for mutually-recursive definitions:  Induction rules have conclusions
  3379 of the form "P foo bar" instead of "P <foo, bar>".
  3380 
  3381 * The constant "sq_le" (with infix syntax "<<" or "\<sqsubseteq>") has
  3382 been renamed to "below".  The name "below" now replaces "less" in many
  3383 theorem names.  (Legacy theorem names using "less" are still supported
  3384 as well.)
  3385 
  3386 * The 'fixrec' package now supports "bottom patterns".  Bottom
  3387 patterns can be used to generate strictness rules, or to make
  3388 functions more strict (much like the bang-patterns supported by the
  3389 Glasgow Haskell Compiler).  See src/HOLCF/ex/Fixrec_ex.thy for
  3390 examples.
  3391 
  3392 
  3393 *** ML ***
  3394 
  3395 * Support for Poly/ML 5.3.0, with improved reporting of compiler
  3396 errors and run-time exceptions, including detailed source positions.
  3397 
  3398 * Structure Name_Space (formerly NameSpace) now manages uniquely
  3399 identified entries, with some additional information such as source
  3400 position, logical grouping etc.
  3401 
  3402 * Theory and context data is now introduced by the simplified and
  3403 modernized functors Theory_Data, Proof_Data, Generic_Data.  Data needs
  3404 to be pure, but the old TheoryDataFun for mutable data (with explicit
  3405 copy operation) is still available for some time.
  3406 
  3407 * Structure Synchronized (cf. src/Pure/Concurrent/synchronized.ML)
  3408 provides a high-level programming interface to synchronized state
  3409 variables with atomic update.  This works via pure function
  3410 application within a critical section -- its runtime should be as
  3411 short as possible; beware of deadlocks if critical code is nested,
  3412 either directly or indirectly via other synchronized variables!
  3413 
  3414 * Structure Unsynchronized (cf. src/Pure/ML-Systems/unsynchronized.ML)
  3415 wraps raw ML references, explicitly indicating their non-thread-safe
  3416 behaviour.  The Isar toplevel keeps this structure open, to
  3417 accommodate Proof General as well as quick and dirty interactive
  3418 experiments with references.
  3419 
  3420 * PARALLEL_CHOICE and PARALLEL_GOALS provide basic support for
  3421 parallel tactical reasoning.
  3422 
  3423 * Tacticals Subgoal.FOCUS, Subgoal.FOCUS_PREMS, Subgoal.FOCUS_PARAMS
  3424 are similar to SUBPROOF, but are slightly more flexible: only the
  3425 specified parts of the subgoal are imported into the context, and the
  3426 body tactic may introduce new subgoals and schematic variables.
  3427 
  3428 * Old tactical METAHYPS, which does not observe the proof context, has
  3429 been renamed to Old_Goals.METAHYPS and awaits deletion.  Use SUBPROOF
  3430 or Subgoal.FOCUS etc.
  3431 
  3432 * Renamed functor TableFun to Table, and GraphFun to Graph.  (Since
  3433 functors have their own ML name space there is no point to mark them
  3434 separately.)  Minor INCOMPATIBILITY.
  3435 
  3436 * Renamed NamedThmsFun to Named_Thms.  INCOMPATIBILITY.
  3437 
  3438 * Renamed several structures FooBar to Foo_Bar.  Occasional,
  3439 INCOMPATIBILITY.
  3440 
  3441 * Operations of structure Skip_Proof no longer require quick_and_dirty
  3442 mode, which avoids critical setmp.
  3443 
  3444 * Eliminated old Attrib.add_attributes, Method.add_methods and related
  3445 combinators for "args".  INCOMPATIBILITY, need to use simplified
  3446 Attrib/Method.setup introduced in Isabelle2009.
  3447 
  3448 * Proper context for simpset_of, claset_of, clasimpset_of.  May fall
  3449 back on global_simpset_of, global_claset_of, global_clasimpset_of as
  3450 last resort.  INCOMPATIBILITY.
  3451 
  3452 * Display.pretty_thm now requires a proper context (cf. former
  3453 ProofContext.pretty_thm).  May fall back on Display.pretty_thm_global
  3454 or even Display.pretty_thm_without_context as last resort.
  3455 INCOMPATIBILITY.
  3456 
  3457 * Discontinued Display.pretty_ctyp/cterm etc.  INCOMPATIBILITY, use
  3458 Syntax.pretty_typ/term directly, preferably with proper context
  3459 instead of global theory.
  3460 
  3461 
  3462 *** System ***
  3463 
  3464 * Further fine tuning of parallel proof checking, scales up to 8 cores
  3465 (max. speedup factor 5.0).  See also Goal.parallel_proofs in ML and
  3466 usedir option -q.
  3467 
  3468 * Support for additional "Isabelle components" via etc/components, see
  3469 also the system manual.
  3470 
  3471 * The isabelle makeall tool now operates on all components with
  3472 IsaMakefile, not just hardwired "logics".
  3473 
  3474 * Removed "compress" option from isabelle-process and isabelle usedir;
  3475 this is always enabled.
  3476 
  3477 * Discontinued support for Poly/ML 4.x versions.
  3478 
  3479 * Isabelle tool "wwwfind" provides web interface for 'find_theorems'
  3480 on a given logic image.  This requires the lighttpd webserver and is
  3481 currently supported on Linux only.
  3482 
  3483 
  3484 
  3485 New in Isabelle2009 (April 2009)
  3486 --------------------------------
  3487 
  3488 *** General ***
  3489 
  3490 * Simplified main Isabelle executables, with less surprises on
  3491 case-insensitive file-systems (such as Mac OS).
  3492 
  3493   - The main Isabelle tool wrapper is now called "isabelle" instead of
  3494     "isatool."
  3495 
  3496   - The former "isabelle" alias for "isabelle-process" has been
  3497     removed (should rarely occur to regular users).
  3498 
  3499   - The former "isabelle-interface" and its alias "Isabelle" have been
  3500     removed (interfaces are now regular Isabelle tools).
  3501 
  3502 Within scripts and make files, the Isabelle environment variables
  3503 ISABELLE_TOOL and ISABELLE_PROCESS replace old ISATOOL and ISABELLE,
  3504 respectively.  (The latter are still available as legacy feature.)
  3505 
  3506 The old isabelle-interface wrapper could react in confusing ways if
  3507 the interface was uninstalled or changed otherwise.  Individual
  3508 interface tool configuration is now more explicit, see also the
  3509 Isabelle system manual.  In particular, Proof General is now available
  3510 via "isabelle emacs".
  3511 
  3512 INCOMPATIBILITY, need to adapt derivative scripts.  Users may need to
  3513 purge installed copies of Isabelle executables and re-run "isabelle
  3514 install -p ...", or use symlinks.
  3515 
  3516 * The default for ISABELLE_HOME_USER is now ~/.isabelle instead of the
  3517 old ~/isabelle, which was slightly non-standard and apt to cause
  3518 surprises on case-insensitive file-systems (such as Mac OS).
  3519 
  3520 INCOMPATIBILITY, need to move existing ~/isabelle/etc,
  3521 ~/isabelle/heaps, ~/isabelle/browser_info to the new place.  Special
  3522 care is required when using older releases of Isabelle.  Note that
  3523 ISABELLE_HOME_USER can be changed in Isabelle/etc/settings of any
  3524 Isabelle distribution, in order to use the new ~/.isabelle uniformly.
  3525 
  3526 * Proofs of fully specified statements are run in parallel on
  3527 multi-core systems.  A speedup factor of 2.5 to 3.2 can be expected on
  3528 a regular 4-core machine, if the initial heap space is made reasonably
  3529 large (cf. Poly/ML option -H).  (Requires Poly/ML 5.2.1 or later.)
  3530 
  3531 * The main reference manuals ("isar-ref", "implementation", and
  3532 "system") have been updated and extended.  Formally checked references
  3533 as hyperlinks are now available uniformly.
  3534 
  3535 
  3536 *** Pure ***
  3537 
  3538 * Complete re-implementation of locales.  INCOMPATIBILITY in several
  3539 respects.  The most important changes are listed below.  See the
  3540 Tutorial on Locales ("locales" manual) for details.
  3541 
  3542 - In locale expressions, instantiation replaces renaming.  Parameters
  3543 must be declared in a for clause.  To aid compatibility with previous
  3544 parameter inheritance, in locale declarations, parameters that are not
  3545 'touched' (instantiation position "_" or omitted) are implicitly added
  3546 with their syntax at the beginning of the for clause.
  3547 
  3548 - Syntax from abbreviations and definitions in locales is available in
  3549 locale expressions and context elements.  The latter is particularly
  3550 useful in locale declarations.
  3551 
  3552 - More flexible mechanisms to qualify names generated by locale
  3553 expressions.  Qualifiers (prefixes) may be specified in locale
  3554 expressions, and can be marked as mandatory (syntax: "name!:") or
  3555 optional (syntax "name?:").  The default depends for plain "name:"
  3556 depends on the situation where a locale expression is used: in
  3557 commands 'locale' and 'sublocale' prefixes are optional, in
  3558 'interpretation' and 'interpret' prefixes are mandatory.  The old
  3559 implicit qualifiers derived from the parameter names of a locale are
  3560 no longer generated.
  3561 
  3562 - Command "sublocale l < e" replaces "interpretation l < e".  The
  3563 instantiation clause in "interpretation" and "interpret" (square
  3564 brackets) is no longer available.  Use locale expressions.
  3565 
  3566 - When converting proof scripts, mandatory qualifiers in
  3567 'interpretation' and 'interpret' should be retained by default, even
  3568 if this is an INCOMPATIBILITY compared to former behavior.  In the
  3569 worst case, use the "name?:" form for non-mandatory ones.  Qualifiers
  3570 in locale expressions range over a single locale instance only.
  3571 
  3572 - Dropped locale element "includes".  This is a major INCOMPATIBILITY.
  3573 In existing theorem specifications replace the includes element by the
  3574 respective context elements of the included locale, omitting those
  3575 that are already present in the theorem specification.  Multiple
  3576 assume elements of a locale should be replaced by a single one
  3577 involving the locale predicate.  In the proof body, declarations (most
  3578 notably theorems) may be regained by interpreting the respective
  3579 locales in the proof context as required (command "interpret").
  3580 
  3581 If using "includes" in replacement of a target solely because the
  3582 parameter types in the theorem are not as general as in the target,
  3583 consider declaring a new locale with additional type constraints on
  3584 the parameters (context element "constrains").
  3585 
  3586 - Discontinued "locale (open)".  INCOMPATIBILITY.
  3587 
  3588 - Locale interpretation commands no longer attempt to simplify goal.
  3589 INCOMPATIBILITY: in rare situations the generated goal differs.  Use
  3590 methods intro_locales and unfold_locales to clarify.
  3591 
  3592 - Locale interpretation commands no longer accept interpretation
  3593 attributes.  INCOMPATIBILITY.
  3594 
  3595 * Class declaration: so-called "base sort" must not be given in import
  3596 list any longer, but is inferred from the specification.  Particularly
  3597 in HOL, write
  3598 
  3599     class foo = ...
  3600 
  3601 instead of
  3602 
  3603     class foo = type + ...
  3604 
  3605 * Class target: global versions of theorems stemming do not carry a
  3606 parameter prefix any longer.  INCOMPATIBILITY.
  3607 
  3608 * Class 'instance' command no longer accepts attached definitions.
  3609 INCOMPATIBILITY, use proper 'instantiation' target instead.
  3610 
  3611 * Recovered hiding of consts, which was accidentally broken in
  3612 Isabelle2007.  Potential INCOMPATIBILITY, ``hide const c'' really
  3613 makes c inaccessible; consider using ``hide (open) const c'' instead.
  3614 
  3615 * Slightly more coherent Pure syntax, with updated documentation in
  3616 isar-ref manual.  Removed locales meta_term_syntax and
  3617 meta_conjunction_syntax: TERM and &&& (formerly &&) are now permanent,
  3618 INCOMPATIBILITY in rare situations.  Note that &&& should not be used
  3619 directly in regular applications.
  3620 
  3621 * There is a new syntactic category "float_const" for signed decimal
  3622 fractions (e.g. 123.45 or -123.45).
  3623 
  3624 * Removed exotic 'token_translation' command.  INCOMPATIBILITY, use ML
  3625 interface with 'setup' command instead.
  3626 
  3627 * Command 'local_setup' is similar to 'setup', but operates on a local
  3628 theory context.
  3629 
  3630 * The 'axiomatization' command now only works within a global theory
  3631 context.  INCOMPATIBILITY.
  3632 
  3633 * Goal-directed proof now enforces strict proof irrelevance wrt. sort
  3634 hypotheses.  Sorts required in the course of reasoning need to be
  3635 covered by the constraints in the initial statement, completed by the
  3636 type instance information of the background theory.  Non-trivial sort
  3637 hypotheses, which rarely occur in practice, may be specified via
  3638 vacuous propositions of the form SORT_CONSTRAINT('a::c).  For example:
  3639 
  3640   lemma assumes "SORT_CONSTRAINT('a::empty)" shows False ...
  3641 
  3642 The result contains an implicit sort hypotheses as before --
  3643 SORT_CONSTRAINT premises are eliminated as part of the canonical rule
  3644 normalization.
  3645 
  3646 * Generalized Isar history, with support for linear undo, direct state
  3647 addressing etc.
  3648 
  3649 * Changed defaults for unify configuration options:
  3650 
  3651   unify_trace_bound = 50 (formerly 25)
  3652   unify_search_bound = 60 (formerly 30)
  3653 
  3654 * Different bookkeeping for code equations (INCOMPATIBILITY):
  3655 
  3656   a) On theory merge, the last set of code equations for a particular
  3657      constant is taken (in accordance with the policy applied by other
  3658      parts of the code generator framework).
  3659 
  3660   b) Code equations stemming from explicit declarations (e.g. code
  3661      attribute) gain priority over default code equations stemming
  3662      from definition, primrec, fun etc.
  3663 
  3664 * Keyword 'code_exception' now named 'code_abort'.  INCOMPATIBILITY.
  3665 
  3666 * Unified theorem tables for both code generators.  Thus [code
  3667 func] has disappeared and only [code] remains.  INCOMPATIBILITY.
  3668 
  3669 * Command 'find_consts' searches for constants based on type and name
  3670 patterns, e.g.
  3671 
  3672     find_consts "_ => bool"
  3673 
  3674 By default, matching is against subtypes, but it may be restricted to
  3675 the whole type.  Searching by name is possible.  Multiple queries are
  3676 conjunctive and queries may be negated by prefixing them with a
  3677 hyphen:
  3678 
  3679     find_consts strict: "_ => bool" name: "Int" -"int => int"
  3680 
  3681 * New 'find_theorems' criterion "solves" matches theorems that
  3682 directly solve the current goal (modulo higher-order unification).
  3683 
  3684 * Auto solve feature for main theorem statements: whenever a new goal
  3685 is stated, "find_theorems solves" is called; any theorems that could
  3686 solve the lemma directly are listed as part of the goal state.
  3687 Cf. associated options in Proof General Isabelle settings menu,
  3688 enabled by default, with reasonable timeout for pathological cases of
  3689 higher-order unification.
  3690 
  3691 
  3692 *** Document preparation ***
  3693 
  3694 * Antiquotation @{lemma} now imitates a regular terminal proof,
  3695 demanding keyword 'by' and supporting the full method expression
  3696 syntax just like the Isar command 'by'.
  3697 
  3698 
  3699 *** HOL ***
  3700 
  3701 * Integrated main parts of former image HOL-Complex with HOL.  Entry
  3702 points Main and Complex_Main remain as before.
  3703 
  3704 * Logic image HOL-Plain provides a minimal HOL with the most important
  3705 tools available (inductive, datatype, primrec, ...).  This facilitates
  3706 experimentation and tool development.  Note that user applications
  3707 (and library theories) should never refer to anything below theory
  3708 Main, as before.
  3709 
  3710 * Logic image HOL-Main stops at theory Main, and thus facilitates
  3711 experimentation due to shorter build times.
  3712 
  3713 * Logic image HOL-NSA contains theories of nonstandard analysis which
  3714 were previously part of former HOL-Complex.  Entry point Hyperreal
  3715 remains valid, but theories formerly using Complex_Main should now use
  3716 new entry point Hypercomplex.
  3717 
  3718 * Generic ATP manager for Sledgehammer, based on ML threads instead of
  3719 Posix processes.  Avoids potentially expensive forking of the ML
  3720 process.  New thread-based implementation also works on non-Unix
  3721 platforms (Cygwin).  Provers are no longer hardwired, but defined
  3722 within the theory via plain ML wrapper functions.  Basic Sledgehammer
  3723 commands are covered in the isar-ref manual.
  3724 
  3725 * Wrapper scripts for remote SystemOnTPTP service allows to use
  3726 sledgehammer without local ATP installation (Vampire etc.). Other
  3727 provers may be included via suitable ML wrappers, see also
  3728 src/HOL/ATP_Linkup.thy.
  3729 
  3730 * ATP selection (E/Vampire/Spass) is now via Proof General's settings
  3731 menu.
  3732 
  3733 * The metis method no longer fails because the theorem is too trivial
  3734 (contains the empty clause).
  3735 
  3736 * The metis method now fails in the usual manner, rather than raising
  3737 an exception, if it determines that it cannot prove the theorem.
  3738 
  3739 * Method "coherent" implements a prover for coherent logic (see also
  3740 src/Tools/coherent.ML).
  3741 
  3742 * Constants "undefined" and "default" replace "arbitrary".  Usually
  3743 "undefined" is the right choice to replace "arbitrary", though
  3744 logically there is no difference.  INCOMPATIBILITY.
  3745 
  3746 * Command "value" now integrates different evaluation mechanisms.  The
  3747 result of the first successful evaluation mechanism is printed.  In
  3748 square brackets a particular named evaluation mechanisms may be
  3749 specified (currently, [SML], [code] or [nbe]).  See further
  3750 src/HOL/ex/Eval_Examples.thy.
  3751 
  3752 * Normalization by evaluation now allows non-leftlinear equations.
  3753 Declare with attribute [code nbe].
  3754 
  3755 * Methods "case_tac" and "induct_tac" now refer to the very same rules
  3756 as the structured Isar versions "cases" and "induct", cf. the
  3757 corresponding "cases" and "induct" attributes.  Mutual induction rules
  3758 are now presented as a list of individual projections
  3759 (e.g. foo_bar.inducts for types foo and bar); the old format with
  3760 explicit HOL conjunction is no longer supported.  INCOMPATIBILITY, in
  3761 rare situations a different rule is selected --- notably nested tuple
  3762 elimination instead of former prod.exhaust: use explicit (case_tac t
  3763 rule: prod.exhaust) here.
  3764 
  3765 * Attributes "cases", "induct", "coinduct" support "del" option.
  3766 
  3767 * Removed fact "case_split_thm", which duplicates "case_split".
  3768 
  3769 * The option datatype has been moved to a new theory Option.  Renamed
  3770 option_map to Option.map, and o2s to Option.set, INCOMPATIBILITY.
  3771 
  3772 * New predicate "strict_mono" classifies strict functions on partial
  3773 orders.  With strict functions on linear orders, reasoning about
  3774 (in)equalities is facilitated by theorems "strict_mono_eq",
  3775 "strict_mono_less_eq" and "strict_mono_less".
  3776 
  3777 * Some set operations are now proper qualified constants with
  3778 authentic syntax.  INCOMPATIBILITY:
  3779 
  3780     op Int ~>   Set.Int
  3781     op Un ~>    Set.Un
  3782     INTER ~>    Set.INTER
  3783     UNION ~>    Set.UNION
  3784     Inter ~>    Set.Inter
  3785     Union ~>    Set.Union
  3786     {} ~>       Set.empty
  3787     UNIV ~>     Set.UNIV
  3788 
  3789 * Class complete_lattice with operations Inf, Sup, INFI, SUPR now in
  3790 theory Set.
  3791 
  3792 * Auxiliary class "itself" has disappeared -- classes without any
  3793 parameter are treated as expected by the 'class' command.
  3794 
  3795 * Leibnitz's Series for Pi and the arcus tangens and logarithm series.
  3796 
  3797 * Common decision procedures (Cooper, MIR, Ferrack, Approximation,
  3798 Dense_Linear_Order) are now in directory HOL/Decision_Procs.
  3799 
  3800 * Theory src/HOL/Decision_Procs/Approximation provides the new proof
  3801 method "approximation".  It proves formulas on real values by using
  3802 interval arithmetic.  In the formulas are also the transcendental
  3803 functions sin, cos, tan, atan, ln, exp and the constant pi are
  3804 allowed. For examples see
  3805 src/HOL/Descision_Procs/ex/Approximation_Ex.thy.
  3806 
  3807 * Theory "Reflection" now resides in HOL/Library.
  3808 
  3809 * Entry point to Word library now simply named "Word".
  3810 INCOMPATIBILITY.
  3811 
  3812 * Made source layout more coherent with logical distribution
  3813 structure:
  3814 
  3815     src/HOL/Library/RType.thy ~> src/HOL/Typerep.thy
  3816     src/HOL/Library/Code_Message.thy ~> src/HOL/
  3817     src/HOL/Library/GCD.thy ~> src/HOL/
  3818     src/HOL/Library/Order_Relation.thy ~> src/HOL/
  3819     src/HOL/Library/Parity.thy ~> src/HOL/
  3820     src/HOL/Library/Univ_Poly.thy ~> src/HOL/
  3821     src/HOL/Real/ContNotDenum.thy ~> src/HOL/Library/
  3822     src/HOL/Real/Lubs.thy ~> src/HOL/
  3823     src/HOL/Real/PReal.thy ~> src/HOL/
  3824     src/HOL/Real/Rational.thy ~> src/HOL/
  3825     src/HOL/Real/RComplete.thy ~> src/HOL/
  3826     src/HOL/Real/RealDef.thy ~> src/HOL/
  3827     src/HOL/Real/RealPow.thy ~> src/HOL/
  3828     src/HOL/Real/Real.thy ~> src/HOL/
  3829     src/HOL/Complex/Complex_Main.thy ~> src/HOL/
  3830     src/HOL/Complex/Complex.thy ~> src/HOL/
  3831     src/HOL/Complex/FrechetDeriv.thy ~> src/HOL/Library/
  3832     src/HOL/Complex/Fundamental_Theorem_Algebra.thy ~> src/HOL/Library/
  3833     src/HOL/Hyperreal/Deriv.thy ~> src/HOL/
  3834     src/HOL/Hyperreal/Fact.thy ~> src/HOL/
  3835     src/HOL/Hyperreal/Integration.thy ~> src/HOL/
  3836     src/HOL/Hyperreal/Lim.thy ~> src/HOL/
  3837     src/HOL/Hyperreal/Ln.thy ~> src/HOL/
  3838     src/HOL/Hyperreal/Log.thy ~> src/HOL/
  3839     src/HOL/Hyperreal/MacLaurin.thy ~> src/HOL/
  3840     src/HOL/Hyperreal/NthRoot.thy ~> src/HOL/
  3841     src/HOL/Hyperreal/Series.thy ~> src/HOL/
  3842     src/HOL/Hyperreal/SEQ.thy ~> src/HOL/
  3843     src/HOL/Hyperreal/Taylor.thy ~> src/HOL/
  3844     src/HOL/Hyperreal/Transcendental.thy ~> src/HOL/
  3845     src/HOL/Real/Float ~> src/HOL/Library/
  3846     src/HOL/Real/HahnBanach ~> src/HOL/HahnBanach
  3847     src/HOL/Real/RealVector.thy ~> src/HOL/
  3848 
  3849     src/HOL/arith_data.ML ~> src/HOL/Tools
  3850     src/HOL/hologic.ML ~> src/HOL/Tools
  3851     src/HOL/simpdata.ML ~> src/HOL/Tools
  3852     src/HOL/int_arith1.ML ~> src/HOL/Tools/int_arith.ML
  3853     src/HOL/int_factor_simprocs.ML ~> src/HOL/Tools
  3854     src/HOL/nat_simprocs.ML ~> src/HOL/Tools
  3855     src/HOL/Real/float_arith.ML ~> src/HOL/Tools
  3856     src/HOL/Real/float_syntax.ML ~> src/HOL/Tools
  3857     src/HOL/Real/rat_arith.ML ~> src/HOL/Tools
  3858     src/HOL/Real/real_arith.ML ~> src/HOL/Tools
  3859 
  3860     src/HOL/Library/Array.thy ~> src/HOL/Imperative_HOL
  3861     src/HOL/Library/Heap_Monad.thy ~> src/HOL/Imperative_HOL
  3862     src/HOL/Library/Heap.thy ~> src/HOL/Imperative_HOL
  3863     src/HOL/Library/Imperative_HOL.thy ~> src/HOL/Imperative_HOL
  3864     src/HOL/Library/Ref.thy ~> src/HOL/Imperative_HOL
  3865     src/HOL/Library/Relational.thy ~> src/HOL/Imperative_HOL
  3866 
  3867 * If methods "eval" and "evaluation" encounter a structured proof
  3868 state with !!/==>, only the conclusion is evaluated to True (if
  3869 possible), avoiding strange error messages.
  3870 
  3871 * Method "sizechange" automates termination proofs using (a
  3872 modification of) the size-change principle.  Requires SAT solver.  See
  3873 src/HOL/ex/Termination.thy for examples.
  3874 
  3875 * Simplifier: simproc for let expressions now unfolds if bound
  3876 variable occurs at most once in let expression body.  INCOMPATIBILITY.
  3877 
  3878 * Method "arith": Linear arithmetic now ignores all inequalities when
  3879 fast_arith_neq_limit is exceeded, instead of giving up entirely.
  3880 
  3881 * New attribute "arith" for facts that should always be used
  3882 automatically by arithmetic. It is intended to be used locally in
  3883 proofs, e.g.
  3884 
  3885   assumes [arith]: "x > 0"
  3886 
  3887 Global usage is discouraged because of possible performance impact.
  3888 
  3889 * New classes "top" and "bot" with corresponding operations "top" and
  3890 "bot" in theory Orderings; instantiation of class "complete_lattice"
  3891 requires instantiation of classes "top" and "bot".  INCOMPATIBILITY.
  3892 
  3893 * Changed definition lemma "less_fun_def" in order to provide an
  3894 instance for preorders on functions; use lemma "less_le" instead.
  3895 INCOMPATIBILITY.
  3896 
  3897 * Theory Orderings: class "wellorder" moved here, with explicit
  3898 induction rule "less_induct" as assumption.  For instantiation of
  3899 "wellorder" by means of predicate "wf", use rule wf_wellorderI.
  3900 INCOMPATIBILITY.
  3901 
  3902 * Theory Orderings: added class "preorder" as superclass of "order".
  3903 INCOMPATIBILITY: Instantiation proofs for order, linorder
  3904 etc. slightly changed.  Some theorems named order_class.* now named
  3905 preorder_class.*.
  3906 
  3907 * Theory Relation: renamed "refl" to "refl_on", "reflexive" to "refl,
  3908 "diag" to "Id_on".
  3909 
  3910 * Theory Finite_Set: added a new fold combinator of type
  3911 
  3912   ('a => 'b => 'b) => 'b => 'a set => 'b
  3913 
  3914 Occasionally this is more convenient than the old fold combinator
  3915 which is now defined in terms of the new one and renamed to
  3916 fold_image.
  3917 
  3918 * Theories Ring_and_Field and OrderedGroup: The lemmas "group_simps"
  3919 and "ring_simps" have been replaced by "algebra_simps" (which can be
  3920 extended with further lemmas!).  At the moment both still exist but
  3921 the former will disappear at some point.
  3922 
  3923 * Theory Power: Lemma power_Suc is now declared as a simp rule in
  3924 class recpower.  Type-specific simp rules for various recpower types
  3925 have been removed.  INCOMPATIBILITY, rename old lemmas as follows:
  3926 
  3927 rat_power_0    -> power_0
  3928 rat_power_Suc  -> power_Suc
  3929 realpow_0      -> power_0
  3930 realpow_Suc    -> power_Suc
  3931 complexpow_0   -> power_0
  3932 complexpow_Suc -> power_Suc
  3933 power_poly_0   -> power_0
  3934 power_poly_Suc -> power_Suc
  3935 
  3936 * Theories Ring_and_Field and Divides: Definition of "op dvd" has been
  3937 moved to separate class dvd in Ring_and_Field; a couple of lemmas on
  3938 dvd has been generalized to class comm_semiring_1.  Likewise a bunch
  3939 of lemmas from Divides has been generalized from nat to class
  3940 semiring_div.  INCOMPATIBILITY.  This involves the following theorem
  3941 renames resulting from duplicate elimination:
  3942 
  3943     dvd_def_mod ~>          dvd_eq_mod_eq_0
  3944     zero_dvd_iff ~>         dvd_0_left_iff
  3945     dvd_0 ~>                dvd_0_right
  3946     DIVISION_BY_ZERO_DIV ~> div_by_0
  3947     DIVISION_BY_ZERO_MOD ~> mod_by_0
  3948     mult_div ~>             div_mult_self2_is_id
  3949     mult_mod ~>             mod_mult_self2_is_0
  3950 
  3951 * Theory IntDiv: removed many lemmas that are instances of class-based
  3952 generalizations (from Divides and Ring_and_Field).  INCOMPATIBILITY,
  3953 rename old lemmas as follows:
  3954 
  3955 dvd_diff               -> nat_dvd_diff
  3956 dvd_zminus_iff         -> dvd_minus_iff
  3957 mod_add1_eq            -> mod_add_eq
  3958 mod_mult1_eq           -> mod_mult_right_eq
  3959 mod_mult1_eq'          -> mod_mult_left_eq
  3960 mod_mult_distrib_mod   -> mod_mult_eq
  3961 nat_mod_add_left_eq    -> mod_add_left_eq
  3962 nat_mod_add_right_eq   -> mod_add_right_eq
  3963 nat_mod_div_trivial    -> mod_div_trivial
  3964 nat_mod_mod_trivial    -> mod_mod_trivial
  3965 zdiv_zadd_self1        -> div_add_self1
  3966 zdiv_zadd_self2        -> div_add_self2
  3967 zdiv_zmult_self1       -> div_mult_self2_is_id
  3968 zdiv_zmult_self2       -> div_mult_self1_is_id
  3969 zdvd_triv_left         -> dvd_triv_left
  3970 zdvd_triv_right        -> dvd_triv_right
  3971 zdvd_zmult_cancel_disj -> dvd_mult_cancel_left
  3972 zmod_eq0_zdvd_iff      -> dvd_eq_mod_eq_0[symmetric]
  3973 zmod_zadd_left_eq      -> mod_add_left_eq
  3974 zmod_zadd_right_eq     -> mod_add_right_eq
  3975 zmod_zadd_self1        -> mod_add_self1
  3976 zmod_zadd_self2        -> mod_add_self2
  3977 zmod_zadd1_eq          -> mod_add_eq
  3978 zmod_zdiff1_eq         -> mod_diff_eq
  3979 zmod_zdvd_zmod         -> mod_mod_cancel
  3980 zmod_zmod_cancel       -> mod_mod_cancel
  3981 zmod_zmult_self1       -> mod_mult_self2_is_0
  3982 zmod_zmult_self2       -> mod_mult_self1_is_0
  3983 zmod_1                 -> mod_by_1
  3984 zdiv_1                 -> div_by_1
  3985 zdvd_abs1              -> abs_dvd_iff
  3986 zdvd_abs2              -> dvd_abs_iff
  3987 zdvd_refl              -> dvd_refl
  3988 zdvd_trans             -> dvd_trans
  3989 zdvd_zadd              -> dvd_add
  3990 zdvd_zdiff             -> dvd_diff
  3991 zdvd_zminus_iff        -> dvd_minus_iff
  3992 zdvd_zminus2_iff       -> minus_dvd_iff
  3993 zdvd_zmultD            -> dvd_mult_right
  3994 zdvd_zmultD2           -> dvd_mult_left
  3995 zdvd_zmult_mono        -> mult_dvd_mono
  3996 zdvd_0_right           -> dvd_0_right
  3997 zdvd_0_left            -> dvd_0_left_iff
  3998 zdvd_1_left            -> one_dvd
  3999 zminus_dvd_iff         -> minus_dvd_iff
  4000 
  4001 * Theory Rational: 'Fract k 0' now equals '0'.  INCOMPATIBILITY.
  4002 
  4003 * The real numbers offer decimal input syntax: 12.34 is translated
  4004 into 1234/10^2. This translation is not reversed upon output.
  4005 
  4006 * Theory Library/Polynomial defines an abstract type 'a poly of
  4007 univariate polynomials with coefficients of type 'a.  In addition to
  4008 the standard ring operations, it also supports div and mod.  Code
  4009 generation is also supported, using list-style constructors.
  4010 
  4011 * Theory Library/Inner_Product defines a class of real_inner for real
  4012 inner product spaces, with an overloaded operation inner :: 'a => 'a
  4013 => real.  Class real_inner is a subclass of real_normed_vector from
  4014 theory RealVector.
  4015 
  4016 * Theory Library/Product_Vector provides instances for the product
  4017 type 'a * 'b of several classes from RealVector and Inner_Product.
  4018 Definitions of addition, subtraction, scalar multiplication, norms,
  4019 and inner products are included.
  4020 
  4021 * Theory Library/Bit defines the field "bit" of integers modulo 2.  In
  4022 addition to the field operations, numerals and case syntax are also
  4023 supported.
  4024 
  4025 * Theory Library/Diagonalize provides constructive version of Cantor's
  4026 first diagonalization argument.
  4027 
  4028 * Theory Library/GCD: Curried operations gcd, lcm (for nat) and zgcd,
  4029 zlcm (for int); carried together from various gcd/lcm developements in
  4030 the HOL Distribution.  Constants zgcd and zlcm replace former igcd and
  4031 ilcm; corresponding theorems renamed accordingly.  INCOMPATIBILITY,
  4032 may recover tupled syntax as follows:
  4033 
  4034     hide (open) const gcd
  4035     abbreviation gcd where
  4036       "gcd == (%(a, b). GCD.gcd a b)"
  4037     notation (output)
  4038       GCD.gcd ("gcd '(_, _')")
  4039 
  4040 The same works for lcm, zgcd, zlcm.
  4041 
  4042 * Theory Library/Nat_Infinity: added addition, numeral syntax and more
  4043 instantiations for algebraic structures.  Removed some duplicate
  4044 theorems.  Changes in simp rules.  INCOMPATIBILITY.
  4045 
  4046 * ML antiquotation @{code} takes a constant as argument and generates
  4047 corresponding code in background and inserts name of the corresponding
  4048 resulting ML value/function/datatype constructor binding in place.
  4049 All occurrences of @{code} with a single ML block are generated
  4050 simultaneously.  Provides a generic and safe interface for
  4051 instrumentalizing code generation.  See
  4052 src/HOL/Decision_Procs/Ferrack.thy for a more ambitious application.
  4053 In future you ought to refrain from ad-hoc compiling generated SML
  4054 code on the ML toplevel.  Note that (for technical reasons) @{code}
  4055 cannot refer to constants for which user-defined serializations are
  4056 set.  Refer to the corresponding ML counterpart directly in that
  4057 cases.
  4058 
  4059 * Command 'rep_datatype': instead of theorem names the command now
  4060 takes a list of terms denoting the constructors of the type to be
  4061 represented as datatype.  The characteristic theorems have to be
  4062 proven.  INCOMPATIBILITY.  Also observe that the following theorems
  4063 have disappeared in favour of existing ones:
  4064 
  4065     unit_induct                 ~> unit.induct
  4066     prod_induct                 ~> prod.induct
  4067     sum_induct                  ~> sum.induct
  4068     Suc_Suc_eq                  ~> nat.inject
  4069     Suc_not_Zero Zero_not_Suc   ~> nat.distinct
  4070 
  4071 
  4072 *** HOL-Algebra ***
  4073 
  4074 * New locales for orders and lattices where the equivalence relation
  4075 is not restricted to equality.  INCOMPATIBILITY: all order and lattice
  4076 locales use a record structure with field eq for the equivalence.
  4077 
  4078 * New theory of factorial domains.
  4079 
  4080 * Units_l_inv and Units_r_inv are now simp rules by default.
  4081 INCOMPATIBILITY.  Simplifier proof that require deletion of l_inv
  4082 and/or r_inv will now also require deletion of these lemmas.
  4083 
  4084 * Renamed the following theorems, INCOMPATIBILITY:
  4085 
  4086 UpperD ~> Upper_memD
  4087 LowerD ~> Lower_memD
  4088 least_carrier ~> least_closed
  4089 greatest_carrier ~> greatest_closed
  4090 greatest_Lower_above ~> greatest_Lower_below
  4091 one_zero ~> carrier_one_zero
  4092 one_not_zero ~> carrier_one_not_zero  (collision with assumption)
  4093 
  4094 
  4095 *** HOL-Nominal ***
  4096 
  4097 * Nominal datatypes can now contain type-variables.
  4098 
  4099 * Commands 'nominal_inductive' and 'equivariance' work with local
  4100 theory targets.
  4101 
  4102 * Nominal primrec can now works with local theory targets and its
  4103 specification syntax now conforms to the general format as seen in
  4104 'inductive' etc.
  4105 
  4106 * Method "perm_simp" honours the standard simplifier attributes
  4107 (no_asm), (no_asm_use) etc.
  4108 
  4109 * The new predicate #* is defined like freshness, except that on the
  4110 left hand side can be a set or list of atoms.
  4111 
  4112 * Experimental command 'nominal_inductive2' derives strong induction
  4113 principles for inductive definitions.  In contrast to
  4114 'nominal_inductive', which can only deal with a fixed number of
  4115 binders, it can deal with arbitrary expressions standing for sets of
  4116 atoms to be avoided.  The only inductive definition we have at the
  4117 moment that needs this generalisation is the typing rule for Lets in
  4118 the algorithm W:
  4119 
  4120  Gamma |- t1 : T1   (x,close Gamma T1)::Gamma |- t2 : T2   x#Gamma
  4121  -----------------------------------------------------------------
  4122          Gamma |- Let x be t1 in t2 : T2
  4123 
  4124 In this rule one wants to avoid all the binders that are introduced by
  4125 "close Gamma T1".  We are looking for other examples where this
  4126 feature might be useful.  Please let us know.
  4127 
  4128 
  4129 *** HOLCF ***
  4130 
  4131 * Reimplemented the simplification procedure for proving continuity
  4132 subgoals.  The new simproc is extensible; users can declare additional
  4133 continuity introduction rules with the attribute [cont2cont].
  4134 
  4135 * The continuity simproc now uses a different introduction rule for
  4136 solving continuity subgoals on terms with lambda abstractions.  In
  4137 some rare cases the new simproc may fail to solve subgoals that the
  4138 old one could solve, and "simp add: cont2cont_LAM" may be necessary.
  4139 Potential INCOMPATIBILITY.
  4140 
  4141 * Command 'fixrec': specification syntax now conforms to the general
  4142 format as seen in 'inductive' etc.  See src/HOLCF/ex/Fixrec_ex.thy for
  4143 examples.  INCOMPATIBILITY.
  4144 
  4145 
  4146 *** ZF ***
  4147 
  4148 * Proof of Zorn's Lemma for partial orders.
  4149 
  4150 
  4151 *** ML ***
  4152 
  4153 * Multithreading for Poly/ML 5.1/5.2 is no longer supported, only for
  4154 Poly/ML 5.2.1 or later.  Important note: the TimeLimit facility
  4155 depends on multithreading, so timouts will not work before Poly/ML
  4156 5.2.1!
  4157 
  4158 * High-level support for concurrent ML programming, see
  4159 src/Pure/Cuncurrent.  The data-oriented model of "future values" is
  4160 particularly convenient to organize independent functional
  4161 computations.  The concept of "synchronized variables" provides a
  4162 higher-order interface for components with shared state, avoiding the
  4163 delicate details of mutexes and condition variables.  (Requires
  4164 Poly/ML 5.2.1 or later.)
  4165 
  4166 * ML bindings produced via Isar commands are stored within the Isar
  4167 context (theory or proof).  Consequently, commands like 'use' and 'ML'
  4168 become thread-safe and work with undo as expected (concerning
  4169 top-level bindings, not side-effects on global references).
  4170 INCOMPATIBILITY, need to provide proper Isar context when invoking the
  4171 compiler at runtime; really global bindings need to be given outside a
  4172 theory.  (Requires Poly/ML 5.2 or later.)
  4173 
  4174 * Command 'ML_prf' is analogous to 'ML' but works within a proof
  4175 context.  Top-level ML bindings are stored within the proof context in
  4176 a purely sequential fashion, disregarding the nested proof structure.
  4177 ML bindings introduced by 'ML_prf' are discarded at the end of the
  4178 proof.  (Requires Poly/ML 5.2 or later.)
  4179 
  4180 * Simplified ML attribute and method setup, cf. functions Attrib.setup
  4181 and Method.setup, as well as Isar commands 'attribute_setup' and
  4182 'method_setup'.  INCOMPATIBILITY for 'method_setup', need to simplify
  4183 existing code accordingly, or use plain 'setup' together with old
  4184 Method.add_method.
  4185 
  4186 * Simplified ML oracle interface Thm.add_oracle promotes 'a -> cterm
  4187 to 'a -> thm, while results are always tagged with an authentic oracle
  4188 name.  The Isar command 'oracle' is now polymorphic, no argument type
  4189 is specified.  INCOMPATIBILITY, need to simplify existing oracle code
  4190 accordingly.  Note that extra performance may be gained by producing
  4191 the cterm carefully, avoiding slow Thm.cterm_of.
  4192 
  4193 * Simplified interface for defining document antiquotations via
  4194 ThyOutput.antiquotation, ThyOutput.output, and optionally
  4195 ThyOutput.maybe_pretty_source.  INCOMPATIBILITY, need to simplify user
  4196 antiquotations accordingly, see src/Pure/Thy/thy_output.ML for common
  4197 examples.
  4198 
  4199 * More systematic treatment of long names, abstract name bindings, and
  4200 name space operations.  Basic operations on qualified names have been
  4201 move from structure NameSpace to Long_Name, e.g. Long_Name.base_name,
  4202 Long_Name.append.  Old type bstring has been mostly replaced by
  4203 abstract type binding (see structure Binding), which supports precise
  4204 qualification by packages and local theory targets, as well as proper
  4205 tracking of source positions.  INCOMPATIBILITY, need to wrap old
  4206 bstring values into Binding.name, or better pass through abstract
  4207 bindings everywhere.  See further src/Pure/General/long_name.ML,
  4208 src/Pure/General/binding.ML and src/Pure/General/name_space.ML
  4209 
  4210 * Result facts (from PureThy.note_thms, ProofContext.note_thms,
  4211 LocalTheory.note etc.) now refer to the *full* internal name, not the
  4212 bstring as before.  INCOMPATIBILITY, not detected by ML type-checking!
  4213 
  4214 * Disposed old type and term read functions (Sign.read_def_typ,
  4215 Sign.read_typ, Sign.read_def_terms, Sign.read_term,
  4216 Thm.read_def_cterms, Thm.read_cterm etc.).  INCOMPATIBILITY, should
  4217 use regular Syntax.read_typ, Syntax.read_term, Syntax.read_typ_global,
  4218 Syntax.read_term_global etc.; see also OldGoals.read_term as last
  4219 resort for legacy applications.
  4220 
  4221 * Disposed old declarations, tactics, tactic combinators that refer to
  4222 the simpset or claset of an implicit theory (such as Addsimps,
  4223 Simp_tac, SIMPSET).  INCOMPATIBILITY, should use @{simpset} etc. in
  4224 embedded ML text, or local_simpset_of with a proper context passed as
  4225 explicit runtime argument.
  4226 
  4227 * Rules and tactics that read instantiations (read_instantiate,
  4228 res_inst_tac, thin_tac, subgoal_tac etc.) now demand a proper proof
  4229 context, which is required for parsing and type-checking.  Moreover,
  4230 the variables are specified as plain indexnames, not string encodings
  4231 thereof.  INCOMPATIBILITY.
  4232 
  4233 * Generic Toplevel.add_hook interface allows to analyze the result of
  4234 transactions.  E.g. see src/Pure/ProofGeneral/proof_general_pgip.ML
  4235 for theorem dependency output of transactions resulting in a new
  4236 theory state.
  4237 
  4238 * ML antiquotations: block-structured compilation context indicated by
  4239 \<lbrace> ... \<rbrace>; additional antiquotation forms:
  4240 
  4241   @{binding name}                         - basic name binding
  4242   @{let ?pat = term}                      - term abbreviation (HO matching)
  4243   @{note name = fact}                     - fact abbreviation
  4244   @{thm fact}                             - singleton fact (with attributes)
  4245   @{thms fact}                            - general fact (with attributes)
  4246   @{lemma prop by method}                 - singleton goal
  4247   @{lemma prop by meth1 meth2}            - singleton goal
  4248   @{lemma prop1 ... propN by method}      - general goal
  4249   @{lemma prop1 ... propN by meth1 meth2} - general goal
  4250   @{lemma (open) ...}                     - open derivation
  4251 
  4252 
  4253 *** System ***
  4254 
  4255 * The Isabelle "emacs" tool provides a specific interface to invoke
  4256 Proof General / Emacs, with more explicit failure if that is not
  4257 installed (the old isabelle-interface script silently falls back on
  4258 isabelle-process).  The PROOFGENERAL_HOME setting determines the
  4259 installation location of the Proof General distribution.
  4260 
  4261 * Isabelle/lib/classes/Pure.jar provides basic support to integrate
  4262 the Isabelle process into a JVM/Scala application.  See
  4263 Isabelle/lib/jedit/plugin for a minimal example.  (The obsolete Java
  4264 process wrapper has been discontinued.)
  4265 
  4266 * Added homegrown Isabelle font with unicode layout, see lib/fonts.
  4267 
  4268 * Various status messages (with exact source position information) are
  4269 emitted, if proper markup print mode is enabled.  This allows
  4270 user-interface components to provide detailed feedback on internal
  4271 prover operations.
  4272 
  4273 
  4274 
  4275 New in Isabelle2008 (June 2008)
  4276 -------------------------------
  4277 
  4278 *** General ***
  4279 
  4280 * The Isabelle/Isar Reference Manual (isar-ref) has been reorganized
  4281 and updated, with formally checked references as hyperlinks.
  4282 
  4283 * Theory loader: use_thy (and similar operations) no longer set the
  4284 implicit ML context, which was occasionally hard to predict and in
  4285 conflict with concurrency.  INCOMPATIBILITY, use ML within Isar which
  4286 provides a proper context already.
  4287 
  4288 * Theory loader: old-style ML proof scripts being *attached* to a thy
  4289 file are no longer supported.  INCOMPATIBILITY, regular 'uses' and
  4290 'use' within a theory file will do the job.
  4291 
  4292 * Name space merge now observes canonical order, i.e. the second space
  4293 is inserted into the first one, while existing entries in the first
  4294 space take precedence.  INCOMPATIBILITY in rare situations, may try to
  4295 swap theory imports.
  4296 
  4297 * Syntax: symbol \<chi> is now considered a letter.  Potential
  4298 INCOMPATIBILITY in identifier syntax etc.
  4299 
  4300 * Outer syntax: string tokens no longer admit escaped white space,
  4301 which was an accidental (undocumented) feature.  INCOMPATIBILITY, use
  4302 white space without escapes.
  4303 
  4304 * Outer syntax: string tokens may contain arbitrary character codes
  4305 specified via 3 decimal digits (as in SML).  E.g. "foo\095bar" for
  4306 "foo_bar".
  4307 
  4308 
  4309 *** Pure ***
  4310 
  4311 * Context-dependent token translations.  Default setup reverts locally
  4312 fixed variables, and adds hilite markup for undeclared frees.
  4313 
  4314 * Unused theorems can be found using the new command 'unused_thms'.
  4315 There are three ways of invoking it:
  4316 
  4317 (1) unused_thms
  4318      Only finds unused theorems in the current theory.
  4319 
  4320 (2) unused_thms thy_1 ... thy_n -
  4321      Finds unused theorems in the current theory and all of its ancestors,
  4322      excluding the theories thy_1 ... thy_n and all of their ancestors.
  4323 
  4324 (3) unused_thms thy_1 ... thy_n - thy'_1 ... thy'_m
  4325      Finds unused theorems in the theories thy'_1 ... thy'_m and all of
  4326      their ancestors, excluding the theories thy_1 ... thy_n and all of
  4327      their ancestors.
  4328 
  4329 In order to increase the readability of the list produced by
  4330 unused_thms, theorems that have been created by a particular instance
  4331 of a theory command such as 'inductive' or 'function' are considered
  4332 to belong to the same "group", meaning that if at least one theorem in
  4333 this group is used, the other theorems in the same group are no longer
  4334 reported as unused.  Moreover, if all theorems in the group are
  4335 unused, only one theorem in the group is displayed.
  4336 
  4337 Note that proof objects have to be switched on in order for
  4338 unused_thms to work properly (i.e. !proofs must be >= 1, which is
  4339 usually the case when using Proof General with the default settings).
  4340 
  4341 * Authentic naming of facts disallows ad-hoc overwriting of previous
  4342 theorems within the same name space.  INCOMPATIBILITY, need to remove
  4343 duplicate fact bindings, or even accidental fact duplications.  Note
  4344 that tools may maintain dynamically scoped facts systematically, using
  4345 PureThy.add_thms_dynamic.
  4346 
  4347 * Command 'hide' now allows to hide from "fact" name space as well.
  4348 
  4349 * Eliminated destructive theorem database, simpset, claset, and
  4350 clasimpset.  Potential INCOMPATIBILITY, really need to observe linear
  4351 update of theories within ML code.
  4352 
  4353 * Eliminated theory ProtoPure and CPure, leaving just one Pure theory.
  4354 INCOMPATIBILITY, object-logics depending on former Pure require
  4355 additional setup PureThy.old_appl_syntax_setup; object-logics
  4356 depending on former CPure need to refer to Pure.
  4357 
  4358 * Commands 'use' and 'ML' are now purely functional, operating on
  4359 theory/local_theory.  Removed former 'ML_setup' (on theory), use 'ML'
  4360 instead.  Added 'ML_val' as mere diagnostic replacement for 'ML'.
  4361 INCOMPATIBILITY.
  4362 
  4363 * Command 'setup': discontinued implicit version with ML reference.
  4364 
  4365 * Instantiation target allows for simultaneous specification of class
  4366 instance operations together with an instantiation proof.
  4367 Type-checking phase allows to refer to class operations uniformly.
  4368 See src/HOL/Complex/Complex.thy for an Isar example and
  4369 src/HOL/Library/Eval.thy for an ML example.
  4370 
  4371 * Indexing of literal facts: be more serious about including only
  4372 facts from the visible specification/proof context, but not the
  4373 background context (locale etc.).  Affects `prop` notation and method
  4374 "fact".  INCOMPATIBILITY: need to name facts explicitly in rare
  4375 situations.
  4376 
  4377 * Method "cases", "induct", "coinduct": removed obsolete/undocumented
  4378 "(open)" option, which used to expose internal bound variables to the
  4379 proof text.
  4380 
  4381 * Isar statements: removed obsolete case "rule_context".
  4382 INCOMPATIBILITY, better use explicit fixes/assumes.
  4383 
  4384 * Locale proofs: default proof step now includes 'unfold_locales';
  4385 hence 'proof' without argument may be used to unfold locale
  4386 predicates.
  4387 
  4388 
  4389 *** Document preparation ***
  4390 
  4391 * Simplified pdfsetup.sty: color/hyperref is used unconditionally for
  4392 both pdf and dvi (hyperlinks usually work in xdvi as well); removed
  4393 obsolete thumbpdf setup (contemporary PDF viewers do this on the
  4394 spot); renamed link color from "darkblue" to "linkcolor" (default
  4395 value unchanged, can be redefined via \definecolor); no longer sets
  4396 "a4paper" option (unnecessary or even intrusive).
  4397 
  4398 * Antiquotation @{lemma A method} proves proposition A by the given
  4399 method (either a method name or a method name plus (optional) method
  4400 arguments in parentheses) and prints A just like @{prop A}.
  4401 
  4402 
  4403 *** HOL ***
  4404 
  4405 * New primrec package.  Specification syntax conforms in style to
  4406 definition/function/....  No separate induction rule is provided.  The
  4407 "primrec" command distinguishes old-style and new-style specifications
  4408 by syntax.  The former primrec package is now named OldPrimrecPackage.
  4409 When adjusting theories, beware: constants stemming from new-style
  4410 primrec specifications have authentic syntax.
  4411 
  4412 * Metis prover is now an order of magnitude faster, and also works
  4413 with multithreading.
  4414 
  4415 * Metis: the maximum number of clauses that can be produced from a
  4416 theorem is now given by the attribute max_clauses.  Theorems that
  4417 exceed this number are ignored, with a warning printed.
  4418 
  4419 * Sledgehammer no longer produces structured proofs by default. To
  4420 enable, declare [[sledgehammer_full = true]].  Attributes
  4421 reconstruction_modulus, reconstruction_sorts renamed
  4422 sledgehammer_modulus, sledgehammer_sorts.  INCOMPATIBILITY.
  4423 
  4424 * Method "induct_scheme" derives user-specified induction rules
  4425 from well-founded induction and completeness of patterns. This factors
  4426 out some operations that are done internally by the function package
  4427 and makes them available separately.  See
  4428 src/HOL/ex/Induction_Scheme.thy for examples.
  4429 
  4430 * More flexible generation of measure functions for termination
  4431 proofs: Measure functions can be declared by proving a rule of the
  4432 form "is_measure f" and giving it the [measure_function] attribute.
  4433 The "is_measure" predicate is logically meaningless (always true), and
  4434 just guides the heuristic.  To find suitable measure functions, the
  4435 termination prover sets up the goal "is_measure ?f" of the appropriate
  4436 type and generates all solutions by prolog-style backwards proof using
  4437 the declared rules.
  4438 
  4439 This setup also deals with rules like 
  4440 
  4441   "is_measure f ==> is_measure (list_size f)"
  4442 
  4443 which accommodates nested datatypes that recurse through lists.
  4444 Similar rules are predeclared for products and option types.
  4445 
  4446 * Turned the type of sets "'a set" into an abbreviation for "'a => bool"
  4447 
  4448   INCOMPATIBILITIES:
  4449 
  4450   - Definitions of overloaded constants on sets have to be replaced by
  4451     definitions on => and bool.
  4452 
  4453   - Some definitions of overloaded operators on sets can now be proved
  4454     using the definitions of the operators on => and bool.  Therefore,
  4455     the following theorems have been renamed:
  4456 
  4457       subset_def   -> subset_eq
  4458       psubset_def  -> psubset_eq
  4459       set_diff_def -> set_diff_eq
  4460       Compl_def    -> Compl_eq
  4461       Sup_set_def  -> Sup_set_eq
  4462       Inf_set_def  -> Inf_set_eq
  4463       sup_set_def  -> sup_set_eq
  4464       inf_set_def  -> inf_set_eq
  4465 
  4466   - Due to the incompleteness of the HO unification algorithm, some
  4467     rules such as subst may require manual instantiation, if some of
  4468     the unknowns in the rule is a set.
  4469 
  4470   - Higher order unification and forward proofs:
  4471     The proof pattern
  4472 
  4473       have "P (S::'a set)" <...>
  4474       then have "EX S. P S" ..
  4475 
  4476     no longer works (due to the incompleteness of the HO unification
  4477     algorithm) and must be replaced by the pattern
  4478 
  4479       have "EX S. P S"
  4480       proof
  4481         show "P S" <...>
  4482       qed
  4483 
  4484   - Calculational reasoning with subst (or similar rules):
  4485     The proof pattern
  4486 
  4487       have "P (S::'a set)" <...>
  4488       also have "S = T" <...>
  4489       finally have "P T" .
  4490 
  4491     no longer works (for similar reasons as the previous example) and
  4492     must be replaced by something like
  4493 
  4494       have "P (S::'a set)" <...>
  4495       moreover have "S = T" <...>
  4496       ultimately have "P T" by simp
  4497 
  4498   - Tactics or packages written in ML code:
  4499     Code performing pattern matching on types via
  4500 
  4501       Type ("set", [T]) => ...
  4502 
  4503     must be rewritten. Moreover, functions like strip_type or
  4504     binder_types no longer return the right value when applied to a
  4505     type of the form
  4506 
  4507       T1 => ... => Tn => U => bool
  4508 
  4509     rather than
  4510 
  4511       T1 => ... => Tn => U set
  4512 
  4513 * Merged theories Wellfounded_Recursion, Accessible_Part and
  4514 Wellfounded_Relations to theory Wellfounded.
  4515 
  4516 * Explicit class "eq" for executable equality.  INCOMPATIBILITY.
  4517 
  4518 * Class finite no longer treats UNIV as class parameter.  Use class
  4519 enum from theory Library/Enum instead to achieve a similar effect.
  4520 INCOMPATIBILITY.
  4521 
  4522 * Theory List: rule list_induct2 now has explicitly named cases "Nil"
  4523 and "Cons".  INCOMPATIBILITY.
  4524 
  4525 * HOL (and FOL): renamed variables in rules imp_elim and swap.
  4526 Potential INCOMPATIBILITY.
  4527 
  4528 * Theory Product_Type: duplicated lemmas split_Pair_apply and
  4529 injective_fst_snd removed, use split_eta and prod_eqI instead.
  4530 Renamed upd_fst to apfst and upd_snd to apsnd.  INCOMPATIBILITY.
  4531 
  4532 * Theory Nat: removed redundant lemmas that merely duplicate lemmas of
  4533 the same name in theory Orderings:
  4534 
  4535   less_trans
  4536   less_linear
  4537   le_imp_less_or_eq
  4538   le_less_trans
  4539   less_le_trans
  4540   less_not_sym
  4541   less_asym
  4542 
  4543 Renamed less_imp_le to less_imp_le_nat, and less_irrefl to
  4544 less_irrefl_nat.  Potential INCOMPATIBILITY due to more general types
  4545 and different variable names.
  4546 
  4547 * Library/Option_ord.thy: Canonical order on option type.
  4548 
  4549 * Library/RBT.thy: Red-black trees, an efficient implementation of
  4550 finite maps.
  4551 
  4552 * Library/Countable.thy: Type class for countable types.
  4553 
  4554 * Theory Int: The representation of numerals has changed.  The infix
  4555 operator BIT and the bit datatype with constructors B0 and B1 have
  4556 disappeared.  INCOMPATIBILITY, use "Int.Bit0 x" and "Int.Bit1 y" in
  4557 place of "x BIT bit.B0" and "y BIT bit.B1", respectively.  Theorems
  4558 involving BIT, B0, or B1 have been renamed with "Bit0" or "Bit1"
  4559 accordingly.
  4560 
  4561 * Theory Nat: definition of <= and < on natural numbers no longer
  4562 depend on well-founded relations.  INCOMPATIBILITY.  Definitions
  4563 le_def and less_def have disappeared.  Consider lemmas not_less
  4564 [symmetric, where ?'a = nat] and less_eq [symmetric] instead.
  4565 
  4566 * Theory Finite_Set: locales ACf, ACe, ACIf, ACIfSL and ACIfSLlin
  4567 (whose purpose mainly is for various fold_set functionals) have been
  4568 abandoned in favor of the existing algebraic classes
  4569 ab_semigroup_mult, comm_monoid_mult, ab_semigroup_idem_mult,
  4570 lower_semilattice (resp. upper_semilattice) and linorder.
  4571 INCOMPATIBILITY.
  4572 
  4573 * Theory Transitive_Closure: induct and cases rules now declare proper
  4574 case_names ("base" and "step").  INCOMPATIBILITY.
  4575 
  4576 * Theorem Inductive.lfp_ordinal_induct generalized to complete
  4577 lattices.  The form set-specific version is available as
  4578 Inductive.lfp_ordinal_induct_set.
  4579 
  4580 * Renamed theorems "power.simps" to "power_int.simps".
  4581 INCOMPATIBILITY.
  4582 
  4583 * Class semiring_div provides basic abstract properties of semirings
  4584 with division and modulo operations.  Subsumes former class dvd_mod.
  4585 
  4586 * Merged theories IntDef, Numeral and IntArith into unified theory
  4587 Int.  INCOMPATIBILITY.
  4588 
  4589 * Theory Library/Code_Index: type "index" now represents natural
  4590 numbers rather than integers.  INCOMPATIBILITY.
  4591 
  4592 * New class "uminus" with operation "uminus" (split of from class
  4593 "minus" which now only has operation "minus", binary).
  4594 INCOMPATIBILITY.
  4595 
  4596 * Constants "card", "internal_split", "option_map" now with authentic
  4597 syntax.  INCOMPATIBILITY.
  4598 
  4599 * Definitions subset_def, psubset_def, set_diff_def, Compl_def,
  4600 le_bool_def, less_bool_def, le_fun_def, less_fun_def, inf_bool_def,
  4601 sup_bool_def, Inf_bool_def, Sup_bool_def, inf_fun_def, sup_fun_def,
  4602 Inf_fun_def, Sup_fun_def, inf_set_def, sup_set_def, Inf_set_def,
  4603 Sup_set_def, le_def, less_def, option_map_def now with object
  4604 equality.  INCOMPATIBILITY.
  4605 
  4606 * Records. Removed K_record, and replaced it by pure lambda term
  4607 %x. c. The simplifier setup is now more robust against eta expansion.
  4608 INCOMPATIBILITY: in cases explicitly referring to K_record.
  4609 
  4610 * Library/Multiset: {#a, b, c#} abbreviates {#a#} + {#b#} + {#c#}.
  4611 
  4612 * Library/ListVector: new theory of arithmetic vector operations.
  4613 
  4614 * Library/Order_Relation: new theory of various orderings as sets of
  4615 pairs.  Defines preorders, partial orders, linear orders and
  4616 well-orders on sets and on types.
  4617 
  4618 
  4619 *** ZF ***
  4620 
  4621 * Renamed some theories to allow to loading both ZF and HOL in the
  4622 same session:
  4623 
  4624   Datatype  -> Datatype_ZF
  4625   Inductive -> Inductive_ZF
  4626   Int       -> Int_ZF
  4627   IntDiv    -> IntDiv_ZF
  4628   Nat       -> Nat_ZF
  4629   List      -> List_ZF
  4630   Main      -> Main_ZF
  4631 
  4632 INCOMPATIBILITY: ZF theories that import individual theories below
  4633 Main might need to be adapted.  Regular theory Main is still
  4634 available, as trivial extension of Main_ZF.
  4635 
  4636 
  4637 *** ML ***
  4638 
  4639 * ML within Isar: antiquotation @{const name} or @{const
  4640 name(typargs)} produces statically-checked Const term.
  4641 
  4642 * Functor NamedThmsFun: data is available to the user as dynamic fact
  4643 (of the same name).  Removed obsolete print command.
  4644 
  4645 * Removed obsolete "use_legacy_bindings" function.
  4646 
  4647 * The ``print mode'' is now a thread-local value derived from a global
  4648 template (the former print_mode reference), thus access becomes
  4649 non-critical.  The global print_mode reference is for session
  4650 management only; user-code should use print_mode_value,
  4651 print_mode_active, PrintMode.setmp etc.  INCOMPATIBILITY.
  4652 
  4653 * Functions system/system_out provide a robust way to invoke external
  4654 shell commands, with propagation of interrupts (requires Poly/ML
  4655 5.2.1).  Do not use OS.Process.system etc. from the basis library!
  4656 
  4657 
  4658 *** System ***
  4659 
  4660 * Default settings: PROOFGENERAL_OPTIONS no longer impose xemacs ---
  4661 in accordance with Proof General 3.7, which prefers GNU emacs.
  4662 
  4663 * isatool tty runs Isabelle process with plain tty interaction;
  4664 optional line editor may be specified via ISABELLE_LINE_EDITOR
  4665 setting, the default settings attempt to locate "ledit" and "rlwrap".
  4666 
  4667 * isatool browser now works with Cygwin as well, using general
  4668 "javapath" function defined in Isabelle process environment.
  4669 
  4670 * YXML notation provides a simple and efficient alternative to
  4671 standard XML transfer syntax.  See src/Pure/General/yxml.ML and
  4672 isatool yxml as described in the Isabelle system manual.
  4673 
  4674 * JVM class isabelle.IsabelleProcess (located in Isabelle/lib/classes)
  4675 provides general wrapper for managing an Isabelle process in a robust
  4676 fashion, with ``cooked'' output from stdin/stderr.
  4677 
  4678 * Rudimentary Isabelle plugin for jEdit (see Isabelle/lib/jedit),
  4679 based on Isabelle/JVM process wrapper (see Isabelle/lib/classes).
  4680 
  4681 * Removed obsolete THIS_IS_ISABELLE_BUILD feature.  NB: the documented
  4682 way of changing the user's settings is via
  4683 ISABELLE_HOME_USER/etc/settings, which is a fully featured bash
  4684 script.
  4685 
  4686 * Multithreading.max_threads := 0 refers to the number of actual CPU
  4687 cores of the underlying machine, which is a good starting point for
  4688 optimal performance tuning.  The corresponding usedir option -M allows
  4689 "max" as an alias for "0".  WARNING: does not work on certain versions
  4690 of Mac OS (with Poly/ML 5.1).
  4691 
  4692 * isabelle-process: non-ML sessions are run with "nice", to reduce the
  4693 adverse effect of Isabelle flooding interactive front-ends (notably
  4694 ProofGeneral / XEmacs).
  4695 
  4696 
  4697 
  4698 New in Isabelle2007 (November 2007)
  4699 -----------------------------------
  4700 
  4701 *** General ***
  4702 
  4703 * More uniform information about legacy features, notably a
  4704 warning/error of "Legacy feature: ...", depending on the state of the
  4705 tolerate_legacy_features flag (default true). FUTURE INCOMPATIBILITY:
  4706 legacy features will disappear eventually.
  4707 
  4708 * Theory syntax: the header format ``theory A = B + C:'' has been
  4709 discontinued in favour of ``theory A imports B C begin''.  Use isatool
  4710 fixheaders to convert existing theory files.  INCOMPATIBILITY.
  4711 
  4712 * Theory syntax: the old non-Isar theory file format has been
  4713 discontinued altogether.  Note that ML proof scripts may still be used
  4714 with Isar theories; migration is usually quite simple with the ML
  4715 function use_legacy_bindings.  INCOMPATIBILITY.
  4716 
  4717 * Theory syntax: some popular names (e.g. 'class', 'declaration',
  4718 'fun', 'help', 'if') are now keywords.  INCOMPATIBILITY, use double
  4719 quotes.
  4720 
  4721 * Theory loader: be more serious about observing the static theory
  4722 header specifications (including optional directories), but not the
  4723 accidental file locations of previously successful loads.  The strict
  4724 update policy of former update_thy is now already performed by
  4725 use_thy, so the former has been removed; use_thys updates several
  4726 theories simultaneously, just as 'imports' within a theory header
  4727 specification, but without merging the results.  Potential
  4728 INCOMPATIBILITY: may need to refine theory headers and commands
  4729 ROOT.ML which depend on load order.
  4730 
  4731 * Theory loader: optional support for content-based file
  4732 identification, instead of the traditional scheme of full physical
  4733 path plus date stamp; configured by the ISABELLE_FILE_IDENT setting
  4734 (cf. the system manual).  The new scheme allows to work with
  4735 non-finished theories in persistent session images, such that source
  4736 files may be moved later on without requiring reloads.
  4737 
  4738 * Theory loader: old-style ML proof scripts being *attached* to a thy
  4739 file (with the same base name as the theory) are considered a legacy
  4740 feature, which will disappear eventually. Even now, the theory loader
  4741 no longer maintains dependencies on such files.
  4742 
  4743 * Syntax: the scope for resolving ambiguities via type-inference is
  4744 now limited to individual terms, instead of whole simultaneous
  4745 specifications as before. This greatly reduces the complexity of the
  4746 syntax module and improves flexibility by separating parsing and
  4747 type-checking. INCOMPATIBILITY: additional type-constraints (explicit
  4748 'fixes' etc.) are required in rare situations.
  4749 
  4750 * Syntax: constants introduced by new-style packages ('definition',
  4751 'abbreviation' etc.) are passed through the syntax module in
  4752 ``authentic mode''. This means that associated mixfix annotations
  4753 really stick to such constants, independently of potential name space
  4754 ambiguities introduced later on. INCOMPATIBILITY: constants in parse
  4755 trees are represented slightly differently, may need to adapt syntax
  4756 translations accordingly. Use CONST marker in 'translations' and
  4757 @{const_syntax} antiquotation in 'parse_translation' etc.
  4758 
  4759 * Legacy goal package: reduced interface to the bare minimum required
  4760 to keep existing proof scripts running.  Most other user-level
  4761 functions are now part of the OldGoals structure, which is *not* open
  4762 by default (consider isatool expandshort before open OldGoals).
  4763 Removed top_sg, prin, printyp, pprint_term/typ altogether, because
  4764 these tend to cause confusion about the actual goal (!) context being
  4765 used here, which is not necessarily the same as the_context().
  4766 
  4767 * Command 'find_theorems': supports "*" wild-card in "name:"
  4768 criterion; "with_dups" option.  Certain ProofGeneral versions might
  4769 support a specific search form (see ProofGeneral/CHANGES).
  4770 
  4771 * The ``prems limit'' option (cf. ProofContext.prems_limit) is now -1
  4772 by default, which means that "prems" (and also "fixed variables") are
  4773 suppressed from proof state output.  Note that the ProofGeneral
  4774 settings mechanism allows to change and save options persistently, but
  4775 older versions of Isabelle will fail to start up if a negative prems
  4776 limit is imposed.
  4777 
  4778 * Local theory targets may be specified by non-nested blocks of
  4779 ``context/locale/class ... begin'' followed by ``end''.  The body may
  4780 contain definitions, theorems etc., including any derived mechanism
  4781 that has been implemented on top of these primitives.  This concept
  4782 generalizes the existing ``theorem (in ...)'' towards more versatility
  4783 and scalability.
  4784 
  4785 * Proof General interface: proper undo of final 'end' command;
  4786 discontinued Isabelle/classic mode (ML proof scripts).
  4787 
  4788 
  4789 *** Document preparation ***
  4790 
  4791 * Added antiquotation @{theory name} which prints the given name,
  4792 after checking that it refers to a valid ancestor theory in the
  4793 current context.
  4794 
  4795 * Added antiquotations @{ML_type text} and @{ML_struct text} which
  4796 check the given source text as ML type/structure, printing verbatim.
  4797 
  4798 * Added antiquotation @{abbrev "c args"} which prints the abbreviation
  4799 "c args == rhs" given in the current context.  (Any number of
  4800 arguments may be given on the LHS.)
  4801 
  4802 
  4803 *** Pure ***
  4804 
  4805 * The 'class' package offers a combination of axclass and locale to
  4806 achieve Haskell-like type classes in Isabelle.  Definitions and
  4807 theorems within a class context produce both relative results (with
  4808 implicit parameters according to the locale context), and polymorphic
  4809 constants with qualified polymorphism (according to the class
  4810 context).  Within the body context of a 'class' target, a separate
  4811 syntax layer ("user space type system") takes care of converting
  4812 between global polymorphic consts and internal locale representation.
  4813 See src/HOL/ex/Classpackage.thy for examples (as well as main HOL).
  4814 "isatool doc classes" provides a tutorial.
  4815 
  4816 * Generic code generator framework allows to generate executable
  4817 code for ML and Haskell (including Isabelle classes).  A short usage
  4818 sketch:
  4819 
  4820     internal compilation:
  4821         export_code <list of constants (term syntax)> in SML
  4822     writing SML code to a file:
  4823         export_code <list of constants (term syntax)> in SML <filename>
  4824     writing OCaml code to a file:
  4825         export_code <list of constants (term syntax)> in OCaml <filename>
  4826     writing Haskell code to a bunch of files:
  4827         export_code <list of constants (term syntax)> in Haskell <filename>
  4828 
  4829     evaluating closed propositions to True/False using code generation:
  4830         method ``eval''
  4831 
  4832 Reasonable default setup of framework in HOL.
  4833 
  4834 Theorem attributs for selecting and transforming function equations theorems:
  4835 
  4836     [code fun]:        select a theorem as function equation for a specific constant
  4837     [code fun del]:    deselect a theorem as function equation for a specific constant
  4838     [code inline]:     select an equation theorem for unfolding (inlining) in place
  4839     [code inline del]: deselect an equation theorem for unfolding (inlining) in place
  4840 
  4841 User-defined serializations (target in {SML, OCaml, Haskell}):
  4842 
  4843     code_const <and-list of constants (term syntax)>
  4844       {(target) <and-list of const target syntax>}+
  4845 
  4846     code_type <and-list of type constructors>
  4847       {(target) <and-list of type target syntax>}+
  4848 
  4849     code_instance <and-list of instances>
  4850       {(target)}+
  4851         where instance ::= <type constructor> :: <class>
  4852 
  4853     code_class <and_list of classes>
  4854       {(target) <and-list of class target syntax>}+
  4855         where class target syntax ::= <class name> {where {<classop> == <target syntax>}+}?
  4856 
  4857 code_instance and code_class only are effective to target Haskell.
  4858 
  4859 For example usage see src/HOL/ex/Codegenerator.thy and
  4860 src/HOL/ex/Codegenerator_Pretty.thy.  A separate tutorial on code
  4861 generation from Isabelle/HOL theories is available via "isatool doc
  4862 codegen".
  4863 
  4864 * Code generator: consts in 'consts_code' Isar commands are now
  4865 referred to by usual term syntax (including optional type
  4866 annotations).
  4867 
  4868 * Command 'no_translations' removes translation rules from theory
  4869 syntax.
  4870 
  4871 * Overloaded definitions are now actually checked for acyclic
  4872 dependencies.  The overloading scheme is slightly more general than
  4873 that of Haskell98, although Isabelle does not demand an exact
  4874 correspondence to type class and instance declarations.
  4875 INCOMPATIBILITY, use ``defs (unchecked overloaded)'' to admit more
  4876 exotic versions of overloading -- at the discretion of the user!
  4877 
  4878 Polymorphic constants are represented via type arguments, i.e. the
  4879 instantiation that matches an instance against the most general
  4880 declaration given in the signature.  For example, with the declaration
  4881 c :: 'a => 'a => 'a, an instance c :: nat => nat => nat is represented
  4882 as c(nat).  Overloading is essentially simultaneous structural
  4883 recursion over such type arguments.  Incomplete specification patterns
  4884 impose global constraints on all occurrences, e.g. c('a * 'a) on the
  4885 LHS means that more general c('a * 'b) will be disallowed on any RHS.
  4886 Command 'print_theory' outputs the normalized system of recursive
  4887 equations, see section "definitions".
  4888 
  4889 * Configuration options are maintained within the theory or proof
  4890 context (with name and type bool/int/string), providing a very simple
  4891 interface to a poor-man's version of general context data.  Tools may
  4892 declare options in ML (e.g. using Attrib.config_int) and then refer to
  4893 these values using Config.get etc.  Users may change options via an
  4894 associated attribute of the same name.  This form of context
  4895 declaration works particularly well with commands 'declare' or
  4896 'using', for example ``declare [[foo = 42]]''.  Thus it has become
  4897 very easy to avoid global references, which would not observe Isar
  4898 toplevel undo/redo and fail to work with multithreading.
  4899 
  4900 Various global ML references of Pure and HOL have been turned into
  4901 configuration options:
  4902 
  4903   Unify.search_bound		unify_search_bound
  4904   Unify.trace_bound		unify_trace_bound
  4905   Unify.trace_simp		unify_trace_simp
  4906   Unify.trace_types		unify_trace_types
  4907   Simplifier.simp_depth_limit	simp_depth_limit
  4908   Blast.depth_limit		blast_depth_limit
  4909   DatatypeProp.dtK		datatype_distinctness_limit
  4910   fast_arith_neq_limit  	fast_arith_neq_limit
  4911   fast_arith_split_limit	fast_arith_split_limit
  4912 
  4913 * Named collections of theorems may be easily installed as context
  4914 data using the functor NamedThmsFun (see also
  4915 src/Pure/Tools/named_thms.ML).  The user may add or delete facts via
  4916 attributes; there is also a toplevel print command.  This facility is
  4917 just a common case of general context data, which is the preferred way
  4918 for anything more complex than just a list of facts in canonical
  4919 order.
  4920 
  4921 * Isar: command 'declaration' augments a local theory by generic
  4922 declaration functions written in ML.  This enables arbitrary content
  4923 being added to the context, depending on a morphism that tells the
  4924 difference of the original declaration context wrt. the application
  4925 context encountered later on.
  4926 
  4927 * Isar: proper interfaces for simplification procedures.  Command
  4928 'simproc_setup' declares named simprocs (with match patterns, and body
  4929 text in ML).  Attribute "simproc" adds/deletes simprocs in the current
  4930 context.  ML antiquotation @{simproc name} retrieves named simprocs.
  4931 
  4932 * Isar: an extra pair of brackets around attribute declarations
  4933 abbreviates a theorem reference involving an internal dummy fact,
  4934 which will be ignored later --- only the effect of the attribute on
  4935 the background context will persist.  This form of in-place
  4936 declarations is particularly useful with commands like 'declare' and
  4937 'using', for example ``have A using [[simproc a]] by simp''.
  4938 
  4939 * Isar: method "assumption" (and implicit closing of subproofs) now
  4940 takes simple non-atomic goal assumptions into account: after applying
  4941 an assumption as a rule the resulting subgoals are solved by atomic
  4942 assumption steps.  This is particularly useful to finish 'obtain'
  4943 goals, such as "!!x. (!!x. P x ==> thesis) ==> P x ==> thesis",
  4944 without referring to the original premise "!!x. P x ==> thesis" in the
  4945 Isar proof context.  POTENTIAL INCOMPATIBILITY: method "assumption" is
  4946 more permissive.
  4947 
  4948 * Isar: implicit use of prems from the Isar proof context is
  4949 considered a legacy feature.  Common applications like ``have A .''
  4950 may be replaced by ``have A by fact'' or ``note `A`''.  In general,
  4951 referencing facts explicitly here improves readability and
  4952 maintainability of proof texts.
  4953 
  4954 * Isar: improper proof element 'guess' is like 'obtain', but derives
  4955 the obtained context from the course of reasoning!  For example:
  4956 
  4957   assume "EX x y. A x & B y"   -- "any previous fact"
  4958   then guess x and y by clarify
  4959 
  4960 This technique is potentially adventurous, depending on the facts and
  4961 proof tools being involved here.
  4962 
  4963 * Isar: known facts from the proof context may be specified as literal
  4964 propositions, using ASCII back-quote syntax.  This works wherever
  4965 named facts used to be allowed so far, in proof commands, proof
  4966 methods, attributes etc.  Literal facts are retrieved from the context
  4967 according to unification of type and term parameters.  For example,
  4968 provided that "A" and "A ==> B" and "!!x. P x ==> Q x" are known
  4969 theorems in the current context, then these are valid literal facts:
  4970 `A` and `A ==> B` and `!!x. P x ==> Q x" as well as `P a ==> Q a` etc.
  4971 
  4972 There is also a proof method "fact" which does the same composition
  4973 for explicit goal states, e.g. the following proof texts coincide with
  4974 certain special cases of literal facts:
  4975 
  4976   have "A" by fact                 ==  note `A`
  4977   have "A ==> B" by fact           ==  note `A ==> B`
  4978   have "!!x. P x ==> Q x" by fact  ==  note `!!x. P x ==> Q x`
  4979   have "P a ==> Q a" by fact       ==  note `P a ==> Q a`
  4980 
  4981 * Isar: ":" (colon) is no longer a symbolic identifier character in
  4982 outer syntax.  Thus symbolic identifiers may be used without
  4983 additional white space in declarations like this: ``assume *: A''.
  4984 
  4985 * Isar: 'print_facts' prints all local facts of the current context,
  4986 both named and unnamed ones.
  4987 
  4988 * Isar: 'def' now admits simultaneous definitions, e.g.:
  4989 
  4990   def x == "t" and y == "u"
  4991 
  4992 * Isar: added command 'unfolding', which is structurally similar to
  4993 'using', but affects both the goal state and facts by unfolding given
  4994 rewrite rules.  Thus many occurrences of the 'unfold' method or
  4995 'unfolded' attribute may be replaced by first-class proof text.
  4996 
  4997 * Isar: methods 'unfold' / 'fold', attributes 'unfolded' / 'folded',
  4998 and command 'unfolding' now all support object-level equalities
  4999 (potentially conditional).  The underlying notion of rewrite rule is
  5000 analogous to the 'rule_format' attribute, but *not* that of the
  5001 Simplifier (which is usually more generous).
  5002 
  5003 * Isar: the new attribute [rotated n] (default n = 1) rotates the
  5004 premises of a theorem by n. Useful in conjunction with drule.
  5005 
  5006 * Isar: the goal restriction operator [N] (default N = 1) evaluates a
  5007 method expression within a sandbox consisting of the first N
  5008 sub-goals, which need to exist.  For example, ``simp_all [3]''
  5009 simplifies the first three sub-goals, while (rule foo, simp_all)[]
  5010 simplifies all new goals that emerge from applying rule foo to the
  5011 originally first one.
  5012 
  5013 * Isar: schematic goals are no longer restricted to higher-order
  5014 patterns; e.g. ``lemma "?P(?x)" by (rule TrueI)'' now works as
  5015 expected.
  5016 
  5017 * Isar: the conclusion of a long theorem statement is now either
  5018 'shows' (a simultaneous conjunction, as before), or 'obtains'
  5019 (essentially a disjunction of cases with local parameters and
  5020 assumptions).  The latter allows to express general elimination rules
  5021 adequately; in this notation common elimination rules look like this:
  5022 
  5023   lemma exE:    -- "EX x. P x ==> (!!x. P x ==> thesis) ==> thesis"
  5024     assumes "EX x. P x"
  5025     obtains x where "P x"
  5026 
  5027   lemma conjE:  -- "A & B ==> (A ==> B ==> thesis) ==> thesis"
  5028     assumes "A & B"
  5029     obtains A and B
  5030 
  5031   lemma disjE:  -- "A | B ==> (A ==> thesis) ==> (B ==> thesis) ==> thesis"
  5032     assumes "A | B"
  5033     obtains
  5034       A
  5035     | B
  5036 
  5037 The subsequent classical rules even refer to the formal "thesis"
  5038 explicitly:
  5039 
  5040   lemma classical:     -- "(~ thesis ==> thesis) ==> thesis"
  5041     obtains "~ thesis"
  5042 
  5043   lemma Peirce's_Law:  -- "((thesis ==> something) ==> thesis) ==> thesis"
  5044     obtains "thesis ==> something"
  5045 
  5046 The actual proof of an 'obtains' statement is analogous to that of the
  5047 Isar proof element 'obtain', only that there may be several cases.
  5048 Optional case names may be specified in parentheses; these will be
  5049 available both in the present proof and as annotations in the
  5050 resulting rule, for later use with the 'cases' method (cf. attribute
  5051 case_names).
  5052 
  5053 * Isar: the assumptions of a long theorem statement are available as
  5054 "assms" fact in the proof context.  This is more appropriate than the
  5055 (historical) "prems", which refers to all assumptions of the current
  5056 context, including those from the target locale, proof body etc.
  5057 
  5058 * Isar: 'print_statement' prints theorems from the current theory or
  5059 proof context in long statement form, according to the syntax of a
  5060 top-level lemma.
  5061 
  5062 * Isar: 'obtain' takes an optional case name for the local context
  5063 introduction rule (default "that").
  5064 
  5065 * Isar: removed obsolete 'concl is' patterns.  INCOMPATIBILITY, use
  5066 explicit (is "_ ==> ?foo") in the rare cases where this still happens
  5067 to occur.
  5068 
  5069 * Pure: syntax "CONST name" produces a fully internalized constant
  5070 according to the current context.  This is particularly useful for
  5071 syntax translations that should refer to internal constant
  5072 representations independently of name spaces.
  5073 
  5074 * Pure: syntax constant for foo (binder "FOO ") is called "foo_binder"
  5075 instead of "FOO ". This allows multiple binder declarations to coexist
  5076 in the same context.  INCOMPATIBILITY.
  5077 
  5078 * Isar/locales: 'notation' provides a robust interface to the 'syntax'
  5079 primitive that also works in a locale context (both for constants and
  5080 fixed variables). Type declaration and internal syntactic representation
  5081 of given constants retrieved from the context. Likewise, the
  5082 'no_notation' command allows to remove given syntax annotations from the
  5083 current context.
  5084 
  5085 * Isar/locales: new derived specification elements 'axiomatization',
  5086 'definition', 'abbreviation', which support type-inference, admit
  5087 object-level specifications (equality, equivalence).  See also the
  5088 isar-ref manual.  Examples:
  5089 
  5090   axiomatization
  5091     eq  (infix "===" 50) where
  5092     eq_refl: "x === x" and eq_subst: "x === y ==> P x ==> P y"
  5093 
  5094   definition "f x y = x + y + 1"
  5095   definition g where "g x = f x x"
  5096 
  5097   abbreviation
  5098     neq  (infix "=!=" 50) where
  5099     "x =!= y == ~ (x === y)"
  5100 
  5101 These specifications may be also used in a locale context.  Then the
  5102 constants being introduced depend on certain fixed parameters, and the
  5103 constant name is qualified by the locale base name.  An internal
  5104 abbreviation takes care for convenient input and output, making the
  5105 parameters implicit and using the original short name.  See also
  5106 src/HOL/ex/Abstract_NAT.thy for an example of deriving polymorphic
  5107 entities from a monomorphic theory.
  5108 
  5109 Presently, abbreviations are only available 'in' a target locale, but
  5110 not inherited by general import expressions.  Also note that
  5111 'abbreviation' may be used as a type-safe replacement for 'syntax' +
  5112 'translations' in common applications.  The "no_abbrevs" print mode
  5113 prevents folding of abbreviations in term output.
  5114 
  5115 Concrete syntax is attached to specified constants in internal form,
  5116 independently of name spaces.  The parse tree representation is
  5117 slightly different -- use 'notation' instead of raw 'syntax', and
  5118 'translations' with explicit "CONST" markup to accommodate this.
  5119 
  5120 * Pure/Isar: unified syntax for new-style specification mechanisms
  5121 (e.g.  'definition', 'abbreviation', or 'inductive' in HOL) admits
  5122 full type inference and dummy patterns ("_").  For example:
  5123 
  5124   definition "K x _ = x"
  5125 
  5126   inductive conj for A B
  5127   where "A ==> B ==> conj A B"
  5128 
  5129 * Pure: command 'print_abbrevs' prints all constant abbreviations of
  5130 the current context.  Print mode "no_abbrevs" prevents inversion of
  5131 abbreviations on output.
  5132 
  5133 * Isar/locales: improved parameter handling: use of locales "var" and
  5134 "struct" no longer necessary; - parameter renamings are no longer
  5135 required to be injective.  For example, this allows to define
  5136 endomorphisms as locale endom = homom mult mult h.
  5137 
  5138 * Isar/locales: changed the way locales with predicates are defined.
  5139 Instead of accumulating the specification, the imported expression is
  5140 now an interpretation.  INCOMPATIBILITY: different normal form of
  5141 locale expressions.  In particular, in interpretations of locales with
  5142 predicates, goals repesenting already interpreted fragments are not
  5143 removed automatically.  Use methods `intro_locales' and
  5144 `unfold_locales'; see below.
  5145 
  5146 * Isar/locales: new methods `intro_locales' and `unfold_locales'
  5147 provide backward reasoning on locales predicates.  The methods are
  5148 aware of interpretations and discharge corresponding goals.
  5149 `intro_locales' is less aggressive then `unfold_locales' and does not
  5150 unfold predicates to assumptions.
  5151 
  5152 * Isar/locales: the order in which locale fragments are accumulated
  5153 has changed.  This enables to override declarations from fragments due
  5154 to interpretations -- for example, unwanted simp rules.
  5155 
  5156 * Isar/locales: interpretation in theories and proof contexts has been
  5157 extended.  One may now specify (and prove) equations, which are
  5158 unfolded in interpreted theorems.  This is useful for replacing
  5159 defined concepts (constants depending on locale parameters) by
  5160 concepts already existing in the target context.  Example:
  5161 
  5162   interpretation partial_order ["op <= :: [int, int] => bool"]
  5163     where "partial_order.less (op <=) (x::int) y = (x < y)"
  5164 
  5165 Typically, the constant `partial_order.less' is created by a
  5166 definition specification element in the context of locale
  5167 partial_order.
  5168 
  5169 * Method "induct": improved internal context management to support
  5170 local fixes and defines on-the-fly. Thus explicit meta-level
  5171 connectives !!  and ==> are rarely required anymore in inductive goals
  5172 (using object-logic connectives for this purpose has been long
  5173 obsolete anyway). Common proof patterns are explained in
  5174 src/HOL/Induct/Common_Patterns.thy, see also
  5175 src/HOL/Isar_examples/Puzzle.thy and src/HOL/Lambda for realistic
  5176 examples.
  5177 
  5178 * Method "induct": improved handling of simultaneous goals. Instead of
  5179 introducing object-level conjunction, the statement is now split into
  5180 several conclusions, while the corresponding symbolic cases are nested
  5181 accordingly. INCOMPATIBILITY, proofs need to be structured explicitly,
  5182 see src/HOL/Induct/Common_Patterns.thy, for example.
  5183 
  5184 * Method "induct": mutual induction rules are now specified as a list
  5185 of rule sharing the same induction cases. HOL packages usually provide
  5186 foo_bar.inducts for mutually defined items foo and bar (e.g. inductive
  5187 predicates/sets or datatypes). INCOMPATIBILITY, users need to specify
  5188 mutual induction rules differently, i.e. like this:
  5189 
  5190   (induct rule: foo_bar.inducts)
  5191   (induct set: foo bar)
  5192   (induct pred: foo bar)
  5193   (induct type: foo bar)
  5194 
  5195 The ML function ProjectRule.projections turns old-style rules into the
  5196 new format.
  5197 
  5198 * Method "coinduct": dual of induction, see
  5199 src/HOL/Library/Coinductive_List.thy for various examples.
  5200 
  5201 * Method "cases", "induct", "coinduct": the ``(open)'' option is
  5202 considered a legacy feature.
  5203 
  5204 * Attribute "symmetric" produces result with standardized schematic
  5205 variables (index 0).  Potential INCOMPATIBILITY.
  5206 
  5207 * Simplifier: by default the simplifier trace only shows top level
  5208 rewrites now. That is, trace_simp_depth_limit is set to 1 by
  5209 default. Thus there is less danger of being flooded by the trace. The
  5210 trace indicates where parts have been suppressed.
  5211   
  5212 * Provers/classical: removed obsolete classical version of elim_format
  5213 attribute; classical elim/dest rules are now treated uniformly when
  5214 manipulating the claset.
  5215 
  5216 * Provers/classical: stricter checks to ensure that supplied intro,
  5217 dest and elim rules are well-formed; dest and elim rules must have at
  5218 least one premise.
  5219 
  5220 * Provers/classical: attributes dest/elim/intro take an optional
  5221 weight argument for the rule (just as the Pure versions).  Weights are
  5222 ignored by automated tools, but determine the search order of single
  5223 rule steps.
  5224 
  5225 * Syntax: input syntax now supports dummy variable binding "%_. b",
  5226 where the body does not mention the bound variable.  Note that dummy
  5227 patterns implicitly depend on their context of bounds, which makes
  5228 "{_. _}" match any set comprehension as expected.  Potential
  5229 INCOMPATIBILITY -- parse translations need to cope with syntactic
  5230 constant "_idtdummy" in the binding position.
  5231 
  5232 * Syntax: removed obsolete syntactic constant "_K" and its associated
  5233 parse translation.  INCOMPATIBILITY -- use dummy abstraction instead,
  5234 for example "A -> B" => "Pi A (%_. B)".
  5235 
  5236 * Pure: 'class_deps' command visualizes the subclass relation, using
  5237 the graph browser tool.
  5238 
  5239 * Pure: 'print_theory' now suppresses certain internal declarations by
  5240 default; use '!' option for full details.
  5241 
  5242 
  5243 *** HOL ***
  5244 
  5245 * Method "metis" proves goals by applying the Metis general-purpose
  5246 resolution prover (see also http://gilith.com/software/metis/).
  5247 Examples are in the directory MetisExamples.  WARNING: the
  5248 Isabelle/HOL-Metis integration does not yet work properly with
  5249 multi-threading.
  5250   
  5251 * Command 'sledgehammer' invokes external automatic theorem provers as
  5252 background processes.  It generates calls to the "metis" method if
  5253 successful. These can be pasted into the proof.  Users do not have to
  5254 wait for the automatic provers to return.  WARNING: does not really
  5255 work with multi-threading.
  5256 
  5257 * New "auto_quickcheck" feature tests outermost goal statements for
  5258 potential counter-examples.  Controlled by ML references
  5259 auto_quickcheck (default true) and auto_quickcheck_time_limit (default
  5260 5000 milliseconds).  Fails silently if statements is outside of
  5261 executable fragment, or any other codgenerator problem occurs.
  5262 
  5263 * New constant "undefined" with axiom "undefined x = undefined".
  5264 
  5265 * Added class "HOL.eq", allowing for code generation with polymorphic
  5266 equality.
  5267 
  5268 * Some renaming of class constants due to canonical name prefixing in
  5269 the new 'class' package:
  5270 
  5271     HOL.abs ~> HOL.abs_class.abs
  5272     HOL.divide ~> HOL.divide_class.divide
  5273     0 ~> HOL.zero_class.zero
  5274     1 ~> HOL.one_class.one
  5275     op + ~> HOL.plus_class.plus
  5276     op - ~> HOL.minus_class.minus
  5277     uminus ~> HOL.minus_class.uminus
  5278     op * ~> HOL.times_class.times
  5279     op < ~> HOL.ord_class.less
  5280     op <= > HOL.ord_class.less_eq
  5281     Nat.power ~> Power.power_class.power
  5282     Nat.size ~> Nat.size_class.size
  5283     Numeral.number_of ~> Numeral.number_class.number_of
  5284     FixedPoint.Inf ~> Lattices.complete_lattice_class.Inf
  5285     FixedPoint.Sup ~> Lattices.complete_lattice_class.Sup
  5286     Orderings.min ~> Orderings.ord_class.min
  5287     Orderings.max ~> Orderings.ord_class.max
  5288     Divides.op div ~> Divides.div_class.div
  5289     Divides.op mod ~> Divides.div_class.mod
  5290     Divides.op dvd ~> Divides.div_class.dvd
  5291 
  5292 INCOMPATIBILITY.  Adaptions may be required in the following cases:
  5293 
  5294 a) User-defined constants using any of the names "plus", "minus",
  5295 "times", "less" or "less_eq". The standard syntax translations for
  5296 "+", "-" and "*" may go wrong.  INCOMPATIBILITY: use more specific
  5297 names.
  5298 
  5299 b) Variables named "plus", "minus", "times", "less", "less_eq"
  5300 INCOMPATIBILITY: use more specific names.
  5301 
  5302 c) Permutative equations (e.g. "a + b = b + a")
  5303 Since the change of names also changes the order of terms, permutative
  5304 rewrite rules may get applied in a different order. Experience shows
  5305 that this is rarely the case (only two adaptions in the whole Isabelle
  5306 distribution).  INCOMPATIBILITY: rewrite proofs
  5307 
  5308 d) ML code directly refering to constant names
  5309 This in general only affects hand-written proof tactics, simprocs and
  5310 so on.  INCOMPATIBILITY: grep your sourcecode and replace names.
  5311 Consider using @{const_name} antiquotation.
  5312 
  5313 * New class "default" with associated constant "default".
  5314 
  5315 * Function "sgn" is now overloaded and available on int, real, complex
  5316 (and other numeric types), using class "sgn".  Two possible defs of
  5317 sgn are given as equational assumptions in the classes sgn_if and
  5318 sgn_div_norm; ordered_idom now also inherits from sgn_if.
  5319 INCOMPATIBILITY.
  5320 
  5321 * Locale "partial_order" now unified with class "order" (cf. theory
  5322 Orderings), added parameter "less".  INCOMPATIBILITY.
  5323 
  5324 * Renamings in classes "order" and "linorder": facts "refl", "trans" and
  5325 "cases" to "order_refl", "order_trans" and "linorder_cases", to avoid
  5326 clashes with HOL "refl" and "trans".  INCOMPATIBILITY.
  5327 
  5328 * Classes "order" and "linorder": potential INCOMPATIBILITY due to
  5329 changed order of proof goals in instance proofs.
  5330 
  5331 * The transitivity reasoner for partial and linear orders is set up
  5332 for classes "order" and "linorder".  Instances of the reasoner are available
  5333 in all contexts importing or interpreting the corresponding locales.
  5334 Method "order" invokes the reasoner separately; the reasoner
  5335 is also integrated with the Simplifier as a solver.  Diagnostic
  5336 command 'print_orders' shows the available instances of the reasoner
  5337 in the current context.
  5338 
  5339 * Localized monotonicity predicate in theory "Orderings"; integrated
  5340 lemmas max_of_mono and min_of_mono with this predicate.
  5341 INCOMPATIBILITY.
  5342 
  5343 * Formulation of theorem "dense" changed slightly due to integration
  5344 with new class dense_linear_order.
  5345 
  5346 * Uniform lattice theory development in HOL.
  5347 
  5348     constants "meet" and "join" now named "inf" and "sup"
  5349     constant "Meet" now named "Inf"
  5350 
  5351     classes "meet_semilorder" and "join_semilorder" now named
  5352       "lower_semilattice" and "upper_semilattice"
  5353     class "lorder" now named "lattice"
  5354     class "comp_lat" now named "complete_lattice"
  5355 
  5356     Instantiation of lattice classes allows explicit definitions
  5357     for "inf" and "sup" operations (or "Inf" and "Sup" for complete lattices).
  5358 
  5359   INCOMPATIBILITY.  Theorem renames:
  5360 
  5361     meet_left_le            ~> inf_le1
  5362     meet_right_le           ~> inf_le2
  5363     join_left_le            ~> sup_ge1
  5364     join_right_le           ~> sup_ge2
  5365     meet_join_le            ~> inf_sup_ord
  5366     le_meetI                ~> le_infI
  5367     join_leI                ~> le_supI
  5368     le_meet                 ~> le_inf_iff
  5369     le_join                 ~> ge_sup_conv
  5370     meet_idempotent         ~> inf_idem
  5371     join_idempotent         ~> sup_idem
  5372     meet_comm               ~> inf_commute
  5373     join_comm               ~> sup_commute
  5374     meet_leI1               ~> le_infI1
  5375     meet_leI2               ~> le_infI2
  5376     le_joinI1               ~> le_supI1
  5377     le_joinI2               ~> le_supI2
  5378     meet_assoc              ~> inf_assoc
  5379     join_assoc              ~> sup_assoc
  5380     meet_left_comm          ~> inf_left_commute
  5381     meet_left_idempotent    ~> inf_left_idem
  5382     join_left_comm          ~> sup_left_commute
  5383     join_left_idempotent    ~> sup_left_idem
  5384     meet_aci                ~> inf_aci
  5385     join_aci                ~> sup_aci
  5386     le_def_meet             ~> le_iff_inf
  5387     le_def_join             ~> le_iff_sup
  5388     join_absorp2            ~> sup_absorb2
  5389     join_absorp1            ~> sup_absorb1
  5390     meet_absorp1            ~> inf_absorb1
  5391     meet_absorp2            ~> inf_absorb2
  5392     meet_join_absorp        ~> inf_sup_absorb
  5393     join_meet_absorp        ~> sup_inf_absorb
  5394     distrib_join_le         ~> distrib_sup_le
  5395     distrib_meet_le         ~> distrib_inf_le
  5396 
  5397     add_meet_distrib_left   ~> add_inf_distrib_left
  5398     add_join_distrib_left   ~> add_sup_distrib_left
  5399     is_join_neg_meet        ~> is_join_neg_inf
  5400     is_meet_neg_join        ~> is_meet_neg_sup
  5401     add_meet_distrib_right  ~> add_inf_distrib_right
  5402     add_join_distrib_right  ~> add_sup_distrib_right
  5403     add_meet_join_distribs  ~> add_sup_inf_distribs
  5404     join_eq_neg_meet        ~> sup_eq_neg_inf
  5405     meet_eq_neg_join        ~> inf_eq_neg_sup
  5406     add_eq_meet_join        ~> add_eq_inf_sup
  5407     meet_0_imp_0            ~> inf_0_imp_0
  5408     join_0_imp_0            ~> sup_0_imp_0
  5409     meet_0_eq_0             ~> inf_0_eq_0
  5410     join_0_eq_0             ~> sup_0_eq_0
  5411     neg_meet_eq_join        ~> neg_inf_eq_sup
  5412     neg_join_eq_meet        ~> neg_sup_eq_inf
  5413     join_eq_if              ~> sup_eq_if
  5414 
  5415     mono_meet               ~> mono_inf
  5416     mono_join               ~> mono_sup
  5417     meet_bool_eq            ~> inf_bool_eq
  5418     join_bool_eq            ~> sup_bool_eq
  5419     meet_fun_eq             ~> inf_fun_eq
  5420     join_fun_eq             ~> sup_fun_eq
  5421     meet_set_eq             ~> inf_set_eq
  5422     join_set_eq             ~> sup_set_eq
  5423     meet1_iff               ~> inf1_iff
  5424     meet2_iff               ~> inf2_iff
  5425     meet1I                  ~> inf1I
  5426     meet2I                  ~> inf2I
  5427     meet1D1                 ~> inf1D1
  5428     meet2D1                 ~> inf2D1
  5429     meet1D2                 ~> inf1D2
  5430     meet2D2                 ~> inf2D2
  5431     meet1E                  ~> inf1E
  5432     meet2E                  ~> inf2E
  5433     join1_iff               ~> sup1_iff
  5434     join2_iff               ~> sup2_iff
  5435     join1I1                 ~> sup1I1
  5436     join2I1                 ~> sup2I1
  5437     join1I1                 ~> sup1I1
  5438     join2I2                 ~> sup1I2
  5439     join1CI                 ~> sup1CI
  5440     join2CI                 ~> sup2CI
  5441     join1E                  ~> sup1E
  5442     join2E                  ~> sup2E
  5443 
  5444     is_meet_Meet            ~> is_meet_Inf
  5445     Meet_bool_def           ~> Inf_bool_def
  5446     Meet_fun_def            ~> Inf_fun_def
  5447     Meet_greatest           ~> Inf_greatest
  5448     Meet_lower              ~> Inf_lower
  5449     Meet_set_def            ~> Inf_set_def
  5450 
  5451     Sup_def                 ~> Sup_Inf
  5452     Sup_bool_eq             ~> Sup_bool_def
  5453     Sup_fun_eq              ~> Sup_fun_def
  5454     Sup_set_eq              ~> Sup_set_def
  5455 
  5456     listsp_meetI            ~> listsp_infI
  5457     listsp_meet_eq          ~> listsp_inf_eq
  5458 
  5459     meet_min                ~> inf_min
  5460     join_max                ~> sup_max
  5461 
  5462 * Added syntactic class "size"; overloaded constant "size" now has
  5463 type "'a::size ==> bool"
  5464 
  5465 * Internal reorganisation of `size' of datatypes: size theorems
  5466 "foo.size" are no longer subsumed by "foo.simps" (but are still
  5467 simplification rules by default!); theorems "prod.size" now named
  5468 "*.size".
  5469 
  5470 * Class "div" now inherits from class "times" rather than "type".
  5471 INCOMPATIBILITY.
  5472 
  5473 * HOL/Finite_Set: "name-space" locales Lattice, Distrib_lattice,
  5474 Linorder etc.  have disappeared; operations defined in terms of
  5475 fold_set now are named Inf_fin, Sup_fin.  INCOMPATIBILITY.
  5476 
  5477 * HOL/Nat: neq0_conv no longer declared as iff.  INCOMPATIBILITY.
  5478 
  5479 * HOL-Word: New extensive library and type for generic, fixed size
  5480 machine words, with arithmetic, bit-wise, shifting and rotating
  5481 operations, reflection into int, nat, and bool lists, automation for
  5482 linear arithmetic (by automatic reflection into nat or int), including
  5483 lemmas on overflow and monotonicity.  Instantiated to all appropriate
  5484 arithmetic type classes, supporting automatic simplification of
  5485 numerals on all operations.
  5486 
  5487 * Library/Boolean_Algebra: locales for abstract boolean algebras.
  5488 
  5489 * Library/Numeral_Type: numbers as types, e.g. TYPE(32).
  5490 
  5491 * Code generator library theories:
  5492   - Code_Integer represents HOL integers by big integer literals in target
  5493     languages.
  5494   - Code_Char represents HOL characters by character literals in target
  5495     languages.
  5496   - Code_Char_chr like Code_Char, but also offers treatment of character
  5497     codes; includes Code_Integer.
  5498   - Executable_Set allows to generate code for finite sets using lists.
  5499   - Executable_Rat implements rational numbers as triples (sign, enumerator,
  5500     denominator).
  5501   - Executable_Real implements a subset of real numbers, namly those
  5502     representable by rational numbers.
  5503   - Efficient_Nat implements natural numbers by integers, which in general will
  5504     result in higher efficency; pattern matching with 0/Suc is eliminated;
  5505     includes Code_Integer.
  5506   - Code_Index provides an additional datatype index which is mapped to
  5507     target-language built-in integers.
  5508   - Code_Message provides an additional datatype message_string which is isomorphic to
  5509     strings; messages are mapped to target-language strings.
  5510 
  5511 * New package for inductive predicates
  5512 
  5513   An n-ary predicate p with m parameters z_1, ..., z_m can now be defined via
  5514 
  5515     inductive
  5516       p :: "U_1 => ... => U_m => T_1 => ... => T_n => bool"
  5517       for z_1 :: U_1 and ... and z_n :: U_m
  5518     where
  5519       rule_1: "... ==> p z_1 ... z_m t_1_1 ... t_1_n"
  5520     | ...
  5521 
  5522   with full support for type-inference, rather than
  5523 
  5524     consts s :: "U_1 => ... => U_m => (T_1 * ... * T_n) set"
  5525 
  5526     abbreviation p :: "U_1 => ... => U_m => T_1 => ... => T_n => bool"
  5527     where "p z_1 ... z_m x_1 ... x_n == (x_1, ..., x_n) : s z_1 ... z_m"
  5528 
  5529     inductive "s z_1 ... z_m"
  5530     intros
  5531       rule_1: "... ==> (t_1_1, ..., t_1_n) : s z_1 ... z_m"
  5532       ...
  5533 
  5534   For backward compatibility, there is a wrapper allowing inductive
  5535   sets to be defined with the new package via
  5536 
  5537     inductive_set
  5538       s :: "U_1 => ... => U_m => (T_1 * ... * T_n) set"
  5539       for z_1 :: U_1 and ... and z_n :: U_m
  5540     where
  5541       rule_1: "... ==> (t_1_1, ..., t_1_n) : s z_1 ... z_m"
  5542     | ...
  5543 
  5544   or
  5545 
  5546     inductive_set
  5547       s :: "U_1 => ... => U_m => (T_1 * ... * T_n) set"
  5548       and p :: "U_1 => ... => U_m => T_1 => ... => T_n => bool"
  5549       for z_1 :: U_1 and ... and z_n :: U_m
  5550     where
  5551       "p z_1 ... z_m x_1 ... x_n == (x_1, ..., x_n) : s z_1 ... z_m"
  5552     | rule_1: "... ==> p z_1 ... z_m t_1_1 ... t_1_n"
  5553     | ...
  5554 
  5555   if the additional syntax "p ..." is required.
  5556 
  5557   Numerous examples can be found in the subdirectories src/HOL/Auth,
  5558   src/HOL/Bali, src/HOL/Induct, and src/HOL/MicroJava.
  5559 
  5560   INCOMPATIBILITIES:
  5561 
  5562   - Since declaration and definition of inductive sets or predicates
  5563     is no longer separated, abbreviations involving the newly
  5564     introduced sets or predicates must be specified together with the
  5565     introduction rules after the 'where' keyword (see above), rather
  5566     than before the actual inductive definition.
  5567 
  5568   - The variables in induction and elimination rules are now
  5569     quantified in the order of their occurrence in the introduction
  5570     rules, rather than in alphabetical order. Since this may break
  5571     some proofs, these proofs either have to be repaired, e.g. by
  5572     reordering the variables a_i_1 ... a_i_{k_i} in Isar 'case'
  5573     statements of the form
  5574 
  5575       case (rule_i a_i_1 ... a_i_{k_i})
  5576 
  5577     or the old order of quantification has to be restored by explicitly adding
  5578     meta-level quantifiers in the introduction rules, i.e.
  5579 
  5580       | rule_i: "!!a_i_1 ... a_i_{k_i}. ... ==> p z_1 ... z_m t_i_1 ... t_i_n"
  5581 
  5582   - The format of the elimination rules is now
  5583 
  5584       p z_1 ... z_m x_1 ... x_n ==>
  5585         (!!a_1_1 ... a_1_{k_1}. x_1 = t_1_1 ==> ... ==> x_n = t_1_n ==> ... ==> P)
  5586         ==> ... ==> P
  5587 
  5588     for predicates and
  5589 
  5590       (x_1, ..., x_n) : s z_1 ... z_m ==>
  5591         (!!a_1_1 ... a_1_{k_1}. x_1 = t_1_1 ==> ... ==> x_n = t_1_n ==> ... ==> P)
  5592         ==> ... ==> P
  5593 
  5594     for sets rather than
  5595 
  5596       x : s z_1 ... z_m ==>
  5597         (!!a_1_1 ... a_1_{k_1}. x = (t_1_1, ..., t_1_n) ==> ... ==> P)
  5598         ==> ... ==> P
  5599 
  5600     This may require terms in goals to be expanded to n-tuples
  5601     (e.g. using case_tac or simplification with the split_paired_all
  5602     rule) before the above elimination rule is applicable.
  5603 
  5604   - The elimination or case analysis rules for (mutually) inductive
  5605     sets or predicates are now called "p_1.cases" ... "p_k.cases". The
  5606     list of rules "p_1_..._p_k.elims" is no longer available.
  5607 
  5608 * New package "function"/"fun" for general recursive functions,
  5609 supporting mutual and nested recursion, definitions in local contexts,
  5610 more general pattern matching and partiality. See HOL/ex/Fundefs.thy
  5611 for small examples, and the separate tutorial on the function
  5612 package. The old recdef "package" is still available as before, but
  5613 users are encouraged to use the new package.
  5614 
  5615 * Method "lexicographic_order" automatically synthesizes termination
  5616 relations as lexicographic combinations of size measures. 
  5617 
  5618 * Case-expressions allow arbitrary constructor-patterns (including
  5619 "_") and take their order into account, like in functional
  5620 programming.  Internally, this is translated into nested
  5621 case-expressions; missing cases are added and mapped to the predefined
  5622 constant "undefined". In complicated cases printing may no longer show
  5623 the original input but the internal form. Lambda-abstractions allow
  5624 the same form of pattern matching: "% pat1 => e1 | ..." is an
  5625 abbreviation for "%x. case x of pat1 => e1 | ..." where x is a new
  5626 variable.
  5627 
  5628 * IntDef: The constant "int :: nat => int" has been removed; now "int"
  5629 is an abbreviation for "of_nat :: nat => int". The simplification
  5630 rules for "of_nat" have been changed to work like "int" did
  5631 previously.  Potential INCOMPATIBILITY:
  5632   - "of_nat (Suc m)" simplifies to "1 + of_nat m" instead of "of_nat m + 1"
  5633   - of_nat_diff and of_nat_mult are no longer default simp rules
  5634 
  5635 * Method "algebra" solves polynomial equations over (semi)rings using
  5636 Groebner bases. The (semi)ring structure is defined by locales and the
  5637 tool setup depends on that generic context. Installing the method for
  5638 a specific type involves instantiating the locale and possibly adding
  5639 declarations for computation on the coefficients.  The method is
  5640 already instantiated for natural numbers and for the axiomatic class
  5641 of idoms with numerals.  See also the paper by Chaieb and Wenzel at
  5642 CALCULEMUS 2007 for the general principles underlying this
  5643 architecture of context-aware proof-tools.
  5644 
  5645 * Method "ferrack" implements quantifier elimination over
  5646 special-purpose dense linear orders using locales (analogous to
  5647 "algebra"). The method is already installed for class
  5648 {ordered_field,recpower,number_ring} which subsumes real, hyperreal,
  5649 rat, etc.
  5650 
  5651 * Former constant "List.op @" now named "List.append".  Use ML
  5652 antiquotations @{const_name List.append} or @{term " ... @ ... "} to
  5653 circumvent possible incompatibilities when working on ML level.
  5654 
  5655 * primrec: missing cases mapped to "undefined" instead of "arbitrary".
  5656 
  5657 * New function listsum :: 'a list => 'a for arbitrary monoids.
  5658 Special syntax: "SUM x <- xs. f x" (and latex variants)
  5659 
  5660 * New syntax for Haskell-like list comprehension (input only), eg.
  5661 [(x,y). x <- xs, y <- ys, x ~= y], see also src/HOL/List.thy.
  5662 
  5663 * The special syntax for function "filter" has changed from [x :
  5664 xs. P] to [x <- xs. P] to avoid an ambiguity caused by list
  5665 comprehension syntax, and for uniformity.  INCOMPATIBILITY.
  5666 
  5667 * [a..b] is now defined for arbitrary linear orders.  It used to be
  5668 defined on nat only, as an abbreviation for [a..<Suc b]
  5669 INCOMPATIBILITY.
  5670 
  5671 * Renamed lemma "set_take_whileD"  to "set_takeWhileD".
  5672 
  5673 * New functions "sorted" and "sort" in src/HOL/List.thy.
  5674 
  5675 * New lemma collection field_simps (an extension of ring_simps) for
  5676 manipulating (in)equations involving division. Multiplies with all
  5677 denominators that can be proved to be non-zero (in equations) or
  5678 positive/negative (in inequations).
  5679 
  5680 * Lemma collections ring_eq_simps, group_eq_simps and ring_distrib
  5681 have been improved and renamed to ring_simps, group_simps and
  5682 ring_distribs.  Removed lemmas field_xyz in theory Ring_and_Field
  5683 because they were subsumed by lemmas xyz.  INCOMPATIBILITY.
  5684 
  5685 * Theory Library/Commutative_Ring: switched from recdef to function
  5686 package; constants add, mul, pow now curried.  Infix syntax for
  5687 algebraic operations.
  5688 
  5689 * Dropped redundant lemma def_imp_eq in favor of meta_eq_to_obj_eq.
  5690 INCOMPATIBILITY.
  5691 
  5692 * Dropped redundant lemma if_def2 in favor of if_bool_eq_conj.
  5693 INCOMPATIBILITY.
  5694 
  5695 * HOL/records: generalised field-update to take a function on the
  5696 field rather than the new value: r(|A := x|) is translated to A_update
  5697 (K x) r The K-combinator that is internally used is called K_record.
  5698 INCOMPATIBILITY: Usage of the plain update functions has to be
  5699 adapted.
  5700  
  5701 * Class "semiring_0" now contains annihilation axioms x * 0 = 0 and 0
  5702 * x = 0, which are required for a semiring.  Richer structures do not
  5703 inherit from semiring_0 anymore, because this property is a theorem
  5704 there, not an axiom.  INCOMPATIBILITY: In instances of semiring_0,
  5705 there is more to prove, but this is mostly trivial.
  5706 
  5707 * Class "recpower" is generalized to arbitrary monoids, not just
  5708 commutative semirings.  INCOMPATIBILITY: may need to incorporate
  5709 commutativity or semiring properties additionally.
  5710 
  5711 * Constant "List.list_all2" in List.thy now uses authentic syntax.
  5712 INCOMPATIBILITY: translations containing list_all2 may go wrong,
  5713 better use 'abbreviation'.
  5714 
  5715 * Renamed constant "List.op mem" to "List.member".  INCOMPATIBILITY.
  5716 
  5717 * Numeral syntax: type 'bin' which was a mere type copy of 'int' has
  5718 been abandoned in favour of plain 'int'.  INCOMPATIBILITY --
  5719 significant changes for setting up numeral syntax for types:
  5720   - New constants Numeral.pred and Numeral.succ instead
  5721       of former Numeral.bin_pred and Numeral.bin_succ.
  5722   - Use integer operations instead of bin_add, bin_mult and so on.
  5723   - Numeral simplification theorems named Numeral.numeral_simps instead of Bin_simps.
  5724   - ML structure Bin_Simprocs now named Int_Numeral_Base_Simprocs.
  5725 
  5726 See src/HOL/Integ/IntArith.thy for an example setup.
  5727 
  5728 * Command 'normal_form' computes the normal form of a term that may
  5729 contain free variables.  For example ``normal_form "rev [a, b, c]"''
  5730 produces ``[b, c, a]'' (without proof).  This command is suitable for
  5731 heavy-duty computations because the functions are compiled to ML
  5732 first.  Correspondingly, a method "normalization" is provided.  See
  5733 further src/HOL/ex/NormalForm.thy and src/Tools/nbe.ML.
  5734 
  5735 * Alternative iff syntax "A <-> B" for equality on bool (with priority
  5736 25 like -->); output depends on the "iff" print_mode, the default is
  5737 "A = B" (with priority 50).
  5738 
  5739 * Relations less (<) and less_eq (<=) are also available on type bool.
  5740 Modified syntax to disallow nesting without explicit parentheses,
  5741 e.g. "(x < y) < z" or "x < (y < z)", but NOT "x < y < z".  Potential
  5742 INCOMPATIBILITY.
  5743 
  5744 * "LEAST x:A. P" expands to "LEAST x. x:A & P" (input only).
  5745 
  5746 * Relation composition operator "op O" now has precedence 75 and binds
  5747 stronger than union and intersection. INCOMPATIBILITY.
  5748 
  5749 * The old set interval syntax "{m..n(}" (and relatives) has been
  5750 removed.  Use "{m..<n}" (and relatives) instead.
  5751 
  5752 * In the context of the assumption "~(s = t)" the Simplifier rewrites
  5753 "t = s" to False (by simproc "neq").  INCOMPATIBILITY, consider using
  5754 ``declare [[simproc del: neq]]''.
  5755 
  5756 * Simplifier: "m dvd n" where m and n are numbers is evaluated to
  5757 True/False.
  5758 
  5759 * Theorem Cons_eq_map_conv no longer declared as "simp".
  5760 
  5761 * Theorem setsum_mult renamed to setsum_right_distrib.
  5762 
  5763 * Prefer ex1I over ex_ex1I in single-step reasoning, e.g. by the
  5764 ``rule'' method.
  5765 
  5766 * Reimplemented methods "sat" and "satx", with several improvements:
  5767 goals no longer need to be stated as "<prems> ==> False", equivalences
  5768 (i.e. "=" on type bool) are handled, variable names of the form
  5769 "lit_<n>" are no longer reserved, significant speedup.
  5770 
  5771 * Methods "sat" and "satx" can now replay MiniSat proof traces.
  5772 zChaff is still supported as well.
  5773 
  5774 * 'inductive' and 'datatype': provide projections of mutual rules,
  5775 bundled as foo_bar.inducts;
  5776 
  5777 * Library: moved theories Parity, GCD, Binomial, Infinite_Set to
  5778 Library.
  5779 
  5780 * Library: moved theory Accessible_Part to main HOL.
  5781 
  5782 * Library: added theory Coinductive_List of potentially infinite lists
  5783 as greatest fixed-point.
  5784 
  5785 * Library: added theory AssocList which implements (finite) maps as
  5786 association lists.
  5787 
  5788 * Method "evaluation" solves goals (i.e. a boolean expression)
  5789 efficiently by compiling it to ML.  The goal is "proved" (via an
  5790 oracle) if it evaluates to True.
  5791 
  5792 * Linear arithmetic now splits certain operators (e.g. min, max, abs)
  5793 also when invoked by the simplifier.  This results in the Simplifier
  5794 being more powerful on arithmetic goals.  INCOMPATIBILITY.
  5795 Configuration option fast_arith_split_limit=0 recovers the old
  5796 behavior.
  5797 
  5798 * Support for hex (0x20) and binary (0b1001) numerals.
  5799 
  5800 * New method: reify eqs (t), where eqs are equations for an
  5801 interpretation I :: 'a list => 'b => 'c and t::'c is an optional
  5802 parameter, computes a term s::'b and a list xs::'a list and proves the
  5803 theorem I xs s = t. This is also known as reification or quoting. The
  5804 resulting theorem is applied to the subgoal to substitute t with I xs
  5805 s.  If t is omitted, the subgoal itself is reified.
  5806 
  5807 * New method: reflection corr_thm eqs (t). The parameters eqs and (t)
  5808 are as explained above. corr_thm is a theorem for I vs (f t) = I vs t,
  5809 where f is supposed to be a computable function (in the sense of code
  5810 generattion). The method uses reify to compute s and xs as above then
  5811 applies corr_thm and uses normalization by evaluation to "prove" f s =
  5812 r and finally gets the theorem t = r, which is again applied to the
  5813 subgoal. An Example is available in src/HOL/ex/ReflectionEx.thy.
  5814 
  5815 * Reflection: Automatic reification now handels binding, an example is
  5816 available in src/HOL/ex/ReflectionEx.thy
  5817 
  5818 * HOL-Statespace: ``State Spaces: The Locale Way'' introduces a
  5819 command 'statespace' that is similar to 'record', but introduces an
  5820 abstract specification based on the locale infrastructure instead of
  5821 HOL types.  This leads to extra flexibility in composing state spaces,
  5822 in particular multiple inheritance and renaming of components.
  5823 
  5824 
  5825 *** HOL-Complex ***
  5826 
  5827 * Hyperreal: Functions root and sqrt are now defined on negative real
  5828 inputs so that root n (- x) = - root n x and sqrt (- x) = - sqrt x.
  5829 Nonnegativity side conditions have been removed from many lemmas, so
  5830 that more subgoals may now be solved by simplification; potential
  5831 INCOMPATIBILITY.
  5832 
  5833 * Real: new type classes formalize real normed vector spaces and
  5834 algebras, using new overloaded constants scaleR :: real => 'a => 'a
  5835 and norm :: 'a => real.
  5836 
  5837 * Real: constant of_real :: real => 'a::real_algebra_1 injects from
  5838 reals into other types. The overloaded constant Reals :: 'a set is now
  5839 defined as range of_real; potential INCOMPATIBILITY.
  5840 
  5841 * Real: proper support for ML code generation, including 'quickcheck'.
  5842 Reals are implemented as arbitrary precision rationals.
  5843 
  5844 * Hyperreal: Several constants that previously worked only for the
  5845 reals have been generalized, so they now work over arbitrary vector
  5846 spaces. Type annotations may need to be added in some cases; potential
  5847 INCOMPATIBILITY.
  5848 
  5849   Infinitesimal  :: ('a::real_normed_vector) star set
  5850   HFinite        :: ('a::real_normed_vector) star set
  5851   HInfinite      :: ('a::real_normed_vector) star set
  5852   approx         :: ('a::real_normed_vector) star => 'a star => bool
  5853   monad          :: ('a::real_normed_vector) star => 'a star set
  5854   galaxy         :: ('a::real_normed_vector) star => 'a star set
  5855   (NS)LIMSEQ     :: [nat => 'a::real_normed_vector, 'a] => bool
  5856   (NS)convergent :: (nat => 'a::real_normed_vector) => bool
  5857   (NS)Bseq       :: (nat => 'a::real_normed_vector) => bool
  5858   (NS)Cauchy     :: (nat => 'a::real_normed_vector) => bool
  5859   (NS)LIM        :: ['a::real_normed_vector => 'b::real_normed_vector, 'a, 'b] => bool
  5860   is(NS)Cont     :: ['a::real_normed_vector => 'b::real_normed_vector, 'a] => bool
  5861   deriv          :: ['a::real_normed_field => 'a, 'a, 'a] => bool
  5862   sgn            :: 'a::real_normed_vector => 'a
  5863   exp            :: 'a::{recpower,real_normed_field,banach} => 'a
  5864 
  5865 * Complex: Some complex-specific constants are now abbreviations for
  5866 overloaded ones: complex_of_real = of_real, cmod = norm, hcmod =
  5867 hnorm.  Other constants have been entirely removed in favor of the
  5868 polymorphic versions (INCOMPATIBILITY):
  5869 
  5870   approx        <-- capprox
  5871   HFinite       <-- CFinite
  5872   HInfinite     <-- CInfinite
  5873   Infinitesimal <-- CInfinitesimal
  5874   monad         <-- cmonad
  5875   galaxy        <-- cgalaxy
  5876   (NS)LIM       <-- (NS)CLIM, (NS)CRLIM
  5877   is(NS)Cont    <-- is(NS)Contc, is(NS)contCR
  5878   (ns)deriv     <-- (ns)cderiv
  5879 
  5880 
  5881 *** HOL-Algebra ***
  5882 
  5883 * Formalisation of ideals and the quotient construction over rings.
  5884 
  5885 * Order and lattice theory no longer based on records.
  5886 INCOMPATIBILITY.
  5887 
  5888 * Renamed lemmas least_carrier -> least_closed and greatest_carrier ->
  5889 greatest_closed.  INCOMPATIBILITY.
  5890 
  5891 * Method algebra is now set up via an attribute.  For examples see
  5892 Ring.thy.  INCOMPATIBILITY: the method is now weaker on combinations
  5893 of algebraic structures.
  5894 
  5895 * Renamed theory CRing to Ring.
  5896 
  5897 
  5898 *** HOL-Nominal ***
  5899 
  5900 * Substantial, yet incomplete support for nominal datatypes (binding
  5901 structures) based on HOL-Nominal logic.  See src/HOL/Nominal and
  5902 src/HOL/Nominal/Examples.  Prospective users should consult
  5903 http://isabelle.in.tum.de/nominal/
  5904 
  5905 
  5906 *** ML ***
  5907 
  5908 * ML basics: just one true type int, which coincides with IntInf.int
  5909 (even on SML/NJ).
  5910 
  5911 * ML within Isar: antiquotations allow to embed statically-checked
  5912 formal entities in the source, referring to the context available at
  5913 compile-time.  For example:
  5914 
  5915 ML {* @{sort "{zero,one}"} *}
  5916 ML {* @{typ "'a => 'b"} *}
  5917 ML {* @{term "%x. x"} *}
  5918 ML {* @{prop "x == y"} *}
  5919 ML {* @{ctyp "'a => 'b"} *}
  5920 ML {* @{cterm "%x. x"} *}
  5921 ML {* @{cprop "x == y"} *}
  5922 ML {* @{thm asm_rl} *}
  5923 ML {* @{thms asm_rl} *}
  5924 ML {* @{type_name c} *}
  5925 ML {* @{type_syntax c} *}
  5926 ML {* @{const_name c} *}
  5927 ML {* @{const_syntax c} *}
  5928 ML {* @{context} *}
  5929 ML {* @{theory} *}
  5930 ML {* @{theory Pure} *}
  5931 ML {* @{theory_ref} *}
  5932 ML {* @{theory_ref Pure} *}
  5933 ML {* @{simpset} *}
  5934 ML {* @{claset} *}
  5935 ML {* @{clasimpset} *}
  5936 
  5937 The same works for sources being ``used'' within an Isar context.
  5938 
  5939 * ML in Isar: improved error reporting; extra verbosity with
  5940 ML_Context.trace enabled.
  5941 
  5942 * Pure/General/table.ML: the join operations now works via exceptions
  5943 DUP/SAME instead of type option. This is simpler in simple cases, and
  5944 admits slightly more efficient complex applications.
  5945 
  5946 * Pure: 'advanced' translation functions (parse_translation etc.) now
  5947 use Context.generic instead of just theory.
  5948 
  5949 * Pure: datatype Context.generic joins theory/Proof.context and
  5950 provides some facilities for code that works in either kind of
  5951 context, notably GenericDataFun for uniform theory and proof data.
  5952 
  5953 * Pure: simplified internal attribute type, which is now always
  5954 Context.generic * thm -> Context.generic * thm. Global (theory) vs.
  5955 local (Proof.context) attributes have been discontinued, while
  5956 minimizing code duplication. Thm.rule_attribute and
  5957 Thm.declaration_attribute build canonical attributes; see also structure
  5958 Context for further operations on Context.generic, notably
  5959 GenericDataFun. INCOMPATIBILITY, need to adapt attribute type
  5960 declarations and definitions.
  5961 
  5962 * Context data interfaces (Theory/Proof/GenericDataFun): removed
  5963 name/print, uninitialized data defaults to ad-hoc copy of empty value,
  5964 init only required for impure data. INCOMPATIBILITY: empty really need
  5965 to be empty (no dependencies on theory content!)
  5966 
  5967 * Pure/kernel: consts certification ignores sort constraints given in
  5968 signature declarations. (This information is not relevant to the
  5969 logic, but only for type inference.) SIGNIFICANT INTERNAL CHANGE,
  5970 potential INCOMPATIBILITY.
  5971 
  5972 * Pure: axiomatic type classes are now purely definitional, with
  5973 explicit proofs of class axioms and super class relations performed
  5974 internally. See Pure/axclass.ML for the main internal interfaces --
  5975 notably AxClass.define_class supercedes AxClass.add_axclass, and
  5976 AxClass.axiomatize_class/classrel/arity supersede
  5977 Sign.add_classes/classrel/arities.
  5978 
  5979 * Pure/Isar: Args/Attrib parsers operate on Context.generic --
  5980 global/local versions on theory vs. Proof.context have been
  5981 discontinued; Attrib.syntax and Method.syntax have been adapted
  5982 accordingly.  INCOMPATIBILITY, need to adapt parser expressions for
  5983 attributes, methods, etc.
  5984 
  5985 * Pure: several functions of signature "... -> theory -> theory * ..."
  5986 have been reoriented to "... -> theory -> ... * theory" in order to
  5987 allow natural usage in combination with the ||>, ||>>, |-> and
  5988 fold_map combinators.
  5989 
  5990 * Pure: official theorem names (closed derivations) and additional
  5991 comments (tags) are now strictly separate.  Name hints -- which are
  5992 maintained as tags -- may be attached any time without affecting the
  5993 derivation.
  5994 
  5995 * Pure: primitive rule lift_rule now takes goal cterm instead of an
  5996 actual goal state (thm).  Use Thm.lift_rule (Thm.cprem_of st i) to
  5997 achieve the old behaviour.
  5998 
  5999 * Pure: the "Goal" constant is now called "prop", supporting a
  6000 slightly more general idea of ``protecting'' meta-level rule
  6001 statements.
  6002 
  6003 * Pure: Logic.(un)varify only works in a global context, which is now
  6004 enforced instead of silently assumed.  INCOMPATIBILITY, may use
  6005 Logic.legacy_(un)varify as temporary workaround.
  6006 
  6007 * Pure: structure Name provides scalable operations for generating
  6008 internal variable names, notably Name.variants etc.  This replaces
  6009 some popular functions from term.ML:
  6010 
  6011   Term.variant		->  Name.variant
  6012   Term.variantlist	->  Name.variant_list
  6013   Term.invent_names	->  Name.invent_list
  6014 
  6015 Note that low-level renaming rarely occurs in new code -- operations
  6016 from structure Variable are used instead (see below).
  6017 
  6018 * Pure: structure Variable provides fundamental operations for proper
  6019 treatment of fixed/schematic variables in a context.  For example,
  6020 Variable.import introduces fixes for schematics of given facts and
  6021 Variable.export reverses the effect (up to renaming) -- this replaces
  6022 various freeze_thaw operations.
  6023 
  6024 * Pure: structure Goal provides simple interfaces for
  6025 init/conclude/finish and tactical prove operations (replacing former
  6026 Tactic.prove).  Goal.prove is the canonical way to prove results
  6027 within a given context; Goal.prove_global is a degraded version for
  6028 theory level goals, including a global Drule.standard.  Note that
  6029 OldGoals.prove_goalw_cterm has long been obsolete, since it is
  6030 ill-behaved in a local proof context (e.g. with local fixes/assumes or
  6031 in a locale context).
  6032 
  6033 * Pure/Syntax: generic interfaces for parsing (Syntax.parse_term etc.)
  6034 and type checking (Syntax.check_term etc.), with common combinations
  6035 (Syntax.read_term etc.). These supersede former Sign.read_term etc.
  6036 which are considered legacy and await removal.
  6037 
  6038 * Pure/Syntax: generic interfaces for type unchecking
  6039 (Syntax.uncheck_terms etc.) and unparsing (Syntax.unparse_term etc.),
  6040 with common combinations (Syntax.pretty_term, Syntax.string_of_term
  6041 etc.).  Former Sign.pretty_term, Sign.string_of_term etc. are still
  6042 available for convenience, but refer to the very same operations using
  6043 a mere theory instead of a full context.
  6044 
  6045 * Isar: simplified treatment of user-level errors, using exception
  6046 ERROR of string uniformly.  Function error now merely raises ERROR,
  6047 without any side effect on output channels.  The Isar toplevel takes
  6048 care of proper display of ERROR exceptions.  ML code may use plain
  6049 handle/can/try; cat_error may be used to concatenate errors like this:
  6050 
  6051   ... handle ERROR msg => cat_error msg "..."
  6052 
  6053 Toplevel ML code (run directly or through the Isar toplevel) may be
  6054 embedded into the Isar toplevel with exception display/debug like
  6055 this:
  6056 
  6057   Isar.toplevel (fn () => ...)
  6058 
  6059 INCOMPATIBILITY, removed special transform_error facilities, removed
  6060 obsolete variants of user-level exceptions (ERROR_MESSAGE,
  6061 Context.PROOF, ProofContext.CONTEXT, Proof.STATE, ProofHistory.FAIL)
  6062 -- use plain ERROR instead.
  6063 
  6064 * Isar: theory setup now has type (theory -> theory), instead of a
  6065 list.  INCOMPATIBILITY, may use #> to compose setup functions.
  6066 
  6067 * Isar: ML toplevel pretty printer for type Proof.context, subject to
  6068 ProofContext.debug/verbose flags.
  6069 
  6070 * Isar: Toplevel.theory_to_proof admits transactions that modify the
  6071 theory before entering a proof state.  Transactions now always see a
  6072 quasi-functional intermediate checkpoint, both in interactive and
  6073 batch mode.
  6074 
  6075 * Isar: simplified interfaces for outer syntax.  Renamed
  6076 OuterSyntax.add_keywords to OuterSyntax.keywords.  Removed
  6077 OuterSyntax.add_parsers -- this functionality is now included in
  6078 OuterSyntax.command etc.  INCOMPATIBILITY.
  6079 
  6080 * Simplifier: the simpset of a running simplification process now
  6081 contains a proof context (cf. Simplifier.the_context), which is the
  6082 very context that the initial simpset has been retrieved from (by
  6083 simpset_of/local_simpset_of).  Consequently, all plug-in components
  6084 (solver, looper etc.) may depend on arbitrary proof data.
  6085 
  6086 * Simplifier.inherit_context inherits the proof context (plus the
  6087 local bounds) of the current simplification process; any simproc
  6088 etc. that calls the Simplifier recursively should do this!  Removed
  6089 former Simplifier.inherit_bounds, which is already included here --
  6090 INCOMPATIBILITY.  Tools based on low-level rewriting may even have to
  6091 specify an explicit context using Simplifier.context/theory_context.
  6092 
  6093 * Simplifier/Classical Reasoner: more abstract interfaces
  6094 change_simpset/claset for modifying the simpset/claset reference of a
  6095 theory; raw versions simpset/claset_ref etc. have been discontinued --
  6096 INCOMPATIBILITY.
  6097 
  6098 * Provers: more generic wrt. syntax of object-logics, avoid hardwired
  6099 "Trueprop" etc.
  6100 
  6101 
  6102 *** System ***
  6103 
  6104 * settings: the default heap location within ISABELLE_HOME_USER now
  6105 includes ISABELLE_IDENTIFIER.  This simplifies use of multiple
  6106 Isabelle installations.
  6107 
  6108 * isabelle-process: option -S (secure mode) disables some critical
  6109 operations, notably runtime compilation and evaluation of ML source
  6110 code.
  6111 
  6112 * Basic Isabelle mode for jEdit, see Isabelle/lib/jedit/.
  6113 
  6114 * Support for parallel execution, using native multicore support of
  6115 Poly/ML 5.1.  The theory loader exploits parallelism when processing
  6116 independent theories, according to the given theory header
  6117 specifications. The maximum number of worker threads is specified via
  6118 usedir option -M or the "max-threads" setting in Proof General. A
  6119 speedup factor of 1.5--3.5 can be expected on a 4-core machine, and up
  6120 to 6 on a 8-core machine.  User-code needs to observe certain
  6121 guidelines for thread-safe programming, see appendix A in the Isar
  6122 Implementation manual.
  6123 
  6124 
  6125 
  6126 New in Isabelle2005 (October 2005)
  6127 ----------------------------------
  6128 
  6129 *** General ***
  6130 
  6131 * Theory headers: the new header syntax for Isar theories is
  6132 
  6133   theory <name>
  6134   imports <theory1> ... <theoryN>
  6135   uses <file1> ... <fileM>
  6136   begin
  6137 
  6138 where the 'uses' part is optional.  The previous syntax
  6139 
  6140   theory <name> = <theory1> + ... + <theoryN>:
  6141 
  6142 will disappear in the next release.  Use isatool fixheaders to convert
  6143 existing theory files.  Note that there is no change in ancient
  6144 non-Isar theories now, but these will disappear soon.
  6145 
  6146 * Theory loader: parent theories can now also be referred to via
  6147 relative and absolute paths.
  6148 
  6149 * Command 'find_theorems' searches for a list of criteria instead of a
  6150 list of constants. Known criteria are: intro, elim, dest, name:string,
  6151 simp:term, and any term. Criteria can be preceded by '-' to select
  6152 theorems that do not match. Intro, elim, dest select theorems that
  6153 match the current goal, name:s selects theorems whose fully qualified
  6154 name contain s, and simp:term selects all simplification rules whose
  6155 lhs match term.  Any other term is interpreted as pattern and selects
  6156 all theorems matching the pattern. Available in ProofGeneral under
  6157 'ProofGeneral -> Find Theorems' or C-c C-f.  Example:
  6158 
  6159   C-c C-f (100) "(_::nat) + _ + _" intro -name: "HOL."
  6160 
  6161 prints the last 100 theorems matching the pattern "(_::nat) + _ + _",
  6162 matching the current goal as introduction rule and not having "HOL."
  6163 in their name (i.e. not being defined in theory HOL).
  6164 
  6165 * Command 'thms_containing' has been discontinued in favour of
  6166 'find_theorems'; INCOMPATIBILITY.
  6167 
  6168 * Communication with Proof General is now 8bit clean, which means that
  6169 Unicode text in UTF-8 encoding may be used within theory texts (both
  6170 formal and informal parts).  Cf. option -U of the Isabelle Proof
  6171 General interface.  Here are some simple examples (cf. src/HOL/ex):
  6172 
  6173   http://isabelle.in.tum.de/library/HOL/ex/Hebrew.html
  6174   http://isabelle.in.tum.de/library/HOL/ex/Chinese.html
  6175 
  6176 * Improved efficiency of the Simplifier and, to a lesser degree, the
  6177 Classical Reasoner.  Typical big applications run around 2 times
  6178 faster.
  6179 
  6180 
  6181 *** Document preparation ***
  6182 
  6183 * Commands 'display_drafts' and 'print_drafts' perform simple output
  6184 of raw sources.  Only those symbols that do not require additional
  6185 LaTeX packages (depending on comments in isabellesym.sty) are
  6186 displayed properly, everything else is left verbatim.  isatool display
  6187 and isatool print are used as front ends (these are subject to the
  6188 DVI/PDF_VIEWER and PRINT_COMMAND settings, respectively).
  6189 
  6190 * Command tags control specific markup of certain regions of text,
  6191 notably folding and hiding.  Predefined tags include "theory" (for
  6192 theory begin and end), "proof" for proof commands, and "ML" for
  6193 commands involving ML code; the additional tags "visible" and
  6194 "invisible" are unused by default.  Users may give explicit tag
  6195 specifications in the text, e.g. ''by %invisible (auto)''.  The
  6196 interpretation of tags is determined by the LaTeX job during document
  6197 preparation: see option -V of isatool usedir, or options -n and -t of
  6198 isatool document, or even the LaTeX macros \isakeeptag, \isafoldtag,
  6199 \isadroptag.
  6200 
  6201 Several document versions may be produced at the same time via isatool
  6202 usedir (the generated index.html will link all of them).  Typical
  6203 specifications include ''-V document=theory,proof,ML'' to present
  6204 theory/proof/ML parts faithfully, ''-V outline=/proof,/ML'' to fold
  6205 proof and ML commands, and ''-V mutilated=-theory,-proof,-ML'' to omit
  6206 these parts without any formal replacement text.  The Isabelle site
  6207 default settings produce ''document'' and ''outline'' versions as
  6208 specified above.
  6209 
  6210 * Several new antiquotations:
  6211 
  6212   @{term_type term} prints a term with its type annotated;
  6213 
  6214   @{typeof term} prints the type of a term;
  6215 
  6216   @{const const} is the same as @{term const}, but checks that the
  6217   argument is a known logical constant;
  6218 
  6219   @{term_style style term} and @{thm_style style thm} print a term or
  6220   theorem applying a "style" to it
  6221 
  6222   @{ML text}
  6223 
  6224 Predefined styles are 'lhs' and 'rhs' printing the lhs/rhs of
  6225 definitions, equations, inequations etc., 'concl' printing only the
  6226 conclusion of a meta-logical statement theorem, and 'prem1' .. 'prem19'
  6227 to print the specified premise.  TermStyle.add_style provides an ML
  6228 interface for introducing further styles.  See also the "LaTeX Sugar"
  6229 document practical applications.  The ML antiquotation prints
  6230 type-checked ML expressions verbatim.
  6231 
  6232 * Markup commands 'chapter', 'section', 'subsection', 'subsubsection',
  6233 and 'text' support optional locale specification '(in loc)', which