NEWS
author wenzelm
Sat Apr 28 18:09:50 2012 +0200 (2012-04-28)
changeset 47829 0e36cc70cb3e
parent 47827 13530d774a21
child 47842 bfc08ce7b7b9
permissions -rw-r--r--
some re-ordering;
     1 Isabelle NEWS -- history user-relevant changes
     2 ==============================================
     3 
     4 New in Isabelle2012 (May 2012)
     5 ------------------------------
     6 
     7 *** General ***
     8 
     9 * Prover IDE (PIDE) improvements:
    10 
    11   - more robust Sledgehammer integration (as before the sledgehammer
    12     command-line needs to be typed into the source buffer)
    13   - markup for bound variables
    14   - markup for types of term variables (displayed as tooltips)
    15   - support for user-defined Isar commands within the running session
    16   - improved support for Unicode outside original 16bit range
    17     e.g. glyph for \<A> (thanks to jEdit 4.5.1)
    18 
    19 * Forward declaration of outer syntax keywords within the theory
    20 header -- minor INCOMPATIBILITY for user-defined commands.  Allow new
    21 commands to be used in the same theory where defined.
    22 
    23 * Auxiliary contexts indicate block structure for specifications with
    24 additional parameters and assumptions.  Such unnamed contexts may be
    25 nested within other targets, like 'theory', 'locale', 'class',
    26 'instantiation' etc.  Results from the local context are generalized
    27 accordingly and applied to the enclosing target context.  Example:
    28 
    29   context
    30     fixes x y z :: 'a
    31     assumes xy: "x = y" and yz: "y = z"
    32   begin
    33 
    34   lemma my_trans: "x = z" using xy yz by simp
    35 
    36   end
    37 
    38   thm my_trans
    39 
    40 The most basic application is to factor-out context elements of
    41 several fixes/assumes/shows theorem statements, e.g. see
    42 ~~/src/HOL/Isar_Examples/Group_Context.thy
    43 
    44 Any other local theory specification element works within the "context
    45 ... begin ... end" block as well.
    46 
    47 * Bundled declarations associate attributed fact expressions with a
    48 given name in the context.  These may be later included in other
    49 contexts.  This allows to manage context extensions casually, without
    50 the logical dependencies of locales and locale interpretation.
    51 
    52 See commands 'bundle', 'include', 'including' etc. in the isar-ref
    53 manual.
    54 
    55 * Commands 'lemmas' and 'theorems' allow local variables using 'for'
    56 declaration, and results are standardized before being stored.  Thus
    57 old-style "standard" after instantiation or composition of facts
    58 becomes obsolete.  Minor INCOMPATIBILITY, due to potential change of
    59 indices of schematic variables.
    60 
    61 * Rule attributes in local theory declarations (e.g. locale or class)
    62 are now statically evaluated: the resulting theorem is stored instead
    63 of the original expression.  INCOMPATIBILITY in rare situations, where
    64 the historic accident of dynamic re-evaluation in interpretations
    65 etc. was exploited.
    66 
    67 * New tutorial "Programming and Proving in Isabelle/HOL"
    68 ("prog-prove").  It completely supersedes "A Tutorial Introduction to
    69 Structured Isar Proofs" ("isar-overview"), which has been removed.  It
    70 also supersedes "Isabelle/HOL, A Proof Assistant for Higher-Order
    71 Logic" as the recommended beginners tutorial, but does not cover all
    72 of the material of that old tutorial.
    73 
    74 * Updated and extended reference manuals: "isar-ref",
    75 "implementation", "system"; reduced remaining material in old "ref"
    76 manual.
    77 
    78 
    79 *** Pure ***
    80 
    81 * Discontinued old "prems" fact, which used to refer to the accidental
    82 collection of foundational premises in the context (already marked as
    83 legacy since Isabelle2011).
    84 
    85 * Rule composition via attribute "OF" (or ML functions OF/MRS) is more
    86 tolerant against multiple unifiers, as long as the final result is
    87 unique.  (As before, rules are composed in canonical right-to-left
    88 order to accommodate newly introduced premises.)
    89 
    90 * Command 'definition' no longer exports the foundational "raw_def"
    91 into the user context.  Minor INCOMPATIBILITY, may use the regular
    92 "def" result with attribute "abs_def" to imitate the old version.
    93 
    94 * Renamed some inner syntax categories:
    95 
    96     num ~> num_token
    97     xnum ~> xnum_token
    98     xstr ~> str_token
    99 
   100 Minor INCOMPATIBILITY.  Note that in practice "num_const" or
   101 "num_position" etc. are mainly used instead (which also include
   102 position information via constraints).
   103 
   104 * Simplified configuration options for syntax ambiguity: see
   105 "syntax_ambiguity_warning" and "syntax_ambiguity_limit" in isar-ref
   106 manual.  Minor INCOMPATIBILITY.
   107 
   108 * Attribute "abs_def" turns an equation of the form "f x y == t" into
   109 "f == %x y. t", which ensures that "simp" or "unfold" steps always
   110 expand it.  This also works for object-logic equality.  (Formerly
   111 undocumented feature.)
   112 
   113 * Obsolete command 'types' has been discontinued.  Use 'type_synonym'
   114 instead.  INCOMPATIBILITY.
   115 
   116 * Old code generator for SML and its commands 'code_module',
   117 'code_library', 'consts_code', 'types_code' have been discontinued.
   118 Use commands of the generic code generator instead.  INCOMPATIBILITY.
   119 
   120 * Redundant attribute "code_inline" has been discontinued. Use
   121 "code_unfold" instead.  INCOMPATIBILITY.
   122 
   123 * Dropped attribute "code_unfold_post" in favor of the its dual
   124 "code_abbrev", which yields a common pattern in definitions like
   125 
   126   definition [code_abbrev]: "f = t"
   127 
   128 INCOMPATIBILITY.
   129 
   130 * Sort constraints are now propagated in simultaneous statements, just
   131 like type constraints.  INCOMPATIBILITY in rare situations, where
   132 distinct sorts used to be assigned accidentally.  For example:
   133 
   134   lemma "P (x::'a::foo)" and "Q (y::'a::bar)"  -- "now illegal"
   135 
   136   lemma "P (x::'a)" and "Q (y::'a::bar)"
   137     -- "now uniform 'a::bar instead of default sort for first occurrence (!)"
   138 
   139 
   140 *** HOL ***
   141 
   142 * Type 'a set is now a proper type constructor (just as before
   143 Isabelle2008).  Definitions mem_def and Collect_def have disappeared.
   144 Non-trivial INCOMPATIBILITY.  For developments keeping predicates and
   145 sets separate, it is often sufficient to rephrase sets S accidentally
   146 used as predicates by "%x. x : S" and predicates P accidentally used
   147 as sets by "{x. P x}".  Corresponding proofs in a first step should be
   148 pruned from any tinkering with former theorems mem_def and Collect_def
   149 as far as possible.
   150 
   151 For developments which deliberately mixed predicates and sets, a
   152 planning step is necessary to determine what should become a predicate
   153 and what a set.  It can be helpful to carry out that step in
   154 Isabelle2011-1 before jumping right into the current release.
   155 
   156 * The representation of numerals has changed.  Datatype "num"
   157 represents strictly positive binary numerals, along with functions
   158 "numeral :: num => 'a" and "neg_numeral :: num => 'a" to represent
   159 positive and negated numeric literals, respectively. (See definitions
   160 in ~~/src/HOL/Num.thy.) Potential INCOMPATIBILITY, some user theories
   161 may require adaptations as follows:
   162 
   163   - Theorems with number_ring or number_semiring constraints: These
   164     classes are gone; use comm_ring_1 or comm_semiring_1 instead.
   165 
   166   - Theories defining numeric types: Remove number, number_semiring,
   167     and number_ring instances. Defer all theorems about numerals until
   168     after classes one and semigroup_add have been instantiated.
   169 
   170   - Numeral-only simp rules: Replace each rule having a "number_of v"
   171     pattern with two copies, one for numeral and one for neg_numeral.
   172 
   173   - Theorems about subclasses of semiring_1 or ring_1: These classes
   174     automatically support numerals now, so more simp rules and
   175     simprocs may now apply within the proof.
   176 
   177   - Definitions and theorems using old constructors Pls/Min/Bit0/Bit1:
   178     Redefine using other integer operations.
   179 
   180 * Code generation by default implements sets as container type rather
   181 than predicates.  INCOMPATIBILITY.
   182 
   183 * HOL-Import: Re-implementation from scratch is faster, simpler, and
   184 more scalable.  Requires a proof bundle, which is available as an
   185 external component.  Discontinued old (and mostly dead) Importer for
   186 HOL4 and HOL Light.  INCOMPATIBILITY.
   187 
   188 * New type synonym 'a rel = ('a * 'a) set
   189 
   190 * New "case_product" attribute to generate a case rule doing multiple
   191 case distinctions at the same time.  E.g.
   192 
   193   list.exhaust [case_product nat.exhaust]
   194 
   195 produces a rule which can be used to perform case distinction on both
   196 a list and a nat.
   197 
   198 * New Transfer package:
   199 
   200   - transfer_rule attribute: Maintains a collection of transfer rules,
   201     which relate constants at two different types. Transfer rules may
   202     relate different type instances of the same polymorphic constant,
   203     or they may relate an operation on a raw type to a corresponding
   204     operation on an abstract type (quotient or subtype). For example:
   205 
   206     ((A ===> B) ===> list_all2 A ===> list_all2 B) map map
   207     (cr_int ===> cr_int ===> cr_int) (%(x,y) (u,v). (x+u, y+v)) plus_int
   208 
   209   - transfer method: Replaces a subgoal on abstract types with an
   210     equivalent subgoal on the corresponding raw types. Constants are
   211     replaced with corresponding ones according to the transfer rules.
   212     Goals are generalized over all free variables by default; this is
   213     necessary for variables whose types changes, but can be overridden
   214     for specific variables with e.g. 'transfer fixing: x y z'.  The
   215     variant transfer' method allows replacing a subgoal with one that
   216     is logically stronger (rather than equivalent).
   217 
   218   - relator_eq attribute: Collects identity laws for relators of
   219     various type constructors, e.g. "list_all2 (op =) = (op =)".  The
   220     transfer method uses these lemmas to infer transfer rules for
   221     non-polymorphic constants on the fly.
   222 
   223   - transfer_prover method: Assists with proving a transfer rule for a
   224     new constant, provided the constant is defined in terms of other
   225     constants that already have transfer rules. It should be applied
   226     after unfolding the constant definitions.
   227 
   228   - HOL/ex/Transfer_Int_Nat.thy: Example theory demonstrating transfer
   229     from type nat to type int.
   230 
   231 * New Lifting package:
   232 
   233   - lift_definition command: Defines operations on an abstract type in
   234     terms of a corresponding operation on a representation
   235     type.  Example syntax:
   236 
   237     lift_definition dlist_insert :: "'a => 'a dlist => 'a dlist"
   238       is List.insert
   239 
   240     Users must discharge a respectfulness proof obligation when each
   241     constant is defined. (For a type copy, i.e. a typedef with UNIV,
   242     the proof is discharged automatically.) The obligation is
   243     presented in a user-friendly, readable form; a respectfulness
   244     theorem in the standard format and a transfer rule are generated
   245     by the package.
   246 
   247   - Integration with code_abstype: For typedefs (e.g. subtypes
   248     corresponding to a datatype invariant, such as dlist),
   249     lift_definition generates a code certificate theorem and sets up
   250     code generation for each constant.
   251 
   252   - setup_lifting command: Sets up the Lifting package to work with a
   253     user-defined type. The user must provide either a quotient theorem
   254     or a type_definition theorem.  The package configures transfer
   255     rules for equality and quantifiers on the type, and sets up the
   256     lift_definition command to work with the type.
   257 
   258   - Usage examples: See Quotient_Examples/Lift_DList.thy,
   259     Quotient_Examples/Lift_RBT.thy, Word/Word.thy and
   260     Library/Float.thy.
   261 
   262 * Quotient package:
   263 
   264   - The 'quotient_type' command now supports a 'morphisms' option with
   265     rep and abs functions, similar to typedef.
   266 
   267   - 'quotient_type' sets up new types to work with the Lifting and
   268     Transfer packages, as with 'setup_lifting'.
   269 
   270   - The 'quotient_definition' command now requires the user to prove a
   271     respectfulness property at the point where the constant is
   272     defined, similar to lift_definition; INCOMPATIBILITY.
   273 
   274   - Renamed predicate 'Quotient' to 'Quotient3', and renamed theorems
   275     accordingly, INCOMPATIBILITY.
   276 
   277 * New diagnostic command 'find_unused_assms' to find potentially
   278 superfluous assumptions in theorems using Quickcheck.
   279 
   280 * Quickcheck:
   281 
   282   - Quickcheck returns variable assignments as counterexamples, which
   283     allows to reveal the underspecification of functions under test.
   284     For example, refuting "hd xs = x", it presents the variable
   285     assignment xs = [] and x = a1 as a counterexample, assuming that
   286     any property is false whenever "hd []" occurs in it.
   287 
   288     These counterexample are marked as potentially spurious, as
   289     Quickcheck also returns "xs = []" as a counterexample to the
   290     obvious theorem "hd xs = hd xs".
   291 
   292     After finding a potentially spurious counterexample, Quickcheck
   293     continues searching for genuine ones.
   294 
   295     By default, Quickcheck shows potentially spurious and genuine
   296     counterexamples. The option "genuine_only" sets quickcheck to only
   297     show genuine counterexamples.
   298 
   299   - The command 'quickcheck_generator' creates random and exhaustive
   300     value generators for a given type and operations.
   301 
   302     It generates values by using the operations as if they were
   303     constructors of that type.
   304 
   305   - Support for multisets.
   306 
   307   - Added "use_subtype" options.
   308 
   309   - Added "quickcheck_locale" configuration to specify how to process
   310     conjectures in a locale context.
   311 
   312 * Nitpick:
   313   - Fixed infinite loop caused by the 'peephole_optim' option and
   314     affecting 'rat' and 'real'.
   315 
   316 * Sledgehammer:
   317   - Integrated more tightly with SPASS, as described in the ITP 2012
   318     paper "More SPASS with Isabelle".
   319   - Made it try "smt" as a fallback if "metis" fails or times out.
   320   - Added support for the following provers: Alt-Ergo (via Why3 and
   321     TFF1), iProver, iProver-Eq.
   322   - Replaced remote E-SInE with remote Satallax in the default setup.
   323   - Sped up the minimizer.
   324   - Added "lam_trans", "uncurry_aliases", and "minimize" options.
   325   - Renamed "slicing" ("no_slicing") option to "slice" ("dont_slice").
   326   - Renamed "sound" option to "strict".
   327 
   328 * Metis:
   329   - Added possibility to specify lambda translations scheme as a
   330     parenthesized argument (e.g., "by (metis (lifting) ...)").
   331 
   332 * SMT:
   333   - Renamed "smt_fixed" option to "smt_read_only_certificates".
   334 
   335 * Command 'try0':
   336   - Renamed from 'try_methods'. INCOMPATIBILITY.
   337 
   338 * New "eventually_elim" method as a generalized variant of the
   339 eventually_elim* rules. Supports structured proofs.
   340 
   341 * Typedef with implicit set definition is considered legacy.  Use
   342 "typedef (open)" form instead, which will eventually become the
   343 default.
   344 
   345 * More default pred/set conversions on a couple of relation operations
   346 and predicates.  Added powers of predicate relations.  Consolidation
   347 of some relation theorems:
   348 
   349   converse_def ~> converse_unfold
   350   rel_comp_def ~> relcomp_unfold
   351   symp_def ~> (modified, use symp_def and sym_def instead)
   352   transp_def ~> transp_trans
   353   Domain_def ~> Domain_unfold
   354   Range_def ~> Domain_converse [symmetric]
   355 
   356 Generalized theorems INF_INT_eq, INF_INT_eq2, SUP_UN_eq, SUP_UN_eq2.
   357 
   358 See theory "Relation" for examples for making use of pred/set
   359 conversions by means of attributes "to_set" and "to_pred".
   360 
   361 INCOMPATIBILITY.
   362 
   363 * Renamed facts about the power operation on relations, i.e., relpow
   364 to match the constant's name:
   365 
   366   rel_pow_1 ~> relpow_1
   367   rel_pow_0_I ~> relpow_0_I
   368   rel_pow_Suc_I ~> relpow_Suc_I
   369   rel_pow_Suc_I2 ~> relpow_Suc_I2
   370   rel_pow_0_E ~> relpow_0_E
   371   rel_pow_Suc_E ~> relpow_Suc_E
   372   rel_pow_E ~> relpow_E
   373   rel_pow_Suc_D2 ~> relpow_Suc_D2
   374   rel_pow_Suc_E2 ~> relpow_Suc_E2
   375   rel_pow_Suc_D2' ~> relpow_Suc_D2'
   376   rel_pow_E2 ~> relpow_E2
   377   rel_pow_add ~> relpow_add
   378   rel_pow_commute ~> relpow
   379   rel_pow_empty ~> relpow_empty:
   380   rtrancl_imp_UN_rel_pow ~> rtrancl_imp_UN_relpow
   381   rel_pow_imp_rtrancl ~> relpow_imp_rtrancl
   382   rtrancl_is_UN_rel_pow ~> rtrancl_is_UN_relpow
   383   rtrancl_imp_rel_pow ~> rtrancl_imp_relpow
   384   rel_pow_fun_conv ~> relpow_fun_conv
   385   rel_pow_finite_bounded1 ~> relpow_finite_bounded1
   386   rel_pow_finite_bounded ~> relpow_finite_bounded
   387   rtrancl_finite_eq_rel_pow ~> rtrancl_finite_eq_relpow
   388   trancl_finite_eq_rel_pow ~> trancl_finite_eq_relpow
   389   single_valued_rel_pow ~> single_valued_relpow
   390 
   391 INCOMPATIBILITY.
   392 
   393 * Theory Relation: Consolidated constant name for relation composition
   394 and corresponding theorem names:
   395 
   396   - Renamed constant rel_comp to relcomp.
   397 
   398   - Dropped abbreviation pred_comp. Use relcompp instead.
   399 
   400   - Renamed theorems:
   401 
   402     rel_compI ~> relcompI
   403     rel_compEpair ~> relcompEpair
   404     rel_compE ~> relcompE
   405     pred_comp_rel_comp_eq ~> relcompp_relcomp_eq
   406     rel_comp_empty1 ~> relcomp_empty1
   407     rel_comp_mono ~> relcomp_mono
   408     rel_comp_subset_Sigma ~> relcomp_subset_Sigma
   409     rel_comp_distrib ~> relcomp_distrib
   410     rel_comp_distrib2 ~> relcomp_distrib2
   411     rel_comp_UNION_distrib ~> relcomp_UNION_distrib
   412     rel_comp_UNION_distrib2 ~> relcomp_UNION_distrib2
   413     single_valued_rel_comp ~> single_valued_relcomp
   414     rel_comp_def ~> relcomp_unfold
   415     converse_rel_comp ~> converse_relcomp
   416     pred_compI ~> relcomppI
   417     pred_compE ~> relcomppE
   418     pred_comp_bot1 ~> relcompp_bot1
   419     pred_comp_bot2 ~> relcompp_bot2
   420     transp_pred_comp_less_eq ~> transp_relcompp_less_eq
   421     pred_comp_mono ~> relcompp_mono
   422     pred_comp_distrib ~> relcompp_distrib
   423     pred_comp_distrib2 ~> relcompp_distrib2
   424     converse_pred_comp ~> converse_relcompp
   425 
   426     finite_rel_comp ~> finite_relcomp
   427 
   428     set_rel_comp ~> set_relcomp
   429 
   430 INCOMPATIBILITY.
   431 
   432 * Theory Divides: Discontinued redundant theorems about div and mod.
   433 INCOMPATIBILITY, use the corresponding generic theorems instead.
   434 
   435   DIVISION_BY_ZERO ~> div_by_0, mod_by_0
   436   zdiv_self ~> div_self
   437   zmod_self ~> mod_self
   438   zdiv_zero ~> div_0
   439   zmod_zero ~> mod_0
   440   zdiv_zmod_equality ~> div_mod_equality2
   441   zdiv_zmod_equality2 ~> div_mod_equality
   442   zmod_zdiv_trivial ~> mod_div_trivial
   443   zdiv_zminus_zminus ~> div_minus_minus
   444   zmod_zminus_zminus ~> mod_minus_minus
   445   zdiv_zminus2 ~> div_minus_right
   446   zmod_zminus2 ~> mod_minus_right
   447   zdiv_minus1_right ~> div_minus1_right
   448   zmod_minus1_right ~> mod_minus1_right
   449   zdvd_mult_div_cancel ~> dvd_mult_div_cancel
   450   zmod_zmult1_eq ~> mod_mult_right_eq
   451   zpower_zmod ~> power_mod
   452   zdvd_zmod ~> dvd_mod
   453   zdvd_zmod_imp_zdvd ~> dvd_mod_imp_dvd
   454   mod_mult_distrib ~> mult_mod_left
   455   mod_mult_distrib2 ~> mult_mod_right
   456 
   457 * Removed redundant theorems nat_mult_2 and nat_mult_2_right; use
   458 generic mult_2 and mult_2_right instead. INCOMPATIBILITY.
   459 
   460 * Finite_Set.fold now qualified.  INCOMPATIBILITY.
   461 
   462 * Consolidated theorem names concerning fold combinators:
   463 
   464   inf_INFI_fold_inf ~> inf_INF_fold_inf
   465   sup_SUPR_fold_sup ~> sup_SUP_fold_sup
   466   INFI_fold_inf ~> INF_fold_inf
   467   SUPR_fold_sup ~> SUP_fold_sup
   468   union_set ~> union_set_fold
   469   minus_set ~> minus_set_fold
   470   INFI_set_fold ~> INF_set_fold
   471   SUPR_set_fold ~> SUP_set_fold
   472   INF_code ~> INF_set_foldr
   473   SUP_code ~> SUP_set_foldr
   474   foldr.simps ~> foldr.simps (in point-free formulation)
   475   foldr_fold_rev ~> foldr_conv_fold
   476   foldl_fold ~> foldl_conv_fold
   477   foldr_foldr ~> foldr_conv_foldl
   478   foldl_foldr ~> foldl_conv_foldr
   479   fold_set_remdups ~> fold_set_fold_remdups
   480   fold_set ~> fold_set_fold
   481   fold1_set ~> fold1_set_fold
   482 
   483 INCOMPATIBILITY.
   484 
   485 * Dropped rarely useful theorems concerning fold combinators:
   486 foldl_apply, foldl_fun_comm, foldl_rev, fold_weak_invariant,
   487 rev_foldl_cons, fold_set_remdups, fold_set, fold_set1,
   488 concat_conv_foldl, foldl_weak_invariant, foldl_invariant,
   489 foldr_invariant, foldl_absorb0, foldl_foldr1_lemma, foldl_foldr1,
   490 listsum_conv_fold, listsum_foldl, sort_foldl_insort, foldl_assoc,
   491 foldr_conv_foldl, start_le_sum, elem_le_sum, sum_eq_0_conv.
   492 INCOMPATIBILITY.  For the common phrases "%xs. List.foldr plus xs 0"
   493 and "List.foldl plus 0", prefer "List.listsum".  Otherwise it can be
   494 useful to boil down "List.foldr" and "List.foldl" to "List.fold" by
   495 unfolding "foldr_conv_fold" and "foldl_conv_fold".
   496 
   497 * Dropped lemmas minus_set_foldr, union_set_foldr, union_coset_foldr,
   498 inter_coset_foldr, Inf_fin_set_foldr, Sup_fin_set_foldr,
   499 Min_fin_set_foldr, Max_fin_set_foldr, Inf_set_foldr, Sup_set_foldr,
   500 INF_set_foldr, SUP_set_foldr.  INCOMPATIBILITY.  Prefer corresponding
   501 lemmas over fold rather than foldr, or make use of lemmas
   502 fold_conv_foldr and fold_rev.
   503 
   504 * Congruence rules Option.map_cong and Option.bind_cong for recursion
   505 through option types.
   506 
   507 * Concrete syntax for case expressions includes constraints for source
   508 positions, and thus produces Prover IDE markup for its bindings.
   509 INCOMPATIBILITY for old-style syntax translations that augment the
   510 pattern notation; e.g. see src/HOL/HOLCF/One.thy for translations of
   511 one_case.
   512 
   513 * Discontinued configuration option "syntax_positions": atomic terms
   514 in parse trees are always annotated by position constraints.
   515 
   516 * HOL/Library/Set_Algebras.thy: Addition and multiplication on sets
   517 are expressed via type classes again. The special syntax
   518 \<oplus>/\<otimes> has been replaced by plain +/*. Removed constant
   519 setsum_set, which is now subsumed by Big_Operators.setsum.
   520 INCOMPATIBILITY.
   521 
   522 * New theory HOL/Library/DAList provides an abstract type for
   523 association lists with distinct keys.
   524 
   525 * 'datatype' specifications allow explicit sort constraints.
   526 
   527 * Theory HOL/Library/Diagonalize has been removed. INCOMPATIBILITY,
   528 use theory HOL/Library/Nat_Bijection instead.
   529 
   530 * Theory HOL/Library/RBT_Impl: Backing implementation of red-black
   531 trees is now inside a type class context.  Names of affected
   532 operations and lemmas have been prefixed by rbt_.  INCOMPATIBILITY for
   533 theories working directly with raw red-black trees, adapt the names as
   534 follows:
   535 
   536   Operations:
   537   bulkload -> rbt_bulkload
   538   del_from_left -> rbt_del_from_left
   539   del_from_right -> rbt_del_from_right
   540   del -> rbt_del
   541   delete -> rbt_delete
   542   ins -> rbt_ins
   543   insert -> rbt_insert
   544   insertw -> rbt_insert_with
   545   insert_with_key -> rbt_insert_with_key
   546   map_entry -> rbt_map_entry
   547   lookup -> rbt_lookup
   548   sorted -> rbt_sorted
   549   tree_greater -> rbt_greater
   550   tree_less -> rbt_less
   551   tree_less_symbol -> rbt_less_symbol
   552   union -> rbt_union
   553   union_with -> rbt_union_with
   554   union_with_key -> rbt_union_with_key
   555 
   556   Lemmas:
   557   balance_left_sorted -> balance_left_rbt_sorted
   558   balance_left_tree_greater -> balance_left_rbt_greater
   559   balance_left_tree_less -> balance_left_rbt_less
   560   balance_right_sorted -> balance_right_rbt_sorted
   561   balance_right_tree_greater -> balance_right_rbt_greater
   562   balance_right_tree_less -> balance_right_rbt_less
   563   balance_sorted -> balance_rbt_sorted
   564   balance_tree_greater -> balance_rbt_greater
   565   balance_tree_less -> balance_rbt_less
   566   bulkload_is_rbt -> rbt_bulkload_is_rbt
   567   combine_sorted -> combine_rbt_sorted
   568   combine_tree_greater -> combine_rbt_greater
   569   combine_tree_less -> combine_rbt_less
   570   delete_in_tree -> rbt_delete_in_tree
   571   delete_is_rbt -> rbt_delete_is_rbt
   572   del_from_left_tree_greater -> rbt_del_from_left_rbt_greater
   573   del_from_left_tree_less -> rbt_del_from_left_rbt_less
   574   del_from_right_tree_greater -> rbt_del_from_right_rbt_greater
   575   del_from_right_tree_less -> rbt_del_from_right_rbt_less
   576   del_in_tree -> rbt_del_in_tree
   577   del_inv1_inv2 -> rbt_del_inv1_inv2
   578   del_sorted -> rbt_del_rbt_sorted
   579   del_tree_greater -> rbt_del_rbt_greater
   580   del_tree_less -> rbt_del_rbt_less
   581   dom_lookup_Branch -> dom_rbt_lookup_Branch
   582   entries_lookup -> entries_rbt_lookup
   583   finite_dom_lookup -> finite_dom_rbt_lookup
   584   insert_sorted -> rbt_insert_rbt_sorted
   585   insertw_is_rbt -> rbt_insertw_is_rbt
   586   insertwk_is_rbt -> rbt_insertwk_is_rbt
   587   insertwk_sorted -> rbt_insertwk_rbt_sorted
   588   insertw_sorted -> rbt_insertw_rbt_sorted
   589   ins_sorted -> ins_rbt_sorted
   590   ins_tree_greater -> ins_rbt_greater
   591   ins_tree_less -> ins_rbt_less
   592   is_rbt_sorted -> is_rbt_rbt_sorted
   593   lookup_balance -> rbt_lookup_balance
   594   lookup_bulkload -> rbt_lookup_rbt_bulkload
   595   lookup_delete -> rbt_lookup_rbt_delete
   596   lookup_Empty -> rbt_lookup_Empty
   597   lookup_from_in_tree -> rbt_lookup_from_in_tree
   598   lookup_in_tree -> rbt_lookup_in_tree
   599   lookup_ins -> rbt_lookup_ins
   600   lookup_insert -> rbt_lookup_rbt_insert
   601   lookup_insertw -> rbt_lookup_rbt_insertw
   602   lookup_insertwk -> rbt_lookup_rbt_insertwk
   603   lookup_keys -> rbt_lookup_keys
   604   lookup_map -> rbt_lookup_map
   605   lookup_map_entry -> rbt_lookup_rbt_map_entry
   606   lookup_tree_greater -> rbt_lookup_rbt_greater
   607   lookup_tree_less -> rbt_lookup_rbt_less
   608   lookup_union -> rbt_lookup_rbt_union
   609   map_entry_color_of -> rbt_map_entry_color_of
   610   map_entry_inv1 -> rbt_map_entry_inv1
   611   map_entry_inv2 -> rbt_map_entry_inv2
   612   map_entry_is_rbt -> rbt_map_entry_is_rbt
   613   map_entry_sorted -> rbt_map_entry_rbt_sorted
   614   map_entry_tree_greater -> rbt_map_entry_rbt_greater
   615   map_entry_tree_less -> rbt_map_entry_rbt_less
   616   map_tree_greater -> map_rbt_greater
   617   map_tree_less -> map_rbt_less
   618   map_sorted -> map_rbt_sorted
   619   paint_sorted -> paint_rbt_sorted
   620   paint_lookup -> paint_rbt_lookup
   621   paint_tree_greater -> paint_rbt_greater
   622   paint_tree_less -> paint_rbt_less
   623   sorted_entries -> rbt_sorted_entries
   624   tree_greater_eq_trans -> rbt_greater_eq_trans
   625   tree_greater_nit -> rbt_greater_nit
   626   tree_greater_prop -> rbt_greater_prop
   627   tree_greater_simps -> rbt_greater_simps
   628   tree_greater_trans -> rbt_greater_trans
   629   tree_less_eq_trans -> rbt_less_eq_trans
   630   tree_less_nit -> rbt_less_nit
   631   tree_less_prop -> rbt_less_prop
   632   tree_less_simps -> rbt_less_simps
   633   tree_less_trans -> rbt_less_trans
   634   tree_ord_props -> rbt_ord_props
   635   union_Branch -> rbt_union_Branch
   636   union_is_rbt -> rbt_union_is_rbt
   637   unionw_is_rbt -> rbt_unionw_is_rbt
   638   unionwk_is_rbt -> rbt_unionwk_is_rbt
   639   unionwk_sorted -> rbt_unionwk_rbt_sorted
   640 
   641 * Theory HOL/Library/Float: Floating point numbers are now defined as
   642 a subset of the real numbers.  All operations are defined using the
   643 lifing-framework and proofs use the transfer method.  INCOMPATIBILITY.
   644 
   645   Changed Operations:
   646   float_abs -> abs
   647   float_nprt -> nprt
   648   float_pprt -> pprt
   649   pow2 -> use powr
   650   round_down -> float_round_down
   651   round_up -> float_round_up
   652   scale -> exponent
   653 
   654   Removed Operations:
   655   ceiling_fl, lb_mult, lb_mod, ub_mult, ub_mod
   656 
   657   Renamed Lemmas:
   658   abs_float_def -> Float.compute_float_abs
   659   bitlen_ge0 -> bitlen_nonneg
   660   bitlen.simps -> Float.compute_bitlen
   661   float_components -> Float_mantissa_exponent
   662   float_divl.simps -> Float.compute_float_divl
   663   float_divr.simps -> Float.compute_float_divr
   664   float_eq_odd -> mult_powr_eq_mult_powr_iff
   665   float_power -> real_of_float_power
   666   lapprox_posrat_def -> Float.compute_lapprox_posrat
   667   lapprox_rat.simps -> Float.compute_lapprox_rat
   668   le_float_def' -> Float.compute_float_le
   669   le_float_def -> less_eq_float.rep_eq
   670   less_float_def' -> Float.compute_float_less
   671   less_float_def -> less_float.rep_eq
   672   normfloat_def -> Float.compute_normfloat
   673   normfloat_imp_odd_or_zero -> mantissa_not_dvd and mantissa_noteq_0
   674   normfloat -> normfloat_def
   675   normfloat_unique -> use normfloat_def
   676   number_of_float_Float -> Float.compute_float_numeral, Float.compute_float_neg_numeral
   677   one_float_def -> Float.compute_float_one
   678   plus_float_def -> Float.compute_float_plus
   679   rapprox_posrat_def -> Float.compute_rapprox_posrat
   680   rapprox_rat.simps -> Float.compute_rapprox_rat
   681   real_of_float_0 -> zero_float.rep_eq
   682   real_of_float_1 -> one_float.rep_eq
   683   real_of_float_abs -> abs_float.rep_eq
   684   real_of_float_add -> plus_float.rep_eq
   685   real_of_float_minus -> uminus_float.rep_eq
   686   real_of_float_mult -> times_float.rep_eq
   687   real_of_float_simp -> Float.rep_eq
   688   real_of_float_sub -> minus_float.rep_eq
   689   round_down.simps -> Float.compute_float_round_down
   690   round_up.simps -> Float.compute_float_round_up
   691   times_float_def -> Float.compute_float_times
   692   uminus_float_def -> Float.compute_float_uminus
   693   zero_float_def -> Float.compute_float_zero
   694 
   695   Lemmas not necessary anymore, use the transfer method:
   696   bitlen_B0, bitlen_B1, bitlen_ge1, bitlen_Min, bitlen_Pls, float_divl,
   697   float_divr, float_le_simp, float_less1_mantissa_bound,
   698   float_less_simp, float_less_zero, float_le_zero,
   699   float_pos_less1_e_neg, float_pos_m_pos, float_split, float_split2,
   700   floor_pos_exp, lapprox_posrat, lapprox_posrat_bottom, lapprox_rat,
   701   lapprox_rat_bottom, normalized_float, rapprox_posrat,
   702   rapprox_posrat_le1, rapprox_rat, real_of_float_ge0_exp,
   703   real_of_float_neg_exp, real_of_float_nge0_exp, round_down floor_fl,
   704   round_up, zero_le_float, zero_less_float
   705 
   706 * Session HOL-Word: Discontinued many redundant theorems specific to
   707 type 'a word. INCOMPATIBILITY, use the corresponding generic theorems
   708 instead.
   709 
   710   word_sub_alt ~> word_sub_wi
   711   word_add_alt ~> word_add_def
   712   word_mult_alt ~> word_mult_def
   713   word_minus_alt ~> word_minus_def
   714   word_0_alt ~> word_0_wi
   715   word_1_alt ~> word_1_wi
   716   word_add_0 ~> add_0_left
   717   word_add_0_right ~> add_0_right
   718   word_mult_1 ~> mult_1_left
   719   word_mult_1_right ~> mult_1_right
   720   word_add_commute ~> add_commute
   721   word_add_assoc ~> add_assoc
   722   word_add_left_commute ~> add_left_commute
   723   word_mult_commute ~> mult_commute
   724   word_mult_assoc ~> mult_assoc
   725   word_mult_left_commute ~> mult_left_commute
   726   word_left_distrib ~> left_distrib
   727   word_right_distrib ~> right_distrib
   728   word_left_minus ~> left_minus
   729   word_diff_0_right ~> diff_0_right
   730   word_diff_self ~> diff_self
   731   word_sub_def ~> diff_minus
   732   word_diff_minus ~> diff_minus
   733   word_add_ac ~> add_ac
   734   word_mult_ac ~> mult_ac
   735   word_plus_ac0 ~> add_0_left add_0_right add_ac
   736   word_times_ac1 ~> mult_1_left mult_1_right mult_ac
   737   word_order_trans ~> order_trans
   738   word_order_refl ~> order_refl
   739   word_order_antisym ~> order_antisym
   740   word_order_linear ~> linorder_linear
   741   lenw1_zero_neq_one ~> zero_neq_one
   742   word_number_of_eq ~> number_of_eq
   743   word_of_int_add_hom ~> wi_hom_add
   744   word_of_int_sub_hom ~> wi_hom_sub
   745   word_of_int_mult_hom ~> wi_hom_mult
   746   word_of_int_minus_hom ~> wi_hom_neg
   747   word_of_int_succ_hom ~> wi_hom_succ
   748   word_of_int_pred_hom ~> wi_hom_pred
   749   word_of_int_0_hom ~> word_0_wi
   750   word_of_int_1_hom ~> word_1_wi
   751 
   752 * Clarified attribute "mono_set": pure declaration without modifying
   753 the result of the fact expression.
   754 
   755 * "Transitive_Closure.ntrancl": bounded transitive closure on
   756 relations.
   757 
   758 * Constant "Set.not_member" now qualified.  INCOMPATIBILITY.
   759 
   760 * Theory Int: Discontinued many legacy theorems specific to type int.
   761 INCOMPATIBILITY, use the corresponding generic theorems instead.
   762 
   763   zminus_zminus ~> minus_minus
   764   zminus_0 ~> minus_zero
   765   zminus_zadd_distrib ~> minus_add_distrib
   766   zadd_commute ~> add_commute
   767   zadd_assoc ~> add_assoc
   768   zadd_left_commute ~> add_left_commute
   769   zadd_ac ~> add_ac
   770   zmult_ac ~> mult_ac
   771   zadd_0 ~> add_0_left
   772   zadd_0_right ~> add_0_right
   773   zadd_zminus_inverse2 ~> left_minus
   774   zmult_zminus ~> mult_minus_left
   775   zmult_commute ~> mult_commute
   776   zmult_assoc ~> mult_assoc
   777   zadd_zmult_distrib ~> left_distrib
   778   zadd_zmult_distrib2 ~> right_distrib
   779   zdiff_zmult_distrib ~> left_diff_distrib
   780   zdiff_zmult_distrib2 ~> right_diff_distrib
   781   zmult_1 ~> mult_1_left
   782   zmult_1_right ~> mult_1_right
   783   zle_refl ~> order_refl
   784   zle_trans ~> order_trans
   785   zle_antisym ~> order_antisym
   786   zle_linear ~> linorder_linear
   787   zless_linear ~> linorder_less_linear
   788   zadd_left_mono ~> add_left_mono
   789   zadd_strict_right_mono ~> add_strict_right_mono
   790   zadd_zless_mono ~> add_less_le_mono
   791   int_0_less_1 ~> zero_less_one
   792   int_0_neq_1 ~> zero_neq_one
   793   zless_le ~> less_le
   794   zpower_zadd_distrib ~> power_add
   795   zero_less_zpower_abs_iff ~> zero_less_power_abs_iff
   796   zero_le_zpower_abs ~> zero_le_power_abs
   797 
   798 * Theory Deriv: Renamed
   799 
   800   DERIV_nonneg_imp_nonincreasing ~> DERIV_nonneg_imp_nondecreasing
   801 
   802 * Theory Library/Multiset: Improved code generation of multisets.
   803 
   804 * Session HOL-Word: New proof method "word_bitwise" for splitting
   805 machine word equalities and inequalities into logical circuits,
   806 defined in HOL/Word/WordBitwise.thy.  Supports addition, subtraction,
   807 multiplication, shifting by constants, bitwise operators and numeric
   808 constants.  Requires fixed-length word types, not 'a word.  Solves
   809 many standard word identies outright and converts more into first
   810 order problems amenable to blast or similar.  See also examples in
   811 HOL/Word/Examples/WordExamples.thy.
   812 
   813 * Session HOL-Probability: Introduced the type "'a measure" to
   814 represent measures, this replaces the records 'a algebra and 'a
   815 measure_space.  The locales based on subset_class now have two
   816 locale-parameters the space \<Omega> and the set of measurables sets
   817 M.  The product of probability spaces uses now the same constant as
   818 the finite product of sigma-finite measure spaces "PiM :: ('i => 'a)
   819 measure".  Most constants are defined now outside of locales and gain
   820 an additional parameter, like null_sets, almost_eventually or \<mu>'.
   821 Measure space constructions for distributions and densities now got
   822 their own constants distr and density.  Instead of using locales to
   823 describe measure spaces with a finite space, the measure count_space
   824 and point_measure is introduced.  INCOMPATIBILITY.
   825 
   826   Renamed constants:
   827   measure -> emeasure
   828   finite_measure.\<mu>' -> measure
   829   product_algebra_generator -> prod_algebra
   830   product_prob_space.emb -> prod_emb
   831   product_prob_space.infprod_algebra -> PiM
   832 
   833   Removed locales:
   834   completeable_measure_space
   835   finite_measure_space
   836   finite_prob_space
   837   finite_product_finite_prob_space
   838   finite_product_sigma_algebra
   839   finite_sigma_algebra
   840   measure_space
   841   pair_finite_prob_space
   842   pair_finite_sigma_algebra
   843   pair_finite_space
   844   pair_sigma_algebra
   845   product_sigma_algebra
   846 
   847   Removed constants:
   848   conditional_space
   849   distribution -> use distr measure, or distributed predicate
   850   image_space
   851   joint_distribution -> use distr measure, or distributed predicate
   852   pair_measure_generator
   853   product_prob_space.infprod_algebra -> use PiM
   854   subvimage
   855 
   856   Replacement theorems:
   857   finite_additivity_sufficient -> ring_of_sets.countably_additiveI_finite
   858   finite_measure.empty_measure -> measure_empty
   859   finite_measure.finite_continuity_from_above -> finite_measure.finite_Lim_measure_decseq
   860   finite_measure.finite_continuity_from_below -> finite_measure.finite_Lim_measure_incseq
   861   finite_measure.finite_measure_countably_subadditive -> finite_measure.finite_measure_subadditive_countably
   862   finite_measure.finite_measure_eq -> finite_measure.emeasure_eq_measure
   863   finite_measure.finite_measure -> finite_measure.emeasure_finite
   864   finite_measure.finite_measure_finite_singleton -> finite_measure.finite_measure_eq_setsum_singleton
   865   finite_measure.positive_measure' -> measure_nonneg
   866   finite_measure.real_measure -> finite_measure.emeasure_real
   867   finite_product_prob_space.finite_measure_times -> finite_product_prob_space.finite_measure_PiM_emb
   868   finite_product_sigma_algebra.in_P -> sets_PiM_I_finite
   869   finite_product_sigma_algebra.P_empty -> space_PiM_empty, sets_PiM_empty
   870   information_space.conditional_entropy_eq -> information_space.conditional_entropy_simple_distributed
   871   information_space.conditional_entropy_positive -> information_space.conditional_entropy_nonneg_simple
   872   information_space.conditional_mutual_information_eq_mutual_information -> information_space.conditional_mutual_information_eq_mutual_information_simple
   873   information_space.conditional_mutual_information_generic_positive -> information_space.conditional_mutual_information_nonneg_simple
   874   information_space.conditional_mutual_information_positive -> information_space.conditional_mutual_information_nonneg_simple
   875   information_space.entropy_commute -> information_space.entropy_commute_simple
   876   information_space.entropy_eq -> information_space.entropy_simple_distributed
   877   information_space.entropy_generic_eq -> information_space.entropy_simple_distributed
   878   information_space.entropy_positive -> information_space.entropy_nonneg_simple
   879   information_space.entropy_uniform_max -> information_space.entropy_uniform
   880   information_space.KL_eq_0_imp -> information_space.KL_eq_0_iff_eq
   881   information_space.KL_eq_0 -> information_space.KL_same_eq_0
   882   information_space.KL_ge_0 -> information_space.KL_nonneg
   883   information_space.mutual_information_eq -> information_space.mutual_information_simple_distributed
   884   information_space.mutual_information_positive -> information_space.mutual_information_nonneg_simple
   885   Int_stable_cuboids -> Int_stable_atLeastAtMost
   886   Int_stable_product_algebra_generator -> positive_integral
   887   measure_preserving -> equality "distr M N f = N" "f : measurable M N"
   888   measure_space.additive -> emeasure_additive
   889   measure_space.AE_iff_null_set -> AE_iff_null
   890   measure_space.almost_everywhere_def -> eventually_ae_filter
   891   measure_space.almost_everywhere_vimage -> AE_distrD
   892   measure_space.continuity_from_above -> INF_emeasure_decseq
   893   measure_space.continuity_from_above_Lim -> Lim_emeasure_decseq
   894   measure_space.continuity_from_below_Lim -> Lim_emeasure_incseq
   895   measure_space.continuity_from_below -> SUP_emeasure_incseq
   896   measure_space_density -> emeasure_density
   897   measure_space.density_is_absolutely_continuous -> absolutely_continuousI_density
   898   measure_space.integrable_vimage -> integrable_distr
   899   measure_space.integral_translated_density -> integral_density
   900   measure_space.integral_vimage -> integral_distr
   901   measure_space.measure_additive -> plus_emeasure
   902   measure_space.measure_compl -> emeasure_compl
   903   measure_space.measure_countable_increasing -> emeasure_countable_increasing
   904   measure_space.measure_countably_subadditive -> emeasure_subadditive_countably
   905   measure_space.measure_decseq -> decseq_emeasure
   906   measure_space.measure_Diff -> emeasure_Diff
   907   measure_space.measure_Diff_null_set -> emeasure_Diff_null_set
   908   measure_space.measure_eq_0 -> emeasure_eq_0
   909   measure_space.measure_finitely_subadditive -> emeasure_subadditive_finite
   910   measure_space.measure_finite_singleton -> emeasure_eq_setsum_singleton
   911   measure_space.measure_incseq -> incseq_emeasure
   912   measure_space.measure_insert -> emeasure_insert
   913   measure_space.measure_mono -> emeasure_mono
   914   measure_space.measure_not_negative -> emeasure_not_MInf
   915   measure_space.measure_preserving_Int_stable -> measure_eqI_generator_eq
   916   measure_space.measure_setsum -> setsum_emeasure
   917   measure_space.measure_setsum_split -> setsum_emeasure_cover
   918   measure_space.measure_space_vimage -> emeasure_distr
   919   measure_space.measure_subadditive_finite -> emeasure_subadditive_finite
   920   measure_space.measure_subadditive -> subadditive
   921   measure_space.measure_top -> emeasure_space
   922   measure_space.measure_UN_eq_0 -> emeasure_UN_eq_0
   923   measure_space.measure_Un_null_set -> emeasure_Un_null_set
   924   measure_space.positive_integral_translated_density -> positive_integral_density
   925   measure_space.positive_integral_vimage -> positive_integral_distr
   926   measure_space.real_continuity_from_above -> Lim_measure_decseq
   927   measure_space.real_continuity_from_below -> Lim_measure_incseq
   928   measure_space.real_measure_countably_subadditive -> measure_subadditive_countably
   929   measure_space.real_measure_Diff -> measure_Diff
   930   measure_space.real_measure_finite_Union -> measure_finite_Union
   931   measure_space.real_measure_setsum_singleton -> measure_eq_setsum_singleton
   932   measure_space.real_measure_subadditive -> measure_subadditive
   933   measure_space.real_measure_Union -> measure_Union
   934   measure_space.real_measure_UNION -> measure_UNION
   935   measure_space.simple_function_vimage -> simple_function_comp
   936   measure_space.simple_integral_vimage -> simple_integral_distr
   937   measure_space.simple_integral_vimage -> simple_integral_distr
   938   measure_unique_Int_stable -> measure_eqI_generator_eq
   939   measure_unique_Int_stable_vimage -> measure_eqI_generator_eq
   940   pair_sigma_algebra.measurable_cut_fst -> sets_Pair1
   941   pair_sigma_algebra.measurable_cut_snd -> sets_Pair2
   942   pair_sigma_algebra.measurable_pair_image_fst -> measurable_Pair1
   943   pair_sigma_algebra.measurable_pair_image_snd -> measurable_Pair2
   944   pair_sigma_algebra.measurable_product_swap -> measurable_pair_swap_iff
   945   pair_sigma_algebra.pair_sigma_algebra_measurable -> measurable_pair_swap
   946   pair_sigma_algebra.pair_sigma_algebra_swap_measurable -> measurable_pair_swap'
   947   pair_sigma_algebra.sets_swap -> sets_pair_swap
   948   pair_sigma_finite.measure_cut_measurable_fst -> pair_sigma_finite.measurable_emeasure_Pair1
   949   pair_sigma_finite.measure_cut_measurable_snd -> pair_sigma_finite.measurable_emeasure_Pair2
   950   pair_sigma_finite.measure_preserving_swap -> pair_sigma_finite.distr_pair_swap
   951   pair_sigma_finite.pair_measure_alt2 -> pair_sigma_finite.emeasure_pair_measure_alt2
   952   pair_sigma_finite.pair_measure_alt -> pair_sigma_finite.emeasure_pair_measure_alt
   953   pair_sigma_finite.pair_measure_times -> pair_sigma_finite.emeasure_pair_measure_Times
   954   prob_space.indep_distribution_eq_measure -> prob_space.indep_vars_iff_distr_eq_PiM
   955   prob_space.indep_var_distributionD -> prob_space.indep_var_distribution_eq
   956   prob_space.measure_space_1 -> prob_space.emeasure_space_1
   957   prob_space.prob_space_vimage -> prob_space_distr
   958   prob_space.random_variable_restrict -> measurable_restrict
   959   prob_space_unique_Int_stable -> measure_eqI_prob_space
   960   product_algebraE -> prod_algebraE_all
   961   product_algebra_generator_der -> prod_algebra_eq_finite
   962   product_algebra_generator_into_space -> prod_algebra_sets_into_space
   963   product_algebraI -> sets_PiM_I_finite
   964   product_measure_exists -> product_sigma_finite.sigma_finite
   965   product_prob_space.finite_index_eq_finite_product -> product_prob_space.sets_PiM_generator
   966   product_prob_space.finite_measure_infprod_emb_Pi -> product_prob_space.measure_PiM_emb
   967   product_prob_space.infprod_spec -> product_prob_space.emeasure_PiM_emb_not_empty
   968   product_prob_space.measurable_component -> measurable_component_singleton
   969   product_prob_space.measurable_emb -> measurable_prod_emb
   970   product_prob_space.measurable_into_infprod_algebra -> measurable_PiM_single
   971   product_prob_space.measurable_singleton_infprod -> measurable_component_singleton
   972   product_prob_space.measure_emb -> emeasure_prod_emb
   973   product_prob_space.measure_preserving_restrict -> product_prob_space.distr_restrict
   974   product_sigma_algebra.product_algebra_into_space -> space_closed
   975   product_sigma_finite.measure_fold -> product_sigma_finite.distr_merge
   976   product_sigma_finite.measure_preserving_component_singelton -> product_sigma_finite.distr_singleton
   977   product_sigma_finite.measure_preserving_merge -> product_sigma_finite.distr_merge
   978   sequence_space.measure_infprod -> sequence_space.measure_PiM_countable
   979   sets_product_algebra -> sets_PiM
   980   sigma_algebra.measurable_sigma -> measurable_measure_of
   981   sigma_finite_measure.disjoint_sigma_finite -> sigma_finite_disjoint
   982   sigma_finite_measure.RN_deriv_vimage -> sigma_finite_measure.RN_deriv_distr
   983   sigma_product_algebra_sigma_eq -> sigma_prod_algebra_sigma_eq
   984   space_product_algebra -> space_PiM
   985 
   986 * HOL/TPTP: support to parse and import TPTP problems (all languages)
   987 into Isabelle/HOL.
   988 
   989 
   990 *** FOL ***
   991 
   992 * New "case_product" attribute (see HOL).
   993 
   994 
   995 *** ZF ***
   996 
   997 * Greater support for structured proofs involving induction or case
   998 analysis.
   999 
  1000 * Much greater use of mathematical symbols.
  1001 
  1002 * Removal of many ML theorem bindings.  INCOMPATIBILITY.
  1003 
  1004 
  1005 *** ML ***
  1006 
  1007 * Antiquotation @{keyword "name"} produces a parser for outer syntax
  1008 from a minor keyword introduced via theory header declaration.
  1009 
  1010 * Antiquotation @{command_spec "name"} produces the
  1011 Outer_Syntax.command_spec from a major keyword introduced via theory
  1012 header declaration; it can be passed to Outer_Syntax.command etc.
  1013 
  1014 * Local_Theory.define no longer hard-wires default theorem name
  1015 "foo_def", but retains the binding as given.  If that is Binding.empty
  1016 / Attrib.empty_binding, the result is not registered as user-level
  1017 fact.  The Local_Theory.define_internal variant allows to specify a
  1018 non-empty name (used for the foundation in the background theory),
  1019 while omitting the fact binding in the user-context.  Potential
  1020 INCOMPATIBILITY for derived definitional packages: need to specify
  1021 naming policy for primitive definitions more explicitly.
  1022 
  1023 * Renamed Thm.capply to Thm.apply, and Thm.cabs to Thm.lambda in
  1024 conformance with similar operations in structure Term and Logic.
  1025 
  1026 * Antiquotation @{attributes [...]} embeds attribute source
  1027 representation into the ML text, which is particularly useful with
  1028 declarations like Local_Theory.note.
  1029 
  1030 * Structure Proof_Context follows standard naming scheme.  Old
  1031 ProofContext has been discontinued.  INCOMPATIBILITY.
  1032 
  1033 * Refined Local_Theory.declaration {syntax, pervasive}, with subtle
  1034 change of semantics: update is applied to auxiliary local theory
  1035 context as well.
  1036 
  1037 * Modernized some old-style infix operations:
  1038 
  1039   addeqcongs    ~> Simplifier.add_eqcong
  1040   deleqcongs    ~> Simplifier.del_eqcong
  1041   addcongs      ~> Simplifier.add_cong
  1042   delcongs      ~> Simplifier.del_cong
  1043   setmksimps    ~> Simplifier.set_mksimps
  1044   setmkcong     ~> Simplifier.set_mkcong
  1045   setmksym      ~> Simplifier.set_mksym
  1046   setmkeqTrue   ~> Simplifier.set_mkeqTrue
  1047   settermless   ~> Simplifier.set_termless
  1048   setsubgoaler  ~> Simplifier.set_subgoaler
  1049   addsplits     ~> Splitter.add_split
  1050   delsplits     ~> Splitter.del_split
  1051 
  1052 
  1053 *** System ***
  1054 
  1055 * USER_HOME settings variable points to cross-platform user home
  1056 directory, which coincides with HOME on POSIX systems only.  Likewise,
  1057 the Isabelle path specification "~" now expands to $USER_HOME, instead
  1058 of former $HOME.  A different default for USER_HOME may be set
  1059 explicitly in shell environment, before Isabelle settings are
  1060 evaluated.  Minor INCOMPATIBILITY: need to adapt Isabelle path where
  1061 the generic user home was intended.
  1062 
  1063 * ISABELLE_HOME_WINDOWS refers to ISABELLE_HOME in windows file name
  1064 notation, which is useful for the jEdit file browser, for example.
  1065 
  1066 * ISABELLE_JDK_HOME settings variable points to JDK with javac and jar
  1067 (not just JRE).
  1068 
  1069 
  1070 
  1071 New in Isabelle2011-1 (October 2011)
  1072 ------------------------------------
  1073 
  1074 *** General ***
  1075 
  1076 * Improved Isabelle/jEdit Prover IDE (PIDE), which can be invoked as
  1077 "isabelle jedit" or "ISABELLE_HOME/Isabelle" on the command line.
  1078 
  1079   - Management of multiple theory files directly from the editor
  1080     buffer store -- bypassing the file-system (no requirement to save
  1081     files for checking).
  1082 
  1083   - Markup of formal entities within the text buffer, with semantic
  1084     highlighting, tooltips and hyperlinks to jump to defining source
  1085     positions.
  1086 
  1087   - Improved text rendering, with sub/superscripts in the source
  1088     buffer (including support for copy/paste wrt. output panel, HTML
  1089     theory output and other non-Isabelle text boxes).
  1090 
  1091   - Refined scheduling of proof checking and printing of results,
  1092     based on interactive editor view.  (Note: jEdit folding and
  1093     narrowing allows to restrict buffer perspectives explicitly.)
  1094 
  1095   - Reduced CPU performance requirements, usable on machines with few
  1096     cores.
  1097 
  1098   - Reduced memory requirements due to pruning of unused document
  1099     versions (garbage collection).
  1100 
  1101 See also ~~/src/Tools/jEdit/README.html for further information,
  1102 including some remaining limitations.
  1103 
  1104 * Theory loader: source files are exclusively located via the master
  1105 directory of each theory node (where the .thy file itself resides).
  1106 The global load path (such as src/HOL/Library) has been discontinued.
  1107 Note that the path element ~~ may be used to reference theories in the
  1108 Isabelle home folder -- for instance, "~~/src/HOL/Library/FuncSet".
  1109 INCOMPATIBILITY.
  1110 
  1111 * Theory loader: source files are identified by content via SHA1
  1112 digests.  Discontinued former path/modtime identification and optional
  1113 ISABELLE_FILE_IDENT plugin scripts.
  1114 
  1115 * Parallelization of nested Isar proofs is subject to
  1116 Goal.parallel_proofs_threshold (default 100).  See also isabelle
  1117 usedir option -Q.
  1118 
  1119 * Name space: former unsynchronized references are now proper
  1120 configuration options, with more conventional names:
  1121 
  1122   long_names   ~> names_long
  1123   short_names  ~> names_short
  1124   unique_names ~> names_unique
  1125 
  1126 Minor INCOMPATIBILITY, need to declare options in context like this:
  1127 
  1128   declare [[names_unique = false]]
  1129 
  1130 * Literal facts `prop` may contain dummy patterns, e.g. `_ = _`.  Note
  1131 that the result needs to be unique, which means fact specifications
  1132 may have to be refined after enriching a proof context.
  1133 
  1134 * Attribute "case_names" has been refined: the assumptions in each case
  1135 can be named now by following the case name with [name1 name2 ...].
  1136 
  1137 * Isabelle/Isar reference manual has been updated and extended:
  1138   - "Synopsis" provides a catalog of main Isar language concepts.
  1139   - Formal references in syntax diagrams, via @{rail} antiquotation.
  1140   - Updated material from classic "ref" manual, notably about
  1141     "Classical Reasoner".
  1142 
  1143 
  1144 *** HOL ***
  1145 
  1146 * Class bot and top require underlying partial order rather than
  1147 preorder: uniqueness of bot and top is guaranteed.  INCOMPATIBILITY.
  1148 
  1149 * Class complete_lattice: generalized a couple of lemmas from sets;
  1150 generalized theorems INF_cong and SUP_cong.  New type classes for
  1151 complete boolean algebras and complete linear orders.  Lemmas
  1152 Inf_less_iff, less_Sup_iff, INF_less_iff, less_SUP_iff now reside in
  1153 class complete_linorder.
  1154 
  1155 Changed proposition of lemmas Inf_bool_def, Sup_bool_def, Inf_fun_def,
  1156 Sup_fun_def, Inf_apply, Sup_apply.
  1157 
  1158 Removed redundant lemmas (the right hand side gives hints how to
  1159 replace them for (metis ...), or (simp only: ...) proofs):
  1160 
  1161   Inf_singleton ~> Inf_insert [where A="{}", unfolded Inf_empty inf_top_right]
  1162   Sup_singleton ~> Sup_insert [where A="{}", unfolded Sup_empty sup_bot_right]
  1163   Inf_binary ~> Inf_insert, Inf_empty, and inf_top_right
  1164   Sup_binary ~> Sup_insert, Sup_empty, and sup_bot_right
  1165   Int_eq_Inter ~> Inf_insert, Inf_empty, and inf_top_right
  1166   Un_eq_Union ~> Sup_insert, Sup_empty, and sup_bot_right
  1167   Inter_def ~> INF_def, image_def
  1168   Union_def ~> SUP_def, image_def
  1169   INT_eq ~> INF_def, and image_def
  1170   UN_eq ~> SUP_def, and image_def
  1171   INF_subset ~> INF_superset_mono [OF _ order_refl]
  1172 
  1173 More consistent and comprehensive names:
  1174 
  1175   INTER_eq_Inter_image ~> INF_def
  1176   UNION_eq_Union_image ~> SUP_def
  1177   INFI_def ~> INF_def
  1178   SUPR_def ~> SUP_def
  1179   INF_leI ~> INF_lower
  1180   INF_leI2 ~> INF_lower2
  1181   le_INFI ~> INF_greatest
  1182   le_SUPI ~> SUP_upper
  1183   le_SUPI2 ~> SUP_upper2
  1184   SUP_leI ~> SUP_least
  1185   INFI_bool_eq ~> INF_bool_eq
  1186   SUPR_bool_eq ~> SUP_bool_eq
  1187   INFI_apply ~> INF_apply
  1188   SUPR_apply ~> SUP_apply
  1189   INTER_def ~> INTER_eq
  1190   UNION_def ~> UNION_eq
  1191 
  1192 INCOMPATIBILITY.
  1193 
  1194 * Renamed theory Complete_Lattice to Complete_Lattices.
  1195 INCOMPATIBILITY.
  1196 
  1197 * Theory Complete_Lattices: lemmas Inf_eq_top_iff, INF_eq_top_iff,
  1198 INF_image, Inf_insert, INF_top, Inf_top_conv, INF_top_conv, SUP_bot,
  1199 Sup_bot_conv, SUP_bot_conv, Sup_eq_top_iff, SUP_eq_top_iff, SUP_image,
  1200 Sup_insert are now declared as [simp].  INCOMPATIBILITY.
  1201 
  1202 * Theory Lattice: lemmas compl_inf_bot, compl_le_comp_iff,
  1203 compl_sup_top, inf_idem, inf_left_idem, inf_sup_absorb, sup_idem,
  1204 sup_inf_absob, sup_left_idem are now declared as [simp].  Minor
  1205 INCOMPATIBILITY.
  1206 
  1207 * Added syntactic classes "inf" and "sup" for the respective
  1208 constants.  INCOMPATIBILITY: Changes in the argument order of the
  1209 (mostly internal) locale predicates for some derived classes.
  1210 
  1211 * Theorem collections ball_simps and bex_simps do not contain theorems
  1212 referring to UNION any longer; these have been moved to collection
  1213 UN_ball_bex_simps.  INCOMPATIBILITY.
  1214 
  1215 * Theory Archimedean_Field: floor now is defined as parameter of a
  1216 separate type class floor_ceiling.
  1217 
  1218 * Theory Finite_Set: more coherent development of fold_set locales:
  1219 
  1220     locale fun_left_comm ~> locale comp_fun_commute
  1221     locale fun_left_comm_idem ~> locale comp_fun_idem
  1222 
  1223 Both use point-free characterization; interpretation proofs may need
  1224 adjustment.  INCOMPATIBILITY.
  1225 
  1226 * Theory Limits: Type "'a net" has been renamed to "'a filter", in
  1227 accordance with standard mathematical terminology. INCOMPATIBILITY.
  1228 
  1229 * Theory Complex_Main: The locale interpretations for the
  1230 bounded_linear and bounded_bilinear locales have been removed, in
  1231 order to reduce the number of duplicate lemmas. Users must use the
  1232 original names for distributivity theorems, potential INCOMPATIBILITY.
  1233 
  1234   divide.add ~> add_divide_distrib
  1235   divide.diff ~> diff_divide_distrib
  1236   divide.setsum ~> setsum_divide_distrib
  1237   mult.add_right ~> right_distrib
  1238   mult.diff_right ~> right_diff_distrib
  1239   mult_right.setsum ~> setsum_right_distrib
  1240   mult_left.diff ~> left_diff_distrib
  1241 
  1242 * Theory Complex_Main: Several redundant theorems have been removed or
  1243 replaced by more general versions. INCOMPATIBILITY.
  1244 
  1245   real_diff_def ~> minus_real_def
  1246   real_divide_def ~> divide_real_def
  1247   real_less_def ~> less_le
  1248   real_abs_def ~> abs_real_def
  1249   real_sgn_def ~> sgn_real_def
  1250   real_mult_commute ~> mult_commute
  1251   real_mult_assoc ~> mult_assoc
  1252   real_mult_1 ~> mult_1_left
  1253   real_add_mult_distrib ~> left_distrib
  1254   real_zero_not_eq_one ~> zero_neq_one
  1255   real_mult_inverse_left ~> left_inverse
  1256   INVERSE_ZERO ~> inverse_zero
  1257   real_le_refl ~> order_refl
  1258   real_le_antisym ~> order_antisym
  1259   real_le_trans ~> order_trans
  1260   real_le_linear ~> linear
  1261   real_le_eq_diff ~> le_iff_diff_le_0
  1262   real_add_left_mono ~> add_left_mono
  1263   real_mult_order ~> mult_pos_pos
  1264   real_mult_less_mono2 ~> mult_strict_left_mono
  1265   real_of_int_real_of_nat ~> real_of_int_of_nat_eq
  1266   real_0_le_divide_iff ~> zero_le_divide_iff
  1267   realpow_two_disj ~> power2_eq_iff
  1268   real_squared_diff_one_factored ~> square_diff_one_factored
  1269   realpow_two_diff ~> square_diff_square_factored
  1270   reals_complete2 ~> complete_real
  1271   real_sum_squared_expand ~> power2_sum
  1272   exp_ln_eq ~> ln_unique
  1273   expi_add ~> exp_add
  1274   expi_zero ~> exp_zero
  1275   lemma_DERIV_subst ~> DERIV_cong
  1276   LIMSEQ_Zfun_iff ~> tendsto_Zfun_iff
  1277   LIMSEQ_const ~> tendsto_const
  1278   LIMSEQ_norm ~> tendsto_norm
  1279   LIMSEQ_add ~> tendsto_add
  1280   LIMSEQ_minus ~> tendsto_minus
  1281   LIMSEQ_minus_cancel ~> tendsto_minus_cancel
  1282   LIMSEQ_diff ~> tendsto_diff
  1283   bounded_linear.LIMSEQ ~> bounded_linear.tendsto
  1284   bounded_bilinear.LIMSEQ ~> bounded_bilinear.tendsto
  1285   LIMSEQ_mult ~> tendsto_mult
  1286   LIMSEQ_inverse ~> tendsto_inverse
  1287   LIMSEQ_divide ~> tendsto_divide
  1288   LIMSEQ_pow ~> tendsto_power
  1289   LIMSEQ_setsum ~> tendsto_setsum
  1290   LIMSEQ_setprod ~> tendsto_setprod
  1291   LIMSEQ_norm_zero ~> tendsto_norm_zero_iff
  1292   LIMSEQ_rabs_zero ~> tendsto_rabs_zero_iff
  1293   LIMSEQ_imp_rabs ~> tendsto_rabs
  1294   LIMSEQ_add_minus ~> tendsto_add [OF _ tendsto_minus]
  1295   LIMSEQ_add_const ~> tendsto_add [OF _ tendsto_const]
  1296   LIMSEQ_diff_const ~> tendsto_diff [OF _ tendsto_const]
  1297   LIMSEQ_Complex ~> tendsto_Complex
  1298   LIM_ident ~> tendsto_ident_at
  1299   LIM_const ~> tendsto_const
  1300   LIM_add ~> tendsto_add
  1301   LIM_add_zero ~> tendsto_add_zero
  1302   LIM_minus ~> tendsto_minus
  1303   LIM_diff ~> tendsto_diff
  1304   LIM_norm ~> tendsto_norm
  1305   LIM_norm_zero ~> tendsto_norm_zero
  1306   LIM_norm_zero_cancel ~> tendsto_norm_zero_cancel
  1307   LIM_norm_zero_iff ~> tendsto_norm_zero_iff
  1308   LIM_rabs ~> tendsto_rabs
  1309   LIM_rabs_zero ~> tendsto_rabs_zero
  1310   LIM_rabs_zero_cancel ~> tendsto_rabs_zero_cancel
  1311   LIM_rabs_zero_iff ~> tendsto_rabs_zero_iff
  1312   LIM_compose ~> tendsto_compose
  1313   LIM_mult ~> tendsto_mult
  1314   LIM_scaleR ~> tendsto_scaleR
  1315   LIM_of_real ~> tendsto_of_real
  1316   LIM_power ~> tendsto_power
  1317   LIM_inverse ~> tendsto_inverse
  1318   LIM_sgn ~> tendsto_sgn
  1319   isCont_LIM_compose ~> isCont_tendsto_compose
  1320   bounded_linear.LIM ~> bounded_linear.tendsto
  1321   bounded_linear.LIM_zero ~> bounded_linear.tendsto_zero
  1322   bounded_bilinear.LIM ~> bounded_bilinear.tendsto
  1323   bounded_bilinear.LIM_prod_zero ~> bounded_bilinear.tendsto_zero
  1324   bounded_bilinear.LIM_left_zero ~> bounded_bilinear.tendsto_left_zero
  1325   bounded_bilinear.LIM_right_zero ~> bounded_bilinear.tendsto_right_zero
  1326   LIM_inverse_fun ~> tendsto_inverse [OF tendsto_ident_at]
  1327 
  1328 * Theory Complex_Main: The definition of infinite series was
  1329 generalized.  Now it is defined on the type class {topological_space,
  1330 comm_monoid_add}.  Hence it is useable also for extended real numbers.
  1331 
  1332 * Theory Complex_Main: The complex exponential function "expi" is now
  1333 a type-constrained abbreviation for "exp :: complex => complex"; thus
  1334 several polymorphic lemmas about "exp" are now applicable to "expi".
  1335 
  1336 * Code generation:
  1337 
  1338   - Theory Library/Code_Char_ord provides native ordering of
  1339     characters in the target language.
  1340 
  1341   - Commands code_module and code_library are legacy, use export_code
  1342     instead.
  1343 
  1344   - Method "evaluation" is legacy, use method "eval" instead.
  1345 
  1346   - Legacy evaluator "SML" is deactivated by default.  May be
  1347     reactivated by the following theory command:
  1348 
  1349       setup {* Value.add_evaluator ("SML", Codegen.eval_term) *}
  1350 
  1351 * Declare ext [intro] by default.  Rare INCOMPATIBILITY.
  1352 
  1353 * New proof method "induction" that gives induction hypotheses the
  1354 name "IH", thus distinguishing them from further hypotheses that come
  1355 from rule induction.  The latter are still called "hyps".  Method
  1356 "induction" is a thin wrapper around "induct" and follows the same
  1357 syntax.
  1358 
  1359 * Method "fastsimp" has been renamed to "fastforce", but "fastsimp" is
  1360 still available as a legacy feature for some time.
  1361 
  1362 * Nitpick:
  1363   - Added "need" and "total_consts" options.
  1364   - Reintroduced "show_skolems" option by popular demand.
  1365   - Renamed attribute: nitpick_def ~> nitpick_unfold.
  1366     INCOMPATIBILITY.
  1367 
  1368 * Sledgehammer:
  1369   - Use quasi-sound (and efficient) translations by default.
  1370   - Added support for the following provers: E-ToFoF, LEO-II,
  1371     Satallax, SNARK, Waldmeister, and Z3 with TPTP syntax.
  1372   - Automatically preplay and minimize proofs before showing them if
  1373     this can be done within reasonable time.
  1374   - sledgehammer available_provers ~> sledgehammer supported_provers.
  1375     INCOMPATIBILITY.
  1376   - Added "preplay_timeout", "slicing", "type_enc", "sound",
  1377     "max_mono_iters", and "max_new_mono_instances" options.
  1378   - Removed "explicit_apply" and "full_types" options as well as "Full
  1379     Types" Proof General menu item. INCOMPATIBILITY.
  1380 
  1381 * Metis:
  1382   - Removed "metisF" -- use "metis" instead. INCOMPATIBILITY.
  1383   - Obsoleted "metisFT" -- use "metis (full_types)" instead.
  1384     INCOMPATIBILITY.
  1385 
  1386 * Command 'try':
  1387   - Renamed 'try_methods' and added "simp:", "intro:", "dest:", and
  1388     "elim:" options. INCOMPATIBILITY.
  1389   - Introduced 'try' that not only runs 'try_methods' but also
  1390     'solve_direct', 'sledgehammer', 'quickcheck', and 'nitpick'.
  1391 
  1392 * Quickcheck:
  1393   - Added "eval" option to evaluate terms for the found counterexample
  1394     (currently only supported by the default (exhaustive) tester).
  1395   - Added post-processing of terms to obtain readable counterexamples
  1396     (currently only supported by the default (exhaustive) tester).
  1397   - New counterexample generator quickcheck[narrowing] enables
  1398     narrowing-based testing.  Requires the Glasgow Haskell compiler
  1399     with its installation location defined in the Isabelle settings
  1400     environment as ISABELLE_GHC.
  1401   - Removed quickcheck tester "SML" based on the SML code generator
  1402     (formly in HOL/Library).
  1403 
  1404 * Function package: discontinued option "tailrec".  INCOMPATIBILITY,
  1405 use 'partial_function' instead.
  1406 
  1407 * Theory Library/Extended_Reals replaces now the positive extended
  1408 reals found in probability theory. This file is extended by
  1409 Multivariate_Analysis/Extended_Real_Limits.
  1410 
  1411 * Theory Library/Old_Recdef: old 'recdef' package has been moved here,
  1412 from where it must be imported explicitly if it is really required.
  1413 INCOMPATIBILITY.
  1414 
  1415 * Theory Library/Wfrec: well-founded recursion combinator "wfrec" has
  1416 been moved here.  INCOMPATIBILITY.
  1417 
  1418 * Theory Library/Saturated provides type of numbers with saturated
  1419 arithmetic.
  1420 
  1421 * Theory Library/Product_Lattice defines a pointwise ordering for the
  1422 product type 'a * 'b, and provides instance proofs for various order
  1423 and lattice type classes.
  1424 
  1425 * Theory Library/Countable now provides the "countable_datatype" proof
  1426 method for proving "countable" class instances for datatypes.
  1427 
  1428 * Theory Library/Cset_Monad allows do notation for computable sets
  1429 (cset) via the generic monad ad-hoc overloading facility.
  1430 
  1431 * Library: Theories of common data structures are split into theories
  1432 for implementation, an invariant-ensuring type, and connection to an
  1433 abstract type. INCOMPATIBILITY.
  1434 
  1435   - RBT is split into RBT and RBT_Mapping.
  1436   - AssocList is split and renamed into AList and AList_Mapping.
  1437   - DList is split into DList_Impl, DList, and DList_Cset.
  1438   - Cset is split into Cset and List_Cset.
  1439 
  1440 * Theory Library/Nat_Infinity has been renamed to
  1441 Library/Extended_Nat, with name changes of the following types and
  1442 constants:
  1443 
  1444   type inat   ~> type enat
  1445   Fin         ~> enat
  1446   Infty       ~> infinity (overloaded)
  1447   iSuc        ~> eSuc
  1448   the_Fin     ~> the_enat
  1449 
  1450 Every theorem name containing "inat", "Fin", "Infty", or "iSuc" has
  1451 been renamed accordingly. INCOMPATIBILITY.
  1452 
  1453 * Session Multivariate_Analysis: The euclidean_space type class now
  1454 fixes a constant "Basis :: 'a set" consisting of the standard
  1455 orthonormal basis for the type. Users now have the option of
  1456 quantifying over this set instead of using the "basis" function, e.g.
  1457 "ALL x:Basis. P x" vs "ALL i<DIM('a). P (basis i)".
  1458 
  1459 * Session Multivariate_Analysis: Type "('a, 'b) cart" has been renamed
  1460 to "('a, 'b) vec" (the syntax "'a ^ 'b" remains unaffected). Constants
  1461 "Cart_nth" and "Cart_lambda" have been respectively renamed to
  1462 "vec_nth" and "vec_lambda"; theorems mentioning those names have
  1463 changed to match. Definition theorems for overloaded constants now use
  1464 the standard "foo_vec_def" naming scheme. A few other theorems have
  1465 been renamed as follows (INCOMPATIBILITY):
  1466 
  1467   Cart_eq          ~> vec_eq_iff
  1468   dist_nth_le_cart ~> dist_vec_nth_le
  1469   tendsto_vector   ~> vec_tendstoI
  1470   Cauchy_vector    ~> vec_CauchyI
  1471 
  1472 * Session Multivariate_Analysis: Several duplicate theorems have been
  1473 removed, and other theorems have been renamed or replaced with more
  1474 general versions. INCOMPATIBILITY.
  1475 
  1476   finite_choice ~> finite_set_choice
  1477   eventually_conjI ~> eventually_conj
  1478   eventually_and ~> eventually_conj_iff
  1479   eventually_false ~> eventually_False
  1480   setsum_norm ~> norm_setsum
  1481   Lim_sequentially ~> LIMSEQ_def
  1482   Lim_ident_at ~> LIM_ident
  1483   Lim_const ~> tendsto_const
  1484   Lim_cmul ~> tendsto_scaleR [OF tendsto_const]
  1485   Lim_neg ~> tendsto_minus
  1486   Lim_add ~> tendsto_add
  1487   Lim_sub ~> tendsto_diff
  1488   Lim_mul ~> tendsto_scaleR
  1489   Lim_vmul ~> tendsto_scaleR [OF _ tendsto_const]
  1490   Lim_null_norm ~> tendsto_norm_zero_iff [symmetric]
  1491   Lim_linear ~> bounded_linear.tendsto
  1492   Lim_component ~> tendsto_euclidean_component
  1493   Lim_component_cart ~> tendsto_vec_nth
  1494   Lim_inner ~> tendsto_inner [OF tendsto_const]
  1495   dot_lsum ~> inner_setsum_left
  1496   dot_rsum ~> inner_setsum_right
  1497   continuous_cmul ~> continuous_scaleR [OF continuous_const]
  1498   continuous_neg ~> continuous_minus
  1499   continuous_sub ~> continuous_diff
  1500   continuous_vmul ~> continuous_scaleR [OF _ continuous_const]
  1501   continuous_mul ~> continuous_scaleR
  1502   continuous_inv ~> continuous_inverse
  1503   continuous_at_within_inv ~> continuous_at_within_inverse
  1504   continuous_at_inv ~> continuous_at_inverse
  1505   continuous_at_norm ~> continuous_norm [OF continuous_at_id]
  1506   continuous_at_infnorm ~> continuous_infnorm [OF continuous_at_id]
  1507   continuous_at_component ~> continuous_component [OF continuous_at_id]
  1508   continuous_on_neg ~> continuous_on_minus
  1509   continuous_on_sub ~> continuous_on_diff
  1510   continuous_on_cmul ~> continuous_on_scaleR [OF continuous_on_const]
  1511   continuous_on_vmul ~> continuous_on_scaleR [OF _ continuous_on_const]
  1512   continuous_on_mul ~> continuous_on_scaleR
  1513   continuous_on_mul_real ~> continuous_on_mult
  1514   continuous_on_inner ~> continuous_on_inner [OF continuous_on_const]
  1515   continuous_on_norm ~> continuous_on_norm [OF continuous_on_id]
  1516   continuous_on_inverse ~> continuous_on_inv
  1517   uniformly_continuous_on_neg ~> uniformly_continuous_on_minus
  1518   uniformly_continuous_on_sub ~> uniformly_continuous_on_diff
  1519   subset_interior ~> interior_mono
  1520   subset_closure ~> closure_mono
  1521   closure_univ ~> closure_UNIV
  1522   real_arch_lt ~> reals_Archimedean2
  1523   real_arch ~> reals_Archimedean3
  1524   real_abs_norm ~> abs_norm_cancel
  1525   real_abs_sub_norm ~> norm_triangle_ineq3
  1526   norm_cauchy_schwarz_abs ~> Cauchy_Schwarz_ineq2
  1527 
  1528 * Session HOL-Probability:
  1529   - Caratheodory's extension lemma is now proved for ring_of_sets.
  1530   - Infinite products of probability measures are now available.
  1531   - Sigma closure is independent, if the generator is independent
  1532   - Use extended reals instead of positive extended
  1533     reals. INCOMPATIBILITY.
  1534 
  1535 * Session HOLCF: Discontinued legacy theorem names, INCOMPATIBILITY.
  1536 
  1537   expand_fun_below ~> fun_below_iff
  1538   below_fun_ext ~> fun_belowI
  1539   expand_cfun_eq ~> cfun_eq_iff
  1540   ext_cfun ~> cfun_eqI
  1541   expand_cfun_below ~> cfun_below_iff
  1542   below_cfun_ext ~> cfun_belowI
  1543   monofun_fun_fun ~> fun_belowD
  1544   monofun_fun_arg ~> monofunE
  1545   monofun_lub_fun ~> adm_monofun [THEN admD]
  1546   cont_lub_fun ~> adm_cont [THEN admD]
  1547   cont2cont_Rep_CFun ~> cont2cont_APP
  1548   cont_Rep_CFun_app ~> cont_APP_app
  1549   cont_Rep_CFun_app_app ~> cont_APP_app_app
  1550   cont_cfun_fun ~> cont_Rep_cfun1 [THEN contE]
  1551   cont_cfun_arg ~> cont_Rep_cfun2 [THEN contE]
  1552   contlub_cfun ~> lub_APP [symmetric]
  1553   contlub_LAM ~> lub_LAM [symmetric]
  1554   thelubI ~> lub_eqI
  1555   UU_I ~> bottomI
  1556   lift_distinct1 ~> lift.distinct(1)
  1557   lift_distinct2 ~> lift.distinct(2)
  1558   Def_not_UU ~> lift.distinct(2)
  1559   Def_inject ~> lift.inject
  1560   below_UU_iff ~> below_bottom_iff
  1561   eq_UU_iff ~> eq_bottom_iff
  1562 
  1563 
  1564 *** Document preparation ***
  1565 
  1566 * Antiquotation @{rail} layouts railroad syntax diagrams, see also
  1567 isar-ref manual, both for description and actual application of the
  1568 same.
  1569 
  1570 * Antiquotation @{value} evaluates the given term and presents its
  1571 result.
  1572 
  1573 * Antiquotations: term style "isub" provides ad-hoc conversion of
  1574 variables x1, y23 into subscripted form x\<^isub>1,
  1575 y\<^isub>2\<^isub>3.
  1576 
  1577 * Predefined LaTeX macros for Isabelle symbols \<bind> and \<then>
  1578 (e.g. see ~~/src/HOL/Library/Monad_Syntax.thy).
  1579 
  1580 * Localized \isabellestyle switch can be used within blocks or groups
  1581 like this:
  1582 
  1583   \isabellestyle{it}  %preferred default
  1584   {\isabellestylett @{text "typewriter stuff"}}
  1585 
  1586 * Discontinued special treatment of hard tabulators.  Implicit
  1587 tab-width is now defined as 1.  Potential INCOMPATIBILITY for visual
  1588 layouts.
  1589 
  1590 
  1591 *** ML ***
  1592 
  1593 * The inner syntax of sort/type/term/prop supports inlined YXML
  1594 representations within quoted string tokens.  By encoding logical
  1595 entities via Term_XML (in ML or Scala) concrete syntax can be
  1596 bypassed, which is particularly useful for producing bits of text
  1597 under external program control.
  1598 
  1599 * Antiquotations for ML and document preparation are managed as theory
  1600 data, which requires explicit setup.
  1601 
  1602 * Isabelle_Process.is_active allows tools to check if the official
  1603 process wrapper is running (Isabelle/Scala/jEdit) or the old TTY loop
  1604 (better known as Proof General).
  1605 
  1606 * Structure Proof_Context follows standard naming scheme.  Old
  1607 ProofContext is still available for some time as legacy alias.
  1608 
  1609 * Structure Timing provides various operations for timing; supersedes
  1610 former start_timing/end_timing etc.
  1611 
  1612 * Path.print is the official way to show file-system paths to users
  1613 (including quotes etc.).
  1614 
  1615 * Inner syntax: identifiers in parse trees of generic categories
  1616 "logic", "aprop", "idt" etc. carry position information (disguised as
  1617 type constraints).  Occasional INCOMPATIBILITY with non-compliant
  1618 translations that choke on unexpected type constraints.  Positions can
  1619 be stripped in ML translations via Syntax.strip_positions /
  1620 Syntax.strip_positions_ast, or via the syntax constant
  1621 "_strip_positions" within parse trees.  As last resort, positions can
  1622 be disabled via the configuration option Syntax.positions, which is
  1623 called "syntax_positions" in Isar attribute syntax.
  1624 
  1625 * Discontinued special status of various ML structures that contribute
  1626 to structure Syntax (Ast, Lexicon, Mixfix, Parser, Printer etc.): less
  1627 pervasive content, no inclusion in structure Syntax.  INCOMPATIBILITY,
  1628 refer directly to Ast.Constant, Lexicon.is_identifier,
  1629 Syntax_Trans.mk_binder_tr etc.
  1630 
  1631 * Typed print translation: discontinued show_sorts argument, which is
  1632 already available via context of "advanced" translation.
  1633 
  1634 * Refined PARALLEL_GOALS tactical: degrades gracefully for schematic
  1635 goal states; body tactic needs to address all subgoals uniformly.
  1636 
  1637 * Slightly more special eq_list/eq_set, with shortcut involving
  1638 pointer equality (assumes that eq relation is reflexive).
  1639 
  1640 * Classical tactics use proper Proof.context instead of historic types
  1641 claset/clasimpset.  Old-style declarations like addIs, addEs, addDs
  1642 operate directly on Proof.context.  Raw type claset retains its use as
  1643 snapshot of the classical context, which can be recovered via
  1644 (put_claset HOL_cs) etc.  Type clasimpset has been discontinued.
  1645 INCOMPATIBILITY, classical tactics and derived proof methods require
  1646 proper Proof.context.
  1647 
  1648 
  1649 *** System ***
  1650 
  1651 * Discontinued support for Poly/ML 5.2, which was the last version
  1652 without proper multithreading and TimeLimit implementation.
  1653 
  1654 * Discontinued old lib/scripts/polyml-platform, which has been
  1655 obsolete since Isabelle2009-2.
  1656 
  1657 * Various optional external tools are referenced more robustly and
  1658 uniformly by explicit Isabelle settings as follows:
  1659 
  1660   ISABELLE_CSDP   (formerly CSDP_EXE)
  1661   ISABELLE_GHC    (formerly EXEC_GHC or GHC_PATH)
  1662   ISABELLE_OCAML  (formerly EXEC_OCAML)
  1663   ISABELLE_SWIPL  (formerly EXEC_SWIPL)
  1664   ISABELLE_YAP    (formerly EXEC_YAP)
  1665 
  1666 Note that automated detection from the file-system or search path has
  1667 been discontinued.  INCOMPATIBILITY.
  1668 
  1669 * Scala layer provides JVM method invocation service for static
  1670 methods of type (String)String, see Invoke_Scala.method in ML.  For
  1671 example:
  1672 
  1673   Invoke_Scala.method "java.lang.System.getProperty" "java.home"
  1674 
  1675 Together with YXML.string_of_body/parse_body and XML.Encode/Decode
  1676 this allows to pass structured values between ML and Scala.
  1677 
  1678 * The IsabelleText fonts includes some further glyphs to support the
  1679 Prover IDE.  Potential INCOMPATIBILITY: users who happen to have
  1680 installed a local copy (which is normally *not* required) need to
  1681 delete or update it from ~~/lib/fonts/.
  1682 
  1683 
  1684 
  1685 New in Isabelle2011 (January 2011)
  1686 ----------------------------------
  1687 
  1688 *** General ***
  1689 
  1690 * Experimental Prover IDE based on Isabelle/Scala and jEdit (see
  1691 src/Tools/jEdit).  This also serves as IDE for Isabelle/ML, with
  1692 useful tooltips and hyperlinks produced from its static analysis.  The
  1693 bundled component provides an executable Isabelle tool that can be run
  1694 like this:
  1695 
  1696   Isabelle2011/bin/isabelle jedit
  1697 
  1698 * Significantly improved Isabelle/Isar implementation manual.
  1699 
  1700 * System settings: ISABELLE_HOME_USER now includes ISABELLE_IDENTIFIER
  1701 (and thus refers to something like $HOME/.isabelle/Isabelle2011),
  1702 while the default heap location within that directory lacks that extra
  1703 suffix.  This isolates multiple Isabelle installations from each
  1704 other, avoiding problems with old settings in new versions.
  1705 INCOMPATIBILITY, need to copy/upgrade old user settings manually.
  1706 
  1707 * Source files are always encoded as UTF-8, instead of old-fashioned
  1708 ISO-Latin-1.  INCOMPATIBILITY.  Isabelle LaTeX documents might require
  1709 the following package declarations:
  1710 
  1711   \usepackage[utf8]{inputenc}
  1712   \usepackage{textcomp}
  1713 
  1714 * Explicit treatment of UTF-8 sequences as Isabelle symbols, such that
  1715 a Unicode character is treated as a single symbol, not a sequence of
  1716 non-ASCII bytes as before.  Since Isabelle/ML string literals may
  1717 contain symbols without further backslash escapes, Unicode can now be
  1718 used here as well.  Recall that Symbol.explode in ML provides a
  1719 consistent view on symbols, while raw explode (or String.explode)
  1720 merely give a byte-oriented representation.
  1721 
  1722 * Theory loader: source files are primarily located via the master
  1723 directory of each theory node (where the .thy file itself resides).
  1724 The global load path is still partially available as legacy feature.
  1725 Minor INCOMPATIBILITY due to subtle change in file lookup: use
  1726 explicit paths, relatively to the theory.
  1727 
  1728 * Special treatment of ML file names has been discontinued.
  1729 Historically, optional extensions .ML or .sml were added on demand --
  1730 at the cost of clarity of file dependencies.  Recall that Isabelle/ML
  1731 files exclusively use the .ML extension.  Minor INCOMPATIBILTY.
  1732 
  1733 * Various options that affect pretty printing etc. are now properly
  1734 handled within the context via configuration options, instead of
  1735 unsynchronized references or print modes.  There are both ML Config.T
  1736 entities and Isar declaration attributes to access these.
  1737 
  1738   ML (Config.T)                 Isar (attribute)
  1739 
  1740   eta_contract                  eta_contract
  1741   show_brackets                 show_brackets
  1742   show_sorts                    show_sorts
  1743   show_types                    show_types
  1744   show_question_marks           show_question_marks
  1745   show_consts                   show_consts
  1746   show_abbrevs                  show_abbrevs
  1747 
  1748   Syntax.ast_trace              syntax_ast_trace
  1749   Syntax.ast_stat               syntax_ast_stat
  1750   Syntax.ambiguity_level        syntax_ambiguity_level
  1751 
  1752   Goal_Display.goals_limit      goals_limit
  1753   Goal_Display.show_main_goal   show_main_goal
  1754 
  1755   Method.rule_trace             rule_trace
  1756 
  1757   Thy_Output.display            thy_output_display
  1758   Thy_Output.quotes             thy_output_quotes
  1759   Thy_Output.indent             thy_output_indent
  1760   Thy_Output.source             thy_output_source
  1761   Thy_Output.break              thy_output_break
  1762 
  1763 Note that corresponding "..._default" references in ML may only be
  1764 changed globally at the ROOT session setup, but *not* within a theory.
  1765 The option "show_abbrevs" supersedes the former print mode
  1766 "no_abbrevs" with inverted meaning.
  1767 
  1768 * More systematic naming of some configuration options.
  1769 INCOMPATIBILITY.
  1770 
  1771   trace_simp  ~>  simp_trace
  1772   debug_simp  ~>  simp_debug
  1773 
  1774 * Support for real valued configuration options, using simplistic
  1775 floating-point notation that coincides with the inner syntax for
  1776 float_token.
  1777 
  1778 * Support for real valued preferences (with approximative PGIP type):
  1779 front-ends need to accept "pgint" values in float notation.
  1780 INCOMPATIBILITY.
  1781 
  1782 * The IsabelleText font now includes Cyrillic, Hebrew, Arabic from
  1783 DejaVu Sans.
  1784 
  1785 * Discontinued support for Poly/ML 5.0 and 5.1 versions.
  1786 
  1787 
  1788 *** Pure ***
  1789 
  1790 * Command 'type_synonym' (with single argument) replaces somewhat
  1791 outdated 'types', which is still available as legacy feature for some
  1792 time.
  1793 
  1794 * Command 'nonterminal' (with 'and' separated list of arguments)
  1795 replaces somewhat outdated 'nonterminals'.  INCOMPATIBILITY.
  1796 
  1797 * Command 'notepad' replaces former 'example_proof' for
  1798 experimentation in Isar without any result.  INCOMPATIBILITY.
  1799 
  1800 * Locale interpretation commands 'interpret' and 'sublocale' accept
  1801 lists of equations to map definitions in a locale to appropriate
  1802 entities in the context of the interpretation.  The 'interpretation'
  1803 command already provided this functionality.
  1804 
  1805 * Diagnostic command 'print_dependencies' prints the locale instances
  1806 that would be activated if the specified expression was interpreted in
  1807 the current context.  Variant "print_dependencies!" assumes a context
  1808 without interpretations.
  1809 
  1810 * Diagnostic command 'print_interps' prints interpretations in proofs
  1811 in addition to interpretations in theories.
  1812 
  1813 * Discontinued obsolete 'global' and 'local' commands to manipulate
  1814 the theory name space.  Rare INCOMPATIBILITY.  The ML functions
  1815 Sign.root_path and Sign.local_path may be applied directly where this
  1816 feature is still required for historical reasons.
  1817 
  1818 * Discontinued obsolete 'constdefs' command.  INCOMPATIBILITY, use
  1819 'definition' instead.
  1820 
  1821 * The "prems" fact, which refers to the accidental collection of
  1822 foundational premises in the context, is now explicitly marked as
  1823 legacy feature and will be discontinued soon.  Consider using "assms"
  1824 of the head statement or reference facts by explicit names.
  1825 
  1826 * Document antiquotations @{class} and @{type} print classes and type
  1827 constructors.
  1828 
  1829 * Document antiquotation @{file} checks file/directory entries within
  1830 the local file system.
  1831 
  1832 
  1833 *** HOL ***
  1834 
  1835 * Coercive subtyping: functions can be declared as coercions and type
  1836 inference will add them as necessary upon input of a term.  Theory
  1837 Complex_Main declares real :: nat => real and real :: int => real as
  1838 coercions. A coercion function f is declared like this:
  1839 
  1840   declare [[coercion f]]
  1841 
  1842 To lift coercions through type constructors (e.g. from nat => real to
  1843 nat list => real list), map functions can be declared, e.g.
  1844 
  1845   declare [[coercion_map map]]
  1846 
  1847 Currently coercion inference is activated only in theories including
  1848 real numbers, i.e. descendants of Complex_Main.  This is controlled by
  1849 the configuration option "coercion_enabled", e.g. it can be enabled in
  1850 other theories like this:
  1851 
  1852   declare [[coercion_enabled]]
  1853 
  1854 * Command 'partial_function' provides basic support for recursive
  1855 function definitions over complete partial orders.  Concrete instances
  1856 are provided for i) the option type, ii) tail recursion on arbitrary
  1857 types, and iii) the heap monad of Imperative_HOL.  See
  1858 src/HOL/ex/Fundefs.thy and src/HOL/Imperative_HOL/ex/Linked_Lists.thy
  1859 for examples.
  1860 
  1861 * Function package: f.psimps rules are no longer implicitly declared
  1862 as [simp].  INCOMPATIBILITY.
  1863 
  1864 * Datatype package: theorems generated for executable equality (class
  1865 "eq") carry proper names and are treated as default code equations.
  1866 
  1867 * Inductive package: now offers command 'inductive_simps' to
  1868 automatically derive instantiated and simplified equations for
  1869 inductive predicates, similar to 'inductive_cases'.
  1870 
  1871 * Command 'enriched_type' allows to register properties of the
  1872 functorial structure of types.
  1873 
  1874 * Improved infrastructure for term evaluation using code generator
  1875 techniques, in particular static evaluation conversions.
  1876 
  1877 * Code generator: Scala (2.8 or higher) has been added to the target
  1878 languages.
  1879 
  1880 * Code generator: globbing constant expressions "*" and "Theory.*"
  1881 have been replaced by the more idiomatic "_" and "Theory._".
  1882 INCOMPATIBILITY.
  1883 
  1884 * Code generator: export_code without explicit file declaration prints
  1885 to standard output.  INCOMPATIBILITY.
  1886 
  1887 * Code generator: do not print function definitions for case
  1888 combinators any longer.
  1889 
  1890 * Code generator: simplification with rules determined with
  1891 src/Tools/Code/code_simp.ML and method "code_simp".
  1892 
  1893 * Code generator for records: more idiomatic representation of record
  1894 types.  Warning: records are not covered by ancient SML code
  1895 generation any longer.  INCOMPATIBILITY.  In cases of need, a suitable
  1896 rep_datatype declaration helps to succeed then:
  1897 
  1898   record 'a foo = ...
  1899   ...
  1900   rep_datatype foo_ext ...
  1901 
  1902 * Records: logical foundation type for records does not carry a
  1903 '_type' suffix any longer (obsolete due to authentic syntax).
  1904 INCOMPATIBILITY.
  1905 
  1906 * Quickcheck now by default uses exhaustive testing instead of random
  1907 testing.  Random testing can be invoked by "quickcheck [random]",
  1908 exhaustive testing by "quickcheck [exhaustive]".
  1909 
  1910 * Quickcheck instantiates polymorphic types with small finite
  1911 datatypes by default. This enables a simple execution mechanism to
  1912 handle quantifiers and function equality over the finite datatypes.
  1913 
  1914 * Quickcheck random generator has been renamed from "code" to
  1915 "random".  INCOMPATIBILITY.
  1916 
  1917 * Quickcheck now has a configurable time limit which is set to 30
  1918 seconds by default. This can be changed by adding [timeout = n] to the
  1919 quickcheck command. The time limit for Auto Quickcheck is still set
  1920 independently.
  1921 
  1922 * Quickcheck in locales considers interpretations of that locale for
  1923 counter example search.
  1924 
  1925 * Sledgehammer:
  1926   - Added "smt" and "remote_smt" provers based on the "smt" proof
  1927     method. See the Sledgehammer manual for details ("isabelle doc
  1928     sledgehammer").
  1929   - Renamed commands:
  1930     sledgehammer atp_info ~> sledgehammer running_provers
  1931     sledgehammer atp_kill ~> sledgehammer kill_provers
  1932     sledgehammer available_atps ~> sledgehammer available_provers
  1933     INCOMPATIBILITY.
  1934   - Renamed options:
  1935     sledgehammer [atps = ...] ~> sledgehammer [provers = ...]
  1936     sledgehammer [atp = ...] ~> sledgehammer [prover = ...]
  1937     sledgehammer [timeout = 77 s] ~> sledgehammer [timeout = 77]
  1938     (and "ms" and "min" are no longer supported)
  1939     INCOMPATIBILITY.
  1940 
  1941 * Nitpick:
  1942   - Renamed options:
  1943     nitpick [timeout = 77 s] ~> nitpick [timeout = 77]
  1944     nitpick [tac_timeout = 777 ms] ~> nitpick [tac_timeout = 0.777]
  1945     INCOMPATIBILITY.
  1946   - Added support for partial quotient types.
  1947   - Added local versions of the "Nitpick.register_xxx" functions.
  1948   - Added "whack" option.
  1949   - Allow registration of quotient types as codatatypes.
  1950   - Improved "merge_type_vars" option to merge more types.
  1951   - Removed unsound "fast_descrs" option.
  1952   - Added custom symmetry breaking for datatypes, making it possible to reach
  1953     higher cardinalities.
  1954   - Prevent the expansion of too large definitions.
  1955 
  1956 * Proof methods "metis" and "meson" now have configuration options
  1957 "meson_trace", "metis_trace", and "metis_verbose" that can be enabled
  1958 to diagnose these tools. E.g.
  1959 
  1960     using [[metis_trace = true]]
  1961 
  1962 * Auto Solve: Renamed "Auto Solve Direct".  The tool is now available
  1963 manually as command 'solve_direct'.
  1964 
  1965 * The default SMT solver Z3 must be enabled explicitly (due to
  1966 licensing issues) by setting the environment variable
  1967 Z3_NON_COMMERCIAL in etc/settings of the component, for example.  For
  1968 commercial applications, the SMT solver CVC3 is provided as fall-back;
  1969 changing the SMT solver is done via the configuration option
  1970 "smt_solver".
  1971 
  1972 * Remote SMT solvers need to be referred to by the "remote_" prefix,
  1973 i.e. "remote_cvc3" and "remote_z3".
  1974 
  1975 * Added basic SMT support for datatypes, records, and typedefs using
  1976 the oracle mode (no proofs).  Direct support of pairs has been dropped
  1977 in exchange (pass theorems fst_conv snd_conv pair_collapse to the SMT
  1978 support for a similar behavior).  Minor INCOMPATIBILITY.
  1979 
  1980 * Changed SMT configuration options:
  1981   - Renamed:
  1982     z3_proofs ~> smt_oracle (with inverted meaning)
  1983     z3_trace_assms ~> smt_trace_used_facts
  1984     INCOMPATIBILITY.
  1985   - Added:
  1986     smt_verbose
  1987     smt_random_seed
  1988     smt_datatypes
  1989     smt_infer_triggers
  1990     smt_monomorph_limit
  1991     cvc3_options
  1992     remote_cvc3_options
  1993     remote_z3_options
  1994     yices_options
  1995 
  1996 * Boogie output files (.b2i files) need to be declared in the theory
  1997 header.
  1998 
  1999 * Simplification procedure "list_to_set_comprehension" rewrites list
  2000 comprehensions applied to List.set to set comprehensions.  Occasional
  2001 INCOMPATIBILITY, may be deactivated like this:
  2002 
  2003   declare [[simproc del: list_to_set_comprehension]]
  2004 
  2005 * Removed old version of primrec package.  INCOMPATIBILITY.
  2006 
  2007 * Removed simplifier congruence rule of "prod_case", as has for long
  2008 been the case with "split".  INCOMPATIBILITY.
  2009 
  2010 * String.literal is a type, but not a datatype.  INCOMPATIBILITY.
  2011 
  2012 * Removed [split_format ... and ... and ...] version of
  2013 [split_format].  Potential INCOMPATIBILITY.
  2014 
  2015 * Predicate "sorted" now defined inductively, with nice induction
  2016 rules.  INCOMPATIBILITY: former sorted.simps now named sorted_simps.
  2017 
  2018 * Constant "contents" renamed to "the_elem", to free the generic name
  2019 contents for other uses.  INCOMPATIBILITY.
  2020 
  2021 * Renamed class eq and constant eq (for code generation) to class
  2022 equal and constant equal, plus renaming of related facts and various
  2023 tuning.  INCOMPATIBILITY.
  2024 
  2025 * Dropped type classes mult_mono and mult_mono1.  INCOMPATIBILITY.
  2026 
  2027 * Removed output syntax "'a ~=> 'b" for "'a => 'b option".
  2028 INCOMPATIBILITY.
  2029 
  2030 * Renamed theory Fset to Cset, type Fset.fset to Cset.set, in order to
  2031 avoid confusion with finite sets.  INCOMPATIBILITY.
  2032 
  2033 * Abandoned locales equiv, congruent and congruent2 for equivalence
  2034 relations.  INCOMPATIBILITY: use equivI rather than equiv_intro (same
  2035 for congruent(2)).
  2036 
  2037 * Some previously unqualified names have been qualified:
  2038 
  2039   types
  2040     bool ~> HOL.bool
  2041     nat ~> Nat.nat
  2042 
  2043   constants
  2044     Trueprop ~> HOL.Trueprop
  2045     True ~> HOL.True
  2046     False ~> HOL.False
  2047     op & ~> HOL.conj
  2048     op | ~> HOL.disj
  2049     op --> ~> HOL.implies
  2050     op = ~> HOL.eq
  2051     Not ~> HOL.Not
  2052     The ~> HOL.The
  2053     All ~> HOL.All
  2054     Ex ~> HOL.Ex
  2055     Ex1 ~> HOL.Ex1
  2056     Let ~> HOL.Let
  2057     If ~> HOL.If
  2058     Ball ~> Set.Ball
  2059     Bex ~> Set.Bex
  2060     Suc ~> Nat.Suc
  2061     Pair ~> Product_Type.Pair
  2062     fst ~> Product_Type.fst
  2063     snd ~> Product_Type.snd
  2064     curry ~> Product_Type.curry
  2065     op : ~> Set.member
  2066     Collect ~> Set.Collect
  2067 
  2068 INCOMPATIBILITY.
  2069 
  2070 * More canonical naming convention for some fundamental definitions:
  2071 
  2072     bot_bool_eq ~> bot_bool_def
  2073     top_bool_eq ~> top_bool_def
  2074     inf_bool_eq ~> inf_bool_def
  2075     sup_bool_eq ~> sup_bool_def
  2076     bot_fun_eq  ~> bot_fun_def
  2077     top_fun_eq  ~> top_fun_def
  2078     inf_fun_eq  ~> inf_fun_def
  2079     sup_fun_eq  ~> sup_fun_def
  2080 
  2081 INCOMPATIBILITY.
  2082 
  2083 * More stylized fact names:
  2084 
  2085   expand_fun_eq ~> fun_eq_iff
  2086   expand_set_eq ~> set_eq_iff
  2087   set_ext       ~> set_eqI
  2088   nat_number    ~> eval_nat_numeral
  2089 
  2090 INCOMPATIBILITY.
  2091 
  2092 * Refactoring of code-generation specific operations in theory List:
  2093 
  2094   constants
  2095     null ~> List.null
  2096 
  2097   facts
  2098     mem_iff ~> member_def
  2099     null_empty ~> null_def
  2100 
  2101 INCOMPATIBILITY.  Note that these were not supposed to be used
  2102 regularly unless for striking reasons; their main purpose was code
  2103 generation.
  2104 
  2105 Various operations from the Haskell prelude are used for generating
  2106 Haskell code.
  2107 
  2108 * Term "bij f" is now an abbreviation of "bij_betw f UNIV UNIV".  Term
  2109 "surj f" is now an abbreviation of "range f = UNIV".  The theorems
  2110 bij_def and surj_def are unchanged.  INCOMPATIBILITY.
  2111 
  2112 * Abolished some non-alphabetic type names: "prod" and "sum" replace
  2113 "*" and "+" respectively.  INCOMPATIBILITY.
  2114 
  2115 * Name "Plus" of disjoint sum operator "<+>" is now hidden.  Write
  2116 "Sum_Type.Plus" instead.
  2117 
  2118 * Constant "split" has been merged with constant "prod_case"; names of
  2119 ML functions, facts etc. involving split have been retained so far,
  2120 though.  INCOMPATIBILITY.
  2121 
  2122 * Dropped old infix syntax "_ mem _" for List.member; use "_ : set _"
  2123 instead.  INCOMPATIBILITY.
  2124 
  2125 * Removed lemma "Option.is_none_none" which duplicates "is_none_def".
  2126 INCOMPATIBILITY.
  2127 
  2128 * Former theory Library/Enum is now part of the HOL-Main image.
  2129 INCOMPATIBILITY: all constants of the Enum theory now have to be
  2130 referred to by its qualified name.
  2131 
  2132   enum    ~>  Enum.enum
  2133   nlists  ~>  Enum.nlists
  2134   product ~>  Enum.product
  2135 
  2136 * Theory Library/Monad_Syntax provides do-syntax for monad types.
  2137 Syntax in Library/State_Monad has been changed to avoid ambiguities.
  2138 INCOMPATIBILITY.
  2139 
  2140 * Theory Library/SetsAndFunctions has been split into
  2141 Library/Function_Algebras and Library/Set_Algebras; canonical names
  2142 for instance definitions for functions; various improvements.
  2143 INCOMPATIBILITY.
  2144 
  2145 * Theory Library/Multiset provides stable quicksort implementation of
  2146 sort_key.
  2147 
  2148 * Theory Library/Multiset: renamed empty_idemp ~> empty_neutral.
  2149 INCOMPATIBILITY.
  2150 
  2151 * Session Multivariate_Analysis: introduced a type class for euclidean
  2152 space.  Most theorems are now stated in terms of euclidean spaces
  2153 instead of finite cartesian products.
  2154 
  2155   types
  2156     real ^ 'n ~>  'a::real_vector
  2157               ~>  'a::euclidean_space
  2158               ~>  'a::ordered_euclidean_space
  2159         (depends on your needs)
  2160 
  2161   constants
  2162      _ $ _        ~> _ $$ _
  2163      \<chi> x. _  ~> \<chi>\<chi> x. _
  2164      CARD('n)     ~> DIM('a)
  2165 
  2166 Also note that the indices are now natural numbers and not from some
  2167 finite type. Finite cartesian products of euclidean spaces, products
  2168 of euclidean spaces the real and complex numbers are instantiated to
  2169 be euclidean_spaces.  INCOMPATIBILITY.
  2170 
  2171 * Session Probability: introduced pextreal as positive extended real
  2172 numbers.  Use pextreal as value for measures.  Introduce the
  2173 Radon-Nikodym derivative, product spaces and Fubini's theorem for
  2174 arbitrary sigma finite measures.  Introduces Lebesgue measure based on
  2175 the integral in Multivariate Analysis.  INCOMPATIBILITY.
  2176 
  2177 * Session Imperative_HOL: revamped, corrected dozens of inadequacies.
  2178 INCOMPATIBILITY.
  2179 
  2180 * Session SPARK (with image HOL-SPARK) provides commands to load and
  2181 prove verification conditions generated by the SPARK Ada program
  2182 verifier.  See also src/HOL/SPARK and src/HOL/SPARK/Examples.
  2183 
  2184 
  2185 *** HOL-Algebra ***
  2186 
  2187 * Theorems for additive ring operations (locale abelian_monoid and
  2188 descendants) are generated by interpretation from their multiplicative
  2189 counterparts.  Names (in particular theorem names) have the mandatory
  2190 qualifier 'add'.  Previous theorem names are redeclared for
  2191 compatibility.
  2192 
  2193 * Structure "int_ring" is now an abbreviation (previously a
  2194 definition).  This fits more natural with advanced interpretations.
  2195 
  2196 
  2197 *** HOLCF ***
  2198 
  2199 * The domain package now runs in definitional mode by default: The
  2200 former command 'new_domain' is now called 'domain'.  To use the domain
  2201 package in its original axiomatic mode, use 'domain (unsafe)'.
  2202 INCOMPATIBILITY.
  2203 
  2204 * The new class "domain" is now the default sort.  Class "predomain"
  2205 is an unpointed version of "domain". Theories can be updated by
  2206 replacing sort annotations as shown below.  INCOMPATIBILITY.
  2207 
  2208   'a::type ~> 'a::countable
  2209   'a::cpo  ~> 'a::predomain
  2210   'a::pcpo ~> 'a::domain
  2211 
  2212 * The old type class "rep" has been superseded by class "domain".
  2213 Accordingly, users of the definitional package must remove any
  2214 "default_sort rep" declarations.  INCOMPATIBILITY.
  2215 
  2216 * The domain package (definitional mode) now supports unpointed
  2217 predomain argument types, as long as they are marked 'lazy'. (Strict
  2218 arguments must be in class "domain".) For example, the following
  2219 domain definition now works:
  2220 
  2221   domain natlist = nil | cons (lazy "nat discr") (lazy "natlist")
  2222 
  2223 * Theory HOLCF/Library/HOL_Cpo provides cpo and predomain class
  2224 instances for types from main HOL: bool, nat, int, char, 'a + 'b,
  2225 'a option, and 'a list.  Additionally, it configures fixrec and the
  2226 domain package to work with these types.  For example:
  2227 
  2228   fixrec isInl :: "('a + 'b) u -> tr"
  2229     where "isInl$(up$(Inl x)) = TT" | "isInl$(up$(Inr y)) = FF"
  2230 
  2231   domain V = VFun (lazy "V -> V") | VCon (lazy "nat") (lazy "V list")
  2232 
  2233 * The "(permissive)" option of fixrec has been replaced with a
  2234 per-equation "(unchecked)" option. See
  2235 src/HOL/HOLCF/Tutorial/Fixrec_ex.thy for examples. INCOMPATIBILITY.
  2236 
  2237 * The "bifinite" class no longer fixes a constant "approx"; the class
  2238 now just asserts that such a function exists.  INCOMPATIBILITY.
  2239 
  2240 * Former type "alg_defl" has been renamed to "defl".  HOLCF no longer
  2241 defines an embedding of type 'a defl into udom by default; instances
  2242 of "bifinite" and "domain" classes are available in
  2243 src/HOL/HOLCF/Library/Defl_Bifinite.thy.
  2244 
  2245 * The syntax "REP('a)" has been replaced with "DEFL('a)".
  2246 
  2247 * The predicate "directed" has been removed.  INCOMPATIBILITY.
  2248 
  2249 * The type class "finite_po" has been removed.  INCOMPATIBILITY.
  2250 
  2251 * The function "cprod_map" has been renamed to "prod_map".
  2252 INCOMPATIBILITY.
  2253 
  2254 * The monadic bind operator on each powerdomain has new binder syntax
  2255 similar to sets, e.g. "\<Union>\<sharp>x\<in>xs. t" represents
  2256 "upper_bind\<cdot>xs\<cdot>(\<Lambda> x. t)".
  2257 
  2258 * The infix syntax for binary union on each powerdomain has changed
  2259 from e.g. "+\<sharp>" to "\<union>\<sharp>", for consistency with set
  2260 syntax.  INCOMPATIBILITY.
  2261 
  2262 * The constant "UU" has been renamed to "bottom".  The syntax "UU" is
  2263 still supported as an input translation.
  2264 
  2265 * Renamed some theorems (the original names are also still available).
  2266 
  2267   expand_fun_below   ~> fun_below_iff
  2268   below_fun_ext      ~> fun_belowI
  2269   expand_cfun_eq     ~> cfun_eq_iff
  2270   ext_cfun           ~> cfun_eqI
  2271   expand_cfun_below  ~> cfun_below_iff
  2272   below_cfun_ext     ~> cfun_belowI
  2273   cont2cont_Rep_CFun ~> cont2cont_APP
  2274 
  2275 * The Abs and Rep functions for various types have changed names.
  2276 Related theorem names have also changed to match. INCOMPATIBILITY.
  2277 
  2278   Rep_CFun  ~> Rep_cfun
  2279   Abs_CFun  ~> Abs_cfun
  2280   Rep_Sprod ~> Rep_sprod
  2281   Abs_Sprod ~> Abs_sprod
  2282   Rep_Ssum  ~> Rep_ssum
  2283   Abs_Ssum  ~> Abs_ssum
  2284 
  2285 * Lemmas with names of the form *_defined_iff or *_strict_iff have
  2286 been renamed to *_bottom_iff.  INCOMPATIBILITY.
  2287 
  2288 * Various changes to bisimulation/coinduction with domain package:
  2289 
  2290   - Definitions of "bisim" constants no longer mention definedness.
  2291   - With mutual recursion, "bisim" predicate is now curried.
  2292   - With mutual recursion, each type gets a separate coind theorem.
  2293   - Variable names in bisim_def and coinduct rules have changed.
  2294 
  2295 INCOMPATIBILITY.
  2296 
  2297 * Case combinators generated by the domain package for type "foo" are
  2298 now named "foo_case" instead of "foo_when".  INCOMPATIBILITY.
  2299 
  2300 * Several theorems have been renamed to more accurately reflect the
  2301 names of constants and types involved.  INCOMPATIBILITY.
  2302 
  2303   thelub_const    ~> lub_const
  2304   lub_const       ~> is_lub_const
  2305   thelubI         ~> lub_eqI
  2306   is_lub_lub      ~> is_lubD2
  2307   lubI            ~> is_lub_lub
  2308   unique_lub      ~> is_lub_unique
  2309   is_ub_lub       ~> is_lub_rangeD1
  2310   lub_bin_chain   ~> is_lub_bin_chain
  2311   lub_fun         ~> is_lub_fun
  2312   thelub_fun      ~> lub_fun
  2313   thelub_cfun     ~> lub_cfun
  2314   thelub_Pair     ~> lub_Pair
  2315   lub_cprod       ~> is_lub_prod
  2316   thelub_cprod    ~> lub_prod
  2317   minimal_cprod   ~> minimal_prod
  2318   inst_cprod_pcpo ~> inst_prod_pcpo
  2319   UU_I            ~> bottomI
  2320   compact_UU      ~> compact_bottom
  2321   deflation_UU    ~> deflation_bottom
  2322   finite_deflation_UU ~> finite_deflation_bottom
  2323 
  2324 * Many legacy theorem names have been discontinued.  INCOMPATIBILITY.
  2325 
  2326   sq_ord_less_eq_trans ~> below_eq_trans
  2327   sq_ord_eq_less_trans ~> eq_below_trans
  2328   refl_less            ~> below_refl
  2329   trans_less           ~> below_trans
  2330   antisym_less         ~> below_antisym
  2331   antisym_less_inverse ~> po_eq_conv [THEN iffD1]
  2332   box_less             ~> box_below
  2333   rev_trans_less       ~> rev_below_trans
  2334   not_less2not_eq      ~> not_below2not_eq
  2335   less_UU_iff          ~> below_UU_iff
  2336   flat_less_iff        ~> flat_below_iff
  2337   adm_less             ~> adm_below
  2338   adm_not_less         ~> adm_not_below
  2339   adm_compact_not_less ~> adm_compact_not_below
  2340   less_fun_def         ~> below_fun_def
  2341   expand_fun_less      ~> fun_below_iff
  2342   less_fun_ext         ~> fun_belowI
  2343   less_discr_def       ~> below_discr_def
  2344   discr_less_eq        ~> discr_below_eq
  2345   less_unit_def        ~> below_unit_def
  2346   less_cprod_def       ~> below_prod_def
  2347   prod_lessI           ~> prod_belowI
  2348   Pair_less_iff        ~> Pair_below_iff
  2349   fst_less_iff         ~> fst_below_iff
  2350   snd_less_iff         ~> snd_below_iff
  2351   expand_cfun_less     ~> cfun_below_iff
  2352   less_cfun_ext        ~> cfun_belowI
  2353   injection_less       ~> injection_below
  2354   less_up_def          ~> below_up_def
  2355   not_Iup_less         ~> not_Iup_below
  2356   Iup_less             ~> Iup_below
  2357   up_less              ~> up_below
  2358   Def_inject_less_eq   ~> Def_below_Def
  2359   Def_less_is_eq       ~> Def_below_iff
  2360   spair_less_iff       ~> spair_below_iff
  2361   less_sprod           ~> below_sprod
  2362   spair_less           ~> spair_below
  2363   sfst_less_iff        ~> sfst_below_iff
  2364   ssnd_less_iff        ~> ssnd_below_iff
  2365   fix_least_less       ~> fix_least_below
  2366   dist_less_one        ~> dist_below_one
  2367   less_ONE             ~> below_ONE
  2368   ONE_less_iff         ~> ONE_below_iff
  2369   less_sinlD           ~> below_sinlD
  2370   less_sinrD           ~> below_sinrD
  2371 
  2372 
  2373 *** FOL and ZF ***
  2374 
  2375 * All constant names are now qualified internally and use proper
  2376 identifiers, e.g. "IFOL.eq" instead of "op =".  INCOMPATIBILITY.
  2377 
  2378 
  2379 *** ML ***
  2380 
  2381 * Antiquotation @{assert} inlines a function bool -> unit that raises
  2382 Fail if the argument is false.  Due to inlining the source position of
  2383 failed assertions is included in the error output.
  2384 
  2385 * Discontinued antiquotation @{theory_ref}, which is obsolete since ML
  2386 text is in practice always evaluated with a stable theory checkpoint.
  2387 Minor INCOMPATIBILITY, use (Theory.check_thy @{theory}) instead.
  2388 
  2389 * Antiquotation @{theory A} refers to theory A from the ancestry of
  2390 the current context, not any accidental theory loader state as before.
  2391 Potential INCOMPATIBILITY, subtle change in semantics.
  2392 
  2393 * Syntax.pretty_priority (default 0) configures the required priority
  2394 of pretty-printed output and thus affects insertion of parentheses.
  2395 
  2396 * Syntax.default_root (default "any") configures the inner syntax
  2397 category (nonterminal symbol) for parsing of terms.
  2398 
  2399 * Former exception Library.UnequalLengths now coincides with
  2400 ListPair.UnequalLengths.
  2401 
  2402 * Renamed structure MetaSimplifier to Raw_Simplifier.  Note that the
  2403 main functionality is provided by structure Simplifier.
  2404 
  2405 * Renamed raw "explode" function to "raw_explode" to emphasize its
  2406 meaning.  Note that internally to Isabelle, Symbol.explode is used in
  2407 almost all situations.
  2408 
  2409 * Discontinued obsolete function sys_error and exception SYS_ERROR.
  2410 See implementation manual for further details on exceptions in
  2411 Isabelle/ML.
  2412 
  2413 * Renamed setmp_noncritical to Unsynchronized.setmp to emphasize its
  2414 meaning.
  2415 
  2416 * Renamed structure PureThy to Pure_Thy and moved most of its
  2417 operations to structure Global_Theory, to emphasize that this is
  2418 rarely-used global-only stuff.
  2419 
  2420 * Discontinued Output.debug.  Minor INCOMPATIBILITY, use plain writeln
  2421 instead (or tracing for high-volume output).
  2422 
  2423 * Configuration option show_question_marks only affects regular pretty
  2424 printing of types and terms, not raw Term.string_of_vname.
  2425 
  2426 * ML_Context.thm and ML_Context.thms are no longer pervasive.  Rare
  2427 INCOMPATIBILITY, superseded by static antiquotations @{thm} and
  2428 @{thms} for most purposes.
  2429 
  2430 * ML structure Unsynchronized is never opened, not even in Isar
  2431 interaction mode as before.  Old Unsynchronized.set etc. have been
  2432 discontinued -- use plain := instead.  This should be *rare* anyway,
  2433 since modern tools always work via official context data, notably
  2434 configuration options.
  2435 
  2436 * Parallel and asynchronous execution requires special care concerning
  2437 interrupts.  Structure Exn provides some convenience functions that
  2438 avoid working directly with raw Interrupt.  User code must not absorb
  2439 interrupts -- intermediate handling (for cleanup etc.) needs to be
  2440 followed by re-raising of the original exception.  Another common
  2441 source of mistakes are "handle _" patterns, which make the meaning of
  2442 the program subject to physical effects of the environment.
  2443 
  2444 
  2445 
  2446 New in Isabelle2009-2 (June 2010)
  2447 ---------------------------------
  2448 
  2449 *** General ***
  2450 
  2451 * Authentic syntax for *all* logical entities (type classes, type
  2452 constructors, term constants): provides simple and robust
  2453 correspondence between formal entities and concrete syntax.  Within
  2454 the parse tree / AST representations, "constants" are decorated by
  2455 their category (class, type, const) and spelled out explicitly with
  2456 their full internal name.
  2457 
  2458 Substantial INCOMPATIBILITY concerning low-level syntax declarations
  2459 and translations (translation rules and translation functions in ML).
  2460 Some hints on upgrading:
  2461 
  2462   - Many existing uses of 'syntax' and 'translations' can be replaced
  2463     by more modern 'type_notation', 'notation' and 'abbreviation',
  2464     which are independent of this issue.
  2465 
  2466   - 'translations' require markup within the AST; the term syntax
  2467     provides the following special forms:
  2468 
  2469       CONST c   -- produces syntax version of constant c from context
  2470       XCONST c  -- literally c, checked as constant from context
  2471       c         -- literally c, if declared by 'syntax'
  2472 
  2473     Plain identifiers are treated as AST variables -- occasionally the
  2474     system indicates accidental variables via the error "rhs contains
  2475     extra variables".
  2476 
  2477     Type classes and type constructors are marked according to their
  2478     concrete syntax.  Some old translations rules need to be written
  2479     for the "type" category, using type constructor application
  2480     instead of pseudo-term application of the default category
  2481     "logic".
  2482 
  2483   - 'parse_translation' etc. in ML may use the following
  2484     antiquotations:
  2485 
  2486       @{class_syntax c}   -- type class c within parse tree / AST
  2487       @{term_syntax c}    -- type constructor c within parse tree / AST
  2488       @{const_syntax c}   -- ML version of "CONST c" above
  2489       @{syntax_const c}   -- literally c (checked wrt. 'syntax' declarations)
  2490 
  2491   - Literal types within 'typed_print_translations', i.e. those *not*
  2492     represented as pseudo-terms are represented verbatim.  Use @{class
  2493     c} or @{type_name c} here instead of the above syntax
  2494     antiquotations.
  2495 
  2496 Note that old non-authentic syntax was based on unqualified base
  2497 names, so all of the above "constant" names would coincide.  Recall
  2498 that 'print_syntax' and ML_command "set Syntax.trace_ast" help to
  2499 diagnose syntax problems.
  2500 
  2501 * Type constructors admit general mixfix syntax, not just infix.
  2502 
  2503 * Concrete syntax may be attached to local entities without a proof
  2504 body, too.  This works via regular mixfix annotations for 'fix',
  2505 'def', 'obtain' etc. or via the explicit 'write' command, which is
  2506 similar to the 'notation' command in theory specifications.
  2507 
  2508 * Discontinued unnamed infix syntax (legacy feature for many years) --
  2509 need to specify constant name and syntax separately.  Internal ML
  2510 datatype constructors have been renamed from InfixName to Infix etc.
  2511 Minor INCOMPATIBILITY.
  2512 
  2513 * Schematic theorem statements need to be explicitly markup as such,
  2514 via commands 'schematic_lemma', 'schematic_theorem',
  2515 'schematic_corollary'.  Thus the relevance of the proof is made
  2516 syntactically clear, which impacts performance in a parallel or
  2517 asynchronous interactive environment.  Minor INCOMPATIBILITY.
  2518 
  2519 * Use of cumulative prems via "!" in some proof methods has been
  2520 discontinued (old legacy feature).
  2521 
  2522 * References 'trace_simp' and 'debug_simp' have been replaced by
  2523 configuration options stored in the context. Enabling tracing (the
  2524 case of debugging is similar) in proofs works via
  2525 
  2526   using [[trace_simp = true]]
  2527 
  2528 Tracing is then active for all invocations of the simplifier in
  2529 subsequent goal refinement steps. Tracing may also still be enabled or
  2530 disabled via the ProofGeneral settings menu.
  2531 
  2532 * Separate commands 'hide_class', 'hide_type', 'hide_const',
  2533 'hide_fact' replace the former 'hide' KIND command.  Minor
  2534 INCOMPATIBILITY.
  2535 
  2536 * Improved parallelism of proof term normalization: usedir -p2 -q0 is
  2537 more efficient than combinations with -q1 or -q2.
  2538 
  2539 
  2540 *** Pure ***
  2541 
  2542 * Proofterms record type-class reasoning explicitly, using the
  2543 "unconstrain" operation internally.  This eliminates all sort
  2544 constraints from a theorem and proof, introducing explicit
  2545 OFCLASS-premises.  On the proof term level, this operation is
  2546 automatically applied at theorem boundaries, such that closed proofs
  2547 are always free of sort constraints.  INCOMPATIBILITY for tools that
  2548 inspect proof terms.
  2549 
  2550 * Local theory specifications may depend on extra type variables that
  2551 are not present in the result type -- arguments TYPE('a) :: 'a itself
  2552 are added internally.  For example:
  2553 
  2554   definition unitary :: bool where "unitary = (ALL (x::'a) y. x = y)"
  2555 
  2556 * Predicates of locales introduced by classes carry a mandatory
  2557 "class" prefix.  INCOMPATIBILITY.
  2558 
  2559 * Vacuous class specifications observe default sort.  INCOMPATIBILITY.
  2560 
  2561 * Old 'axclass' command has been discontinued.  INCOMPATIBILITY, use
  2562 'class' instead.
  2563 
  2564 * Command 'code_reflect' allows to incorporate generated ML code into
  2565 runtime environment; replaces immature code_datatype antiquotation.
  2566 INCOMPATIBILITY.
  2567 
  2568 * Code generator: simple concept for abstract datatypes obeying
  2569 invariants.
  2570 
  2571 * Code generator: details of internal data cache have no impact on the
  2572 user space functionality any longer.
  2573 
  2574 * Methods "unfold_locales" and "intro_locales" ignore non-locale
  2575 subgoals.  This is more appropriate for interpretations with 'where'.
  2576 INCOMPATIBILITY.
  2577 
  2578 * Command 'example_proof' opens an empty proof body.  This allows to
  2579 experiment with Isar, without producing any persistent result.
  2580 
  2581 * Commands 'type_notation' and 'no_type_notation' declare type syntax
  2582 within a local theory context, with explicit checking of the
  2583 constructors involved (in contrast to the raw 'syntax' versions).
  2584 
  2585 * Commands 'types' and 'typedecl' now work within a local theory
  2586 context -- without introducing dependencies on parameters or
  2587 assumptions, which is not possible in Isabelle/Pure.
  2588 
  2589 * Command 'defaultsort' has been renamed to 'default_sort', it works
  2590 within a local theory context.  Minor INCOMPATIBILITY.
  2591 
  2592 
  2593 *** HOL ***
  2594 
  2595 * Command 'typedef' now works within a local theory context -- without
  2596 introducing dependencies on parameters or assumptions, which is not
  2597 possible in Isabelle/Pure/HOL.  Note that the logical environment may
  2598 contain multiple interpretations of local typedefs (with different
  2599 non-emptiness proofs), even in a global theory context.
  2600 
  2601 * New package for quotient types.  Commands 'quotient_type' and
  2602 'quotient_definition' may be used for defining types and constants by
  2603 quotient constructions.  An example is the type of integers created by
  2604 quotienting pairs of natural numbers:
  2605 
  2606   fun
  2607     intrel :: "(nat * nat) => (nat * nat) => bool"
  2608   where
  2609     "intrel (x, y) (u, v) = (x + v = u + y)"
  2610 
  2611   quotient_type int = "nat * nat" / intrel
  2612     by (auto simp add: equivp_def expand_fun_eq)
  2613 
  2614   quotient_definition
  2615     "0::int" is "(0::nat, 0::nat)"
  2616 
  2617 The method "lifting" can be used to lift of theorems from the
  2618 underlying "raw" type to the quotient type.  The example
  2619 src/HOL/Quotient_Examples/FSet.thy includes such a quotient
  2620 construction and provides a reasoning infrastructure for finite sets.
  2621 
  2622 * Renamed Library/Quotient.thy to Library/Quotient_Type.thy to avoid
  2623 clash with new theory Quotient in Main HOL.
  2624 
  2625 * Moved the SMT binding into the main HOL session, eliminating
  2626 separate HOL-SMT session.
  2627 
  2628 * List membership infix mem operation is only an input abbreviation.
  2629 INCOMPATIBILITY.
  2630 
  2631 * Theory Library/Word.thy has been removed.  Use library Word/Word.thy
  2632 for future developements; former Library/Word.thy is still present in
  2633 the AFP entry RSAPPS.
  2634 
  2635 * Theorem Int.int_induct renamed to Int.int_of_nat_induct and is no
  2636 longer shadowed.  INCOMPATIBILITY.
  2637 
  2638 * Dropped theorem duplicate comp_arith; use semiring_norm instead.
  2639 INCOMPATIBILITY.
  2640 
  2641 * Dropped theorem RealPow.real_sq_order; use power2_le_imp_le instead.
  2642 INCOMPATIBILITY.
  2643 
  2644 * Dropped normalizing_semiring etc; use the facts in semiring classes
  2645 instead.  INCOMPATIBILITY.
  2646 
  2647 * Dropped several real-specific versions of lemmas about floor and
  2648 ceiling; use the generic lemmas from theory "Archimedean_Field"
  2649 instead.  INCOMPATIBILITY.
  2650 
  2651   floor_number_of_eq         ~> floor_number_of
  2652   le_floor_eq_number_of      ~> number_of_le_floor
  2653   le_floor_eq_zero           ~> zero_le_floor
  2654   le_floor_eq_one            ~> one_le_floor
  2655   floor_less_eq_number_of    ~> floor_less_number_of
  2656   floor_less_eq_zero         ~> floor_less_zero
  2657   floor_less_eq_one          ~> floor_less_one
  2658   less_floor_eq_number_of    ~> number_of_less_floor
  2659   less_floor_eq_zero         ~> zero_less_floor
  2660   less_floor_eq_one          ~> one_less_floor
  2661   floor_le_eq_number_of      ~> floor_le_number_of
  2662   floor_le_eq_zero           ~> floor_le_zero
  2663   floor_le_eq_one            ~> floor_le_one
  2664   floor_subtract_number_of   ~> floor_diff_number_of
  2665   floor_subtract_one         ~> floor_diff_one
  2666   ceiling_number_of_eq       ~> ceiling_number_of
  2667   ceiling_le_eq_number_of    ~> ceiling_le_number_of
  2668   ceiling_le_zero_eq         ~> ceiling_le_zero
  2669   ceiling_le_eq_one          ~> ceiling_le_one
  2670   less_ceiling_eq_number_of  ~> number_of_less_ceiling
  2671   less_ceiling_eq_zero       ~> zero_less_ceiling
  2672   less_ceiling_eq_one        ~> one_less_ceiling
  2673   ceiling_less_eq_number_of  ~> ceiling_less_number_of
  2674   ceiling_less_eq_zero       ~> ceiling_less_zero
  2675   ceiling_less_eq_one        ~> ceiling_less_one
  2676   le_ceiling_eq_number_of    ~> number_of_le_ceiling
  2677   le_ceiling_eq_zero         ~> zero_le_ceiling
  2678   le_ceiling_eq_one          ~> one_le_ceiling
  2679   ceiling_subtract_number_of ~> ceiling_diff_number_of
  2680   ceiling_subtract_one       ~> ceiling_diff_one
  2681 
  2682 * Theory "Finite_Set": various folding_XXX locales facilitate the
  2683 application of the various fold combinators on finite sets.
  2684 
  2685 * Library theory "RBT" renamed to "RBT_Impl"; new library theory "RBT"
  2686 provides abstract red-black tree type which is backed by "RBT_Impl" as
  2687 implementation.  INCOMPATIBILTY.
  2688 
  2689 * Theory Library/Coinductive_List has been removed -- superseded by
  2690 AFP/thys/Coinductive.
  2691 
  2692 * Theory PReal, including the type "preal" and related operations, has
  2693 been removed.  INCOMPATIBILITY.
  2694 
  2695 * Real: new development using Cauchy Sequences.
  2696 
  2697 * Split off theory "Big_Operators" containing setsum, setprod,
  2698 Inf_fin, Sup_fin, Min, Max from theory Finite_Set.  INCOMPATIBILITY.
  2699 
  2700 * Theory "Rational" renamed to "Rat", for consistency with "Nat",
  2701 "Int" etc.  INCOMPATIBILITY.
  2702 
  2703 * Constant Rat.normalize needs to be qualified.  INCOMPATIBILITY.
  2704 
  2705 * New set of rules "ac_simps" provides combined assoc / commute
  2706 rewrites for all interpretations of the appropriate generic locales.
  2707 
  2708 * Renamed theory "OrderedGroup" to "Groups" and split theory
  2709 "Ring_and_Field" into theories "Rings" and "Fields"; for more
  2710 appropriate and more consistent names suitable for name prefixes
  2711 within the HOL theories.  INCOMPATIBILITY.
  2712 
  2713 * Some generic constants have been put to appropriate theories:
  2714   - less_eq, less: Orderings
  2715   - zero, one, plus, minus, uminus, times, abs, sgn: Groups
  2716   - inverse, divide: Rings
  2717 INCOMPATIBILITY.
  2718 
  2719 * More consistent naming of type classes involving orderings (and
  2720 lattices):
  2721 
  2722     lower_semilattice                   ~> semilattice_inf
  2723     upper_semilattice                   ~> semilattice_sup
  2724 
  2725     dense_linear_order                  ~> dense_linorder
  2726 
  2727     pordered_ab_group_add               ~> ordered_ab_group_add
  2728     pordered_ab_group_add_abs           ~> ordered_ab_group_add_abs
  2729     pordered_ab_semigroup_add           ~> ordered_ab_semigroup_add
  2730     pordered_ab_semigroup_add_imp_le    ~> ordered_ab_semigroup_add_imp_le
  2731     pordered_cancel_ab_semigroup_add    ~> ordered_cancel_ab_semigroup_add
  2732     pordered_cancel_comm_semiring       ~> ordered_cancel_comm_semiring
  2733     pordered_cancel_semiring            ~> ordered_cancel_semiring
  2734     pordered_comm_monoid_add            ~> ordered_comm_monoid_add
  2735     pordered_comm_ring                  ~> ordered_comm_ring
  2736     pordered_comm_semiring              ~> ordered_comm_semiring
  2737     pordered_ring                       ~> ordered_ring
  2738     pordered_ring_abs                   ~> ordered_ring_abs
  2739     pordered_semiring                   ~> ordered_semiring
  2740 
  2741     ordered_ab_group_add                ~> linordered_ab_group_add
  2742     ordered_ab_semigroup_add            ~> linordered_ab_semigroup_add
  2743     ordered_cancel_ab_semigroup_add     ~> linordered_cancel_ab_semigroup_add
  2744     ordered_comm_semiring_strict        ~> linordered_comm_semiring_strict
  2745     ordered_field                       ~> linordered_field
  2746     ordered_field_no_lb                 ~> linordered_field_no_lb
  2747     ordered_field_no_ub                 ~> linordered_field_no_ub
  2748     ordered_field_dense_linear_order    ~> dense_linordered_field
  2749     ordered_idom                        ~> linordered_idom
  2750     ordered_ring                        ~> linordered_ring
  2751     ordered_ring_le_cancel_factor       ~> linordered_ring_le_cancel_factor
  2752     ordered_ring_less_cancel_factor     ~> linordered_ring_less_cancel_factor
  2753     ordered_ring_strict                 ~> linordered_ring_strict
  2754     ordered_semidom                     ~> linordered_semidom
  2755     ordered_semiring                    ~> linordered_semiring
  2756     ordered_semiring_1                  ~> linordered_semiring_1
  2757     ordered_semiring_1_strict           ~> linordered_semiring_1_strict
  2758     ordered_semiring_strict             ~> linordered_semiring_strict
  2759 
  2760   The following slightly odd type classes have been moved to a
  2761   separate theory Library/Lattice_Algebras:
  2762 
  2763     lordered_ab_group_add               ~> lattice_ab_group_add
  2764     lordered_ab_group_add_abs           ~> lattice_ab_group_add_abs
  2765     lordered_ab_group_add_meet          ~> semilattice_inf_ab_group_add
  2766     lordered_ab_group_add_join          ~> semilattice_sup_ab_group_add
  2767     lordered_ring                       ~> lattice_ring
  2768 
  2769 INCOMPATIBILITY.
  2770 
  2771 * Refined field classes:
  2772   - classes division_ring_inverse_zero, field_inverse_zero,
  2773     linordered_field_inverse_zero include rule inverse 0 = 0 --
  2774     subsumes former division_by_zero class;
  2775   - numerous lemmas have been ported from field to division_ring.
  2776 INCOMPATIBILITY.
  2777 
  2778 * Refined algebra theorem collections:
  2779   - dropped theorem group group_simps, use algebra_simps instead;
  2780   - dropped theorem group ring_simps, use field_simps instead;
  2781   - proper theorem collection field_simps subsumes former theorem
  2782     groups field_eq_simps and field_simps;
  2783   - dropped lemma eq_minus_self_iff which is a duplicate for
  2784     equal_neg_zero.
  2785 INCOMPATIBILITY.
  2786 
  2787 * Theory Finite_Set and List: some lemmas have been generalized from
  2788 sets to lattices:
  2789 
  2790   fun_left_comm_idem_inter      ~> fun_left_comm_idem_inf
  2791   fun_left_comm_idem_union      ~> fun_left_comm_idem_sup
  2792   inter_Inter_fold_inter        ~> inf_Inf_fold_inf
  2793   union_Union_fold_union        ~> sup_Sup_fold_sup
  2794   Inter_fold_inter              ~> Inf_fold_inf
  2795   Union_fold_union              ~> Sup_fold_sup
  2796   inter_INTER_fold_inter        ~> inf_INFI_fold_inf
  2797   union_UNION_fold_union        ~> sup_SUPR_fold_sup
  2798   INTER_fold_inter              ~> INFI_fold_inf
  2799   UNION_fold_union              ~> SUPR_fold_sup
  2800 
  2801 * Theory "Complete_Lattice": lemmas top_def and bot_def have been
  2802 replaced by the more convenient lemmas Inf_empty and Sup_empty.
  2803 Dropped lemmas Inf_insert_simp and Sup_insert_simp, which are subsumed
  2804 by Inf_insert and Sup_insert.  Lemmas Inf_UNIV and Sup_UNIV replace
  2805 former Inf_Univ and Sup_Univ.  Lemmas inf_top_right and sup_bot_right
  2806 subsume inf_top and sup_bot respectively.  INCOMPATIBILITY.
  2807 
  2808 * Reorganized theory Multiset: swapped notation of pointwise and
  2809 multiset order:
  2810 
  2811   - pointwise ordering is instance of class order with standard syntax
  2812     <= and <;
  2813   - multiset ordering has syntax <=# and <#; partial order properties
  2814     are provided by means of interpretation with prefix
  2815     multiset_order;
  2816   - less duplication, less historical organization of sections,
  2817     conversion from associations lists to multisets, rudimentary code
  2818     generation;
  2819   - use insert_DiffM2 [symmetric] instead of elem_imp_eq_diff_union,
  2820     if needed.
  2821 
  2822 Renamed:
  2823 
  2824   multiset_eq_conv_count_eq  ~>  multiset_ext_iff
  2825   multi_count_ext  ~>  multiset_ext
  2826   diff_union_inverse2  ~>  diff_union_cancelR
  2827 
  2828 INCOMPATIBILITY.
  2829 
  2830 * Theory Permutation: replaced local "remove" by List.remove1.
  2831 
  2832 * Code generation: ML and OCaml code is decorated with signatures.
  2833 
  2834 * Theory List: added transpose.
  2835 
  2836 * Library/Nat_Bijection.thy is a collection of bijective functions
  2837 between nat and other types, which supersedes the older libraries
  2838 Library/Nat_Int_Bij.thy and HOLCF/NatIso.thy.  INCOMPATIBILITY.
  2839 
  2840   Constants:
  2841   Nat_Int_Bij.nat2_to_nat         ~> prod_encode
  2842   Nat_Int_Bij.nat_to_nat2         ~> prod_decode
  2843   Nat_Int_Bij.int_to_nat_bij      ~> int_encode
  2844   Nat_Int_Bij.nat_to_int_bij      ~> int_decode
  2845   Countable.pair_encode           ~> prod_encode
  2846   NatIso.prod2nat                 ~> prod_encode
  2847   NatIso.nat2prod                 ~> prod_decode
  2848   NatIso.sum2nat                  ~> sum_encode
  2849   NatIso.nat2sum                  ~> sum_decode
  2850   NatIso.list2nat                 ~> list_encode
  2851   NatIso.nat2list                 ~> list_decode
  2852   NatIso.set2nat                  ~> set_encode
  2853   NatIso.nat2set                  ~> set_decode
  2854 
  2855   Lemmas:
  2856   Nat_Int_Bij.bij_nat_to_int_bij  ~> bij_int_decode
  2857   Nat_Int_Bij.nat2_to_nat_inj     ~> inj_prod_encode
  2858   Nat_Int_Bij.nat2_to_nat_surj    ~> surj_prod_encode
  2859   Nat_Int_Bij.nat_to_nat2_inj     ~> inj_prod_decode
  2860   Nat_Int_Bij.nat_to_nat2_surj    ~> surj_prod_decode
  2861   Nat_Int_Bij.i2n_n2i_id          ~> int_encode_inverse
  2862   Nat_Int_Bij.n2i_i2n_id          ~> int_decode_inverse
  2863   Nat_Int_Bij.surj_nat_to_int_bij ~> surj_int_encode
  2864   Nat_Int_Bij.surj_int_to_nat_bij ~> surj_int_decode
  2865   Nat_Int_Bij.inj_nat_to_int_bij  ~> inj_int_encode
  2866   Nat_Int_Bij.inj_int_to_nat_bij  ~> inj_int_decode
  2867   Nat_Int_Bij.bij_nat_to_int_bij  ~> bij_int_encode
  2868   Nat_Int_Bij.bij_int_to_nat_bij  ~> bij_int_decode
  2869 
  2870 * Sledgehammer:
  2871   - Renamed ATP commands:
  2872     atp_info     ~> sledgehammer running_atps
  2873     atp_kill     ~> sledgehammer kill_atps
  2874     atp_messages ~> sledgehammer messages
  2875     atp_minimize ~> sledgehammer minimize
  2876     print_atps   ~> sledgehammer available_atps
  2877     INCOMPATIBILITY.
  2878   - Added user's manual ("isabelle doc sledgehammer").
  2879   - Added option syntax and "sledgehammer_params" to customize
  2880     Sledgehammer's behavior.  See the manual for details.
  2881   - Modified the Isar proof reconstruction code so that it produces
  2882     direct proofs rather than proofs by contradiction.  (This feature
  2883     is still experimental.)
  2884   - Made Isar proof reconstruction work for SPASS, remote ATPs, and in
  2885     full-typed mode.
  2886   - Added support for TPTP syntax for SPASS via the "spass_tptp" ATP.
  2887 
  2888 * Nitpick:
  2889   - Added and implemented "binary_ints" and "bits" options.
  2890   - Added "std" option and implemented support for nonstandard models.
  2891   - Added and implemented "finitize" option to improve the precision
  2892     of infinite datatypes based on a monotonicity analysis.
  2893   - Added support for quotient types.
  2894   - Added support for "specification" and "ax_specification"
  2895     constructs.
  2896   - Added support for local definitions (for "function" and
  2897     "termination" proofs).
  2898   - Added support for term postprocessors.
  2899   - Optimized "Multiset.multiset" and "FinFun.finfun".
  2900   - Improved efficiency of "destroy_constrs" optimization.
  2901   - Fixed soundness bugs related to "destroy_constrs" optimization and
  2902     record getters.
  2903   - Fixed soundness bug related to higher-order constructors.
  2904   - Fixed soundness bug when "full_descrs" is enabled.
  2905   - Improved precision of set constructs.
  2906   - Added "atoms" option.
  2907   - Added cache to speed up repeated Kodkod invocations on the same
  2908     problems.
  2909   - Renamed "MiniSatJNI", "zChaffJNI", "BerkMinAlloy", and
  2910     "SAT4JLight" to "MiniSat_JNI", "zChaff_JNI", "BerkMin_Alloy", and
  2911     "SAT4J_Light".  INCOMPATIBILITY.
  2912   - Removed "skolemize", "uncurry", "sym_break", "flatten_prop",
  2913     "sharing_depth", and "show_skolems" options.  INCOMPATIBILITY.
  2914   - Removed "nitpick_intro" attribute.  INCOMPATIBILITY.
  2915 
  2916 * Method "induct" now takes instantiations of the form t, where t is not
  2917   a variable, as a shorthand for "x == t", where x is a fresh variable.
  2918   If this is not intended, t has to be enclosed in parentheses.
  2919   By default, the equalities generated by definitional instantiations
  2920   are pre-simplified, which may cause parameters of inductive cases
  2921   to disappear, or may even delete some of the inductive cases.
  2922   Use "induct (no_simp)" instead of "induct" to restore the old
  2923   behaviour. The (no_simp) option is also understood by the "cases"
  2924   and "nominal_induct" methods, which now perform pre-simplification, too.
  2925   INCOMPATIBILITY.
  2926 
  2927 
  2928 *** HOLCF ***
  2929 
  2930 * Variable names in lemmas generated by the domain package have
  2931 changed; the naming scheme is now consistent with the HOL datatype
  2932 package.  Some proof scripts may be affected, INCOMPATIBILITY.
  2933 
  2934 * The domain package no longer defines the function "foo_copy" for
  2935 recursive domain "foo".  The reach lemma is now stated directly in
  2936 terms of "foo_take".  Lemmas and proofs that mention "foo_copy" must
  2937 be reformulated in terms of "foo_take", INCOMPATIBILITY.
  2938 
  2939 * Most definedness lemmas generated by the domain package (previously
  2940 of the form "x ~= UU ==> foo$x ~= UU") now have an if-and-only-if form
  2941 like "foo$x = UU <-> x = UU", which works better as a simp rule.
  2942 Proofs that used definedness lemmas as intro rules may break,
  2943 potential INCOMPATIBILITY.
  2944 
  2945 * Induction and casedist rules generated by the domain package now
  2946 declare proper case_names (one called "bottom", and one named for each
  2947 constructor).  INCOMPATIBILITY.
  2948 
  2949 * For mutually-recursive domains, separate "reach" and "take_lemma"
  2950 rules are generated for each domain, INCOMPATIBILITY.
  2951 
  2952   foo_bar.reach       ~> foo.reach  bar.reach
  2953   foo_bar.take_lemmas ~> foo.take_lemma  bar.take_lemma
  2954 
  2955 * Some lemmas generated by the domain package have been renamed for
  2956 consistency with the datatype package, INCOMPATIBILITY.
  2957 
  2958   foo.ind        ~> foo.induct
  2959   foo.finite_ind ~> foo.finite_induct
  2960   foo.coind      ~> foo.coinduct
  2961   foo.casedist   ~> foo.exhaust
  2962   foo.exhaust    ~> foo.nchotomy
  2963 
  2964 * For consistency with other definition packages, the fixrec package
  2965 now generates qualified theorem names, INCOMPATIBILITY.
  2966 
  2967   foo_simps  ~> foo.simps
  2968   foo_unfold ~> foo.unfold
  2969   foo_induct ~> foo.induct
  2970 
  2971 * The "fixrec_simp" attribute has been removed.  The "fixrec_simp"
  2972 method and internal fixrec proofs now use the default simpset instead.
  2973 INCOMPATIBILITY.
  2974 
  2975 * The "contlub" predicate has been removed.  Proof scripts should use
  2976 lemma contI2 in place of monocontlub2cont, INCOMPATIBILITY.
  2977 
  2978 * The "admw" predicate has been removed, INCOMPATIBILITY.
  2979 
  2980 * The constants cpair, cfst, and csnd have been removed in favor of
  2981 Pair, fst, and snd from Isabelle/HOL, INCOMPATIBILITY.
  2982 
  2983 
  2984 *** ML ***
  2985 
  2986 * Antiquotations for basic formal entities:
  2987 
  2988     @{class NAME}         -- type class
  2989     @{class_syntax NAME}  -- syntax representation of the above
  2990 
  2991     @{type_name NAME}     -- logical type
  2992     @{type_abbrev NAME}   -- type abbreviation
  2993     @{nonterminal NAME}   -- type of concrete syntactic category
  2994     @{type_syntax NAME}   -- syntax representation of any of the above
  2995 
  2996     @{const_name NAME}    -- logical constant (INCOMPATIBILITY)
  2997     @{const_abbrev NAME}  -- abbreviated constant
  2998     @{const_syntax NAME}  -- syntax representation of any of the above
  2999 
  3000 * Antiquotation @{syntax_const NAME} ensures that NAME refers to a raw
  3001 syntax constant (cf. 'syntax' command).
  3002 
  3003 * Antiquotation @{make_string} inlines a function to print arbitrary
  3004 values similar to the ML toplevel.  The result is compiler dependent
  3005 and may fall back on "?" in certain situations.
  3006 
  3007 * Diagnostic commands 'ML_val' and 'ML_command' may refer to
  3008 antiquotations @{Isar.state} and @{Isar.goal}.  This replaces impure
  3009 Isar.state() and Isar.goal(), which belong to the old TTY loop and do
  3010 not work with the asynchronous Isar document model.
  3011 
  3012 * Configuration options now admit dynamic default values, depending on
  3013 the context or even global references.
  3014 
  3015 * SHA1.digest digests strings according to SHA-1 (see RFC 3174).  It
  3016 uses an efficient external library if available (for Poly/ML).
  3017 
  3018 * Renamed some important ML structures, while keeping the old names
  3019 for some time as aliases within the structure Legacy:
  3020 
  3021   OuterKeyword  ~>  Keyword
  3022   OuterLex      ~>  Token
  3023   OuterParse    ~>  Parse
  3024   OuterSyntax   ~>  Outer_Syntax
  3025   PrintMode     ~>  Print_Mode
  3026   SpecParse     ~>  Parse_Spec
  3027   ThyInfo       ~>  Thy_Info
  3028   ThyLoad       ~>  Thy_Load
  3029   ThyOutput     ~>  Thy_Output
  3030   TypeInfer     ~>  Type_Infer
  3031 
  3032 Note that "open Legacy" simplifies porting of sources, but forgetting
  3033 to remove it again will complicate porting again in the future.
  3034 
  3035 * Most operations that refer to a global context are named
  3036 accordingly, e.g. Simplifier.global_context or
  3037 ProofContext.init_global.  There are some situations where a global
  3038 context actually works, but under normal circumstances one needs to
  3039 pass the proper local context through the code!
  3040 
  3041 * Discontinued old TheoryDataFun with its copy/init operation -- data
  3042 needs to be pure.  Functor Theory_Data_PP retains the traditional
  3043 Pretty.pp argument to merge, which is absent in the standard
  3044 Theory_Data version.
  3045 
  3046 * Sorts.certify_sort and derived "cert" operations for types and terms
  3047 no longer minimize sorts.  Thus certification at the boundary of the
  3048 inference kernel becomes invariant under addition of class relations,
  3049 which is an important monotonicity principle.  Sorts are now minimized
  3050 in the syntax layer only, at the boundary between the end-user and the
  3051 system.  Subtle INCOMPATIBILITY, may have to use Sign.minimize_sort
  3052 explicitly in rare situations.
  3053 
  3054 * Renamed old-style Drule.standard to Drule.export_without_context, to
  3055 emphasize that this is in no way a standard operation.
  3056 INCOMPATIBILITY.
  3057 
  3058 * Subgoal.FOCUS (and variants): resulting goal state is normalized as
  3059 usual for resolution.  Rare INCOMPATIBILITY.
  3060 
  3061 * Renamed varify/unvarify operations to varify_global/unvarify_global
  3062 to emphasize that these only work in a global situation (which is
  3063 quite rare).
  3064 
  3065 * Curried take and drop in library.ML; negative length is interpreted
  3066 as infinity (as in chop).  Subtle INCOMPATIBILITY.
  3067 
  3068 * Proof terms: type substitutions on proof constants now use canonical
  3069 order of type variables.  INCOMPATIBILITY for tools working with proof
  3070 terms.
  3071 
  3072 * Raw axioms/defs may no longer carry sort constraints, and raw defs
  3073 may no longer carry premises.  User-level specifications are
  3074 transformed accordingly by Thm.add_axiom/add_def.
  3075 
  3076 
  3077 *** System ***
  3078 
  3079 * Discontinued special HOL_USEDIR_OPTIONS for the main HOL image;
  3080 ISABELLE_USEDIR_OPTIONS applies uniformly to all sessions.  Note that
  3081 proof terms are enabled unconditionally in the new HOL-Proofs image.
  3082 
  3083 * Discontinued old ISABELLE and ISATOOL environment settings (legacy
  3084 feature since Isabelle2009).  Use ISABELLE_PROCESS and ISABELLE_TOOL,
  3085 respectively.
  3086 
  3087 * Old lib/scripts/polyml-platform is superseded by the
  3088 ISABELLE_PLATFORM setting variable, which defaults to the 32 bit
  3089 variant, even on a 64 bit machine.  The following example setting
  3090 prefers 64 bit if available:
  3091 
  3092   ML_PLATFORM="${ISABELLE_PLATFORM64:-$ISABELLE_PLATFORM}"
  3093 
  3094 * The preliminary Isabelle/jEdit application demonstrates the emerging
  3095 Isabelle/Scala layer for advanced prover interaction and integration.
  3096 See src/Tools/jEdit or "isabelle jedit" provided by the properly built
  3097 component.
  3098 
  3099 * "IsabelleText" is a Unicode font derived from Bitstream Vera Mono
  3100 and Bluesky TeX fonts.  It provides the usual Isabelle symbols,
  3101 similar to the default assignment of the document preparation system
  3102 (cf. isabellesym.sty).  The Isabelle/Scala class Isabelle_System
  3103 provides some operations for direct access to the font without asking
  3104 the user for manual installation.
  3105 
  3106 
  3107 
  3108 New in Isabelle2009-1 (December 2009)
  3109 -------------------------------------
  3110 
  3111 *** General ***
  3112 
  3113 * Discontinued old form of "escaped symbols" such as \\<forall>.  Only
  3114 one backslash should be used, even in ML sources.
  3115 
  3116 
  3117 *** Pure ***
  3118 
  3119 * Locale interpretation propagates mixins along the locale hierarchy.
  3120 The currently only available mixins are the equations used to map
  3121 local definitions to terms of the target domain of an interpretation.
  3122 
  3123 * Reactivated diagnostic command 'print_interps'.  Use "print_interps
  3124 loc" to print all interpretations of locale "loc" in the theory.
  3125 Interpretations in proofs are not shown.
  3126 
  3127 * Thoroughly revised locales tutorial.  New section on conditional
  3128 interpretation.
  3129 
  3130 * On instantiation of classes, remaining undefined class parameters
  3131 are formally declared.  INCOMPATIBILITY.
  3132 
  3133 
  3134 *** Document preparation ***
  3135 
  3136 * New generalized style concept for printing terms: @{foo (style) ...}
  3137 instead of @{foo_style style ...}  (old form is still retained for
  3138 backward compatibility).  Styles can be also applied for
  3139 antiquotations prop, term_type and typeof.
  3140 
  3141 
  3142 *** HOL ***
  3143 
  3144 * New proof method "smt" for a combination of first-order logic with
  3145 equality, linear and nonlinear (natural/integer/real) arithmetic, and
  3146 fixed-size bitvectors; there is also basic support for higher-order
  3147 features (esp. lambda abstractions).  It is an incomplete decision
  3148 procedure based on external SMT solvers using the oracle mechanism;
  3149 for the SMT solver Z3, this method is proof-producing.  Certificates
  3150 are provided to avoid calling the external solvers solely for
  3151 re-checking proofs.  Due to a remote SMT service there is no need for
  3152 installing SMT solvers locally.  See src/HOL/SMT.
  3153 
  3154 * New commands to load and prove verification conditions generated by
  3155 the Boogie program verifier or derived systems (e.g. the Verifying C
  3156 Compiler (VCC) or Spec#).  See src/HOL/Boogie.
  3157 
  3158 * New counterexample generator tool 'nitpick' based on the Kodkod
  3159 relational model finder.  See src/HOL/Tools/Nitpick and
  3160 src/HOL/Nitpick_Examples.
  3161 
  3162 * New commands 'code_pred' and 'values' to invoke the predicate
  3163 compiler and to enumerate values of inductive predicates.
  3164 
  3165 * A tabled implementation of the reflexive transitive closure.
  3166 
  3167 * New implementation of quickcheck uses generic code generator;
  3168 default generators are provided for all suitable HOL types, records
  3169 and datatypes.  Old quickcheck can be re-activated importing theory
  3170 Library/SML_Quickcheck.
  3171 
  3172 * New testing tool Mirabelle for automated proof tools.  Applies
  3173 several tools and tactics like sledgehammer, metis, or quickcheck, to
  3174 every proof step in a theory.  To be used in batch mode via the
  3175 "mirabelle" utility.
  3176 
  3177 * New proof method "sos" (sum of squares) for nonlinear real
  3178 arithmetic (originally due to John Harison). It requires theory
  3179 Library/Sum_Of_Squares.  It is not a complete decision procedure but
  3180 works well in practice on quantifier-free real arithmetic with +, -,
  3181 *, ^, =, <= and <, i.e. boolean combinations of equalities and
  3182 inequalities between polynomials.  It makes use of external
  3183 semidefinite programming solvers.  Method "sos" generates a
  3184 certificate that can be pasted into the proof thus avoiding the need
  3185 to call an external tool every time the proof is checked.  See
  3186 src/HOL/Library/Sum_Of_Squares.
  3187 
  3188 * New method "linarith" invokes existing linear arithmetic decision
  3189 procedure only.
  3190 
  3191 * New command 'atp_minimal' reduces result produced by Sledgehammer.
  3192 
  3193 * New Sledgehammer option "Full Types" in Proof General settings menu.
  3194 Causes full type information to be output to the ATPs.  This slows
  3195 ATPs down considerably but eliminates a source of unsound "proofs"
  3196 that fail later.
  3197 
  3198 * New method "metisFT": A version of metis that uses full type
  3199 information in order to avoid failures of proof reconstruction.
  3200 
  3201 * New evaluator "approximate" approximates an real valued term using
  3202 the same method as the approximation method.
  3203 
  3204 * Method "approximate" now supports arithmetic expressions as
  3205 boundaries of intervals and implements interval splitting and Taylor
  3206 series expansion.
  3207 
  3208 * ML antiquotation @{code_datatype} inserts definition of a datatype
  3209 generated by the code generator; e.g. see src/HOL/Predicate.thy.
  3210 
  3211 * New theory SupInf of the supremum and infimum operators for sets of
  3212 reals.
  3213 
  3214 * New theory Probability, which contains a development of measure
  3215 theory, eventually leading to Lebesgue integration and probability.
  3216 
  3217 * Extended Multivariate Analysis to include derivation and Brouwer's
  3218 fixpoint theorem.
  3219 
  3220 * Reorganization of number theory, INCOMPATIBILITY:
  3221   - new number theory development for nat and int, in theories Divides
  3222     and GCD as well as in new session Number_Theory
  3223   - some constants and facts now suffixed with _nat and _int
  3224     accordingly
  3225   - former session NumberTheory now named Old_Number_Theory, including
  3226     theories Legacy_GCD and Primes (prefer Number_Theory if possible)
  3227   - moved theory Pocklington from src/HOL/Library to
  3228     src/HOL/Old_Number_Theory
  3229 
  3230 * Theory GCD includes functions Gcd/GCD and Lcm/LCM for the gcd and
  3231 lcm of finite and infinite sets. It is shown that they form a complete
  3232 lattice.
  3233 
  3234 * Class semiring_div requires superclass no_zero_divisors and proof of
  3235 div_mult_mult1; theorems div_mult_mult1, div_mult_mult2,
  3236 div_mult_mult1_if, div_mult_mult1 and div_mult_mult2 have been
  3237 generalized to class semiring_div, subsuming former theorems
  3238 zdiv_zmult_zmult1, zdiv_zmult_zmult1_if, zdiv_zmult_zmult1 and
  3239 zdiv_zmult_zmult2.  div_mult_mult1 is now [simp] by default.
  3240 INCOMPATIBILITY.
  3241 
  3242 * Refinements to lattice classes and sets:
  3243   - less default intro/elim rules in locale variant, more default
  3244     intro/elim rules in class variant: more uniformity
  3245   - lemma ge_sup_conv renamed to le_sup_iff, in accordance with
  3246     le_inf_iff
  3247   - dropped lemma alias inf_ACI for inf_aci (same for sup_ACI and
  3248     sup_aci)
  3249   - renamed ACI to inf_sup_aci
  3250   - new class "boolean_algebra"
  3251   - class "complete_lattice" moved to separate theory
  3252     "Complete_Lattice"; corresponding constants (and abbreviations)
  3253     renamed and with authentic syntax:
  3254     Set.Inf ~>    Complete_Lattice.Inf
  3255     Set.Sup ~>    Complete_Lattice.Sup
  3256     Set.INFI ~>   Complete_Lattice.INFI
  3257     Set.SUPR ~>   Complete_Lattice.SUPR
  3258     Set.Inter ~>  Complete_Lattice.Inter
  3259     Set.Union ~>  Complete_Lattice.Union
  3260     Set.INTER ~>  Complete_Lattice.INTER
  3261     Set.UNION ~>  Complete_Lattice.UNION
  3262   - authentic syntax for
  3263     Set.Pow
  3264     Set.image
  3265   - mere abbreviations:
  3266     Set.empty               (for bot)
  3267     Set.UNIV                (for top)
  3268     Set.inter               (for inf, formerly Set.Int)
  3269     Set.union               (for sup, formerly Set.Un)
  3270     Complete_Lattice.Inter  (for Inf)
  3271     Complete_Lattice.Union  (for Sup)
  3272     Complete_Lattice.INTER  (for INFI)
  3273     Complete_Lattice.UNION  (for SUPR)
  3274   - object-logic definitions as far as appropriate
  3275 
  3276 INCOMPATIBILITY.  Care is required when theorems Int_subset_iff or
  3277 Un_subset_iff are explicitly deleted as default simp rules; then also
  3278 their lattice counterparts le_inf_iff and le_sup_iff have to be
  3279 deleted to achieve the desired effect.
  3280 
  3281 * Rules inf_absorb1, inf_absorb2, sup_absorb1, sup_absorb2 are no simp
  3282 rules by default any longer; the same applies to min_max.inf_absorb1
  3283 etc.  INCOMPATIBILITY.
  3284 
  3285 * Rules sup_Int_eq and sup_Un_eq are no longer declared as
  3286 pred_set_conv by default.  INCOMPATIBILITY.
  3287 
  3288 * Power operations on relations and functions are now one dedicated
  3289 constant "compow" with infix syntax "^^".  Power operation on
  3290 multiplicative monoids retains syntax "^" and is now defined generic
  3291 in class power.  INCOMPATIBILITY.
  3292 
  3293 * Relation composition "R O S" now has a more standard argument order:
  3294 "R O S = {(x, z). EX y. (x, y) : R & (y, z) : S}".  INCOMPATIBILITY,
  3295 rewrite propositions with "S O R" --> "R O S". Proofs may occasionally
  3296 break, since the O_assoc rule was not rewritten like this.  Fix using
  3297 O_assoc[symmetric].  The same applies to the curried version "R OO S".
  3298 
  3299 * Function "Inv" is renamed to "inv_into" and function "inv" is now an
  3300 abbreviation for "inv_into UNIV".  Lemmas are renamed accordingly.
  3301 INCOMPATIBILITY.
  3302 
  3303 * Most rules produced by inductive and datatype package have mandatory
  3304 prefixes.  INCOMPATIBILITY.
  3305 
  3306 * Changed "DERIV_intros" to a dynamic fact, which can be augmented by
  3307 the attribute of the same name.  Each of the theorems in the list
  3308 DERIV_intros assumes composition with an additional function and
  3309 matches a variable to the derivative, which has to be solved by the
  3310 Simplifier.  Hence (auto intro!: DERIV_intros) computes the derivative
  3311 of most elementary terms.  Former Maclauren.DERIV_tac and
  3312 Maclauren.deriv_tac should be replaced by (auto intro!: DERIV_intros).
  3313 INCOMPATIBILITY.
  3314 
  3315 * Code generator attributes follow the usual underscore convention:
  3316     code_unfold     replaces    code unfold
  3317     code_post       replaces    code post
  3318     etc.
  3319   INCOMPATIBILITY.
  3320 
  3321 * Renamed methods:
  3322     sizechange -> size_change
  3323     induct_scheme -> induction_schema
  3324   INCOMPATIBILITY.
  3325 
  3326 * Discontinued abbreviation "arbitrary" of constant "undefined".
  3327 INCOMPATIBILITY, use "undefined" directly.
  3328 
  3329 * Renamed theorems:
  3330     Suc_eq_add_numeral_1 -> Suc_eq_plus1
  3331     Suc_eq_add_numeral_1_left -> Suc_eq_plus1_left
  3332     Suc_plus1 -> Suc_eq_plus1
  3333     *anti_sym -> *antisym*
  3334     vector_less_eq_def -> vector_le_def
  3335   INCOMPATIBILITY.
  3336 
  3337 * Added theorem List.map_map as [simp].  Removed List.map_compose.
  3338 INCOMPATIBILITY.
  3339 
  3340 * Removed predicate "M hassize n" (<--> card M = n & finite M).
  3341 INCOMPATIBILITY.
  3342 
  3343 
  3344 *** HOLCF ***
  3345 
  3346 * Theory Representable defines a class "rep" of domains that are
  3347 representable (via an ep-pair) in the universal domain type "udom".
  3348 Instances are provided for all type constructors defined in HOLCF.
  3349 
  3350 * The 'new_domain' command is a purely definitional version of the
  3351 domain package, for representable domains.  Syntax is identical to the
  3352 old domain package.  The 'new_domain' package also supports indirect
  3353 recursion using previously-defined type constructors.  See
  3354 src/HOLCF/ex/New_Domain.thy for examples.
  3355 
  3356 * Method "fixrec_simp" unfolds one step of a fixrec-defined constant
  3357 on the left-hand side of an equation, and then performs
  3358 simplification.  Rewriting is done using rules declared with the
  3359 "fixrec_simp" attribute.  The "fixrec_simp" method is intended as a
  3360 replacement for "fixpat"; see src/HOLCF/ex/Fixrec_ex.thy for examples.
  3361 
  3362 * The pattern-match compiler in 'fixrec' can now handle constructors
  3363 with HOL function types.  Pattern-match combinators for the Pair
  3364 constructor are pre-configured.
  3365 
  3366 * The 'fixrec' package now produces better fixed-point induction rules
  3367 for mutually-recursive definitions:  Induction rules have conclusions
  3368 of the form "P foo bar" instead of "P <foo, bar>".
  3369 
  3370 * The constant "sq_le" (with infix syntax "<<" or "\<sqsubseteq>") has
  3371 been renamed to "below".  The name "below" now replaces "less" in many
  3372 theorem names.  (Legacy theorem names using "less" are still supported
  3373 as well.)
  3374 
  3375 * The 'fixrec' package now supports "bottom patterns".  Bottom
  3376 patterns can be used to generate strictness rules, or to make
  3377 functions more strict (much like the bang-patterns supported by the
  3378 Glasgow Haskell Compiler).  See src/HOLCF/ex/Fixrec_ex.thy for
  3379 examples.
  3380 
  3381 
  3382 *** ML ***
  3383 
  3384 * Support for Poly/ML 5.3.0, with improved reporting of compiler
  3385 errors and run-time exceptions, including detailed source positions.
  3386 
  3387 * Structure Name_Space (formerly NameSpace) now manages uniquely
  3388 identified entries, with some additional information such as source
  3389 position, logical grouping etc.
  3390 
  3391 * Theory and context data is now introduced by the simplified and
  3392 modernized functors Theory_Data, Proof_Data, Generic_Data.  Data needs
  3393 to be pure, but the old TheoryDataFun for mutable data (with explicit
  3394 copy operation) is still available for some time.
  3395 
  3396 * Structure Synchronized (cf. src/Pure/Concurrent/synchronized.ML)
  3397 provides a high-level programming interface to synchronized state
  3398 variables with atomic update.  This works via pure function
  3399 application within a critical section -- its runtime should be as
  3400 short as possible; beware of deadlocks if critical code is nested,
  3401 either directly or indirectly via other synchronized variables!
  3402 
  3403 * Structure Unsynchronized (cf. src/Pure/ML-Systems/unsynchronized.ML)
  3404 wraps raw ML references, explicitly indicating their non-thread-safe
  3405 behaviour.  The Isar toplevel keeps this structure open, to
  3406 accommodate Proof General as well as quick and dirty interactive
  3407 experiments with references.
  3408 
  3409 * PARALLEL_CHOICE and PARALLEL_GOALS provide basic support for
  3410 parallel tactical reasoning.
  3411 
  3412 * Tacticals Subgoal.FOCUS, Subgoal.FOCUS_PREMS, Subgoal.FOCUS_PARAMS
  3413 are similar to SUBPROOF, but are slightly more flexible: only the
  3414 specified parts of the subgoal are imported into the context, and the
  3415 body tactic may introduce new subgoals and schematic variables.
  3416 
  3417 * Old tactical METAHYPS, which does not observe the proof context, has
  3418 been renamed to Old_Goals.METAHYPS and awaits deletion.  Use SUBPROOF
  3419 or Subgoal.FOCUS etc.
  3420 
  3421 * Renamed functor TableFun to Table, and GraphFun to Graph.  (Since
  3422 functors have their own ML name space there is no point to mark them
  3423 separately.)  Minor INCOMPATIBILITY.
  3424 
  3425 * Renamed NamedThmsFun to Named_Thms.  INCOMPATIBILITY.
  3426 
  3427 * Renamed several structures FooBar to Foo_Bar.  Occasional,
  3428 INCOMPATIBILITY.
  3429 
  3430 * Operations of structure Skip_Proof no longer require quick_and_dirty
  3431 mode, which avoids critical setmp.
  3432 
  3433 * Eliminated old Attrib.add_attributes, Method.add_methods and related
  3434 combinators for "args".  INCOMPATIBILITY, need to use simplified
  3435 Attrib/Method.setup introduced in Isabelle2009.
  3436 
  3437 * Proper context for simpset_of, claset_of, clasimpset_of.  May fall
  3438 back on global_simpset_of, global_claset_of, global_clasimpset_of as
  3439 last resort.  INCOMPATIBILITY.
  3440 
  3441 * Display.pretty_thm now requires a proper context (cf. former
  3442 ProofContext.pretty_thm).  May fall back on Display.pretty_thm_global
  3443 or even Display.pretty_thm_without_context as last resort.
  3444 INCOMPATIBILITY.
  3445 
  3446 * Discontinued Display.pretty_ctyp/cterm etc.  INCOMPATIBILITY, use
  3447 Syntax.pretty_typ/term directly, preferably with proper context
  3448 instead of global theory.
  3449 
  3450 
  3451 *** System ***
  3452 
  3453 * Further fine tuning of parallel proof checking, scales up to 8 cores
  3454 (max. speedup factor 5.0).  See also Goal.parallel_proofs in ML and
  3455 usedir option -q.
  3456 
  3457 * Support for additional "Isabelle components" via etc/components, see
  3458 also the system manual.
  3459 
  3460 * The isabelle makeall tool now operates on all components with
  3461 IsaMakefile, not just hardwired "logics".
  3462 
  3463 * Removed "compress" option from isabelle-process and isabelle usedir;
  3464 this is always enabled.
  3465 
  3466 * Discontinued support for Poly/ML 4.x versions.
  3467 
  3468 * Isabelle tool "wwwfind" provides web interface for 'find_theorems'
  3469 on a given logic image.  This requires the lighttpd webserver and is
  3470 currently supported on Linux only.
  3471 
  3472 
  3473 
  3474 New in Isabelle2009 (April 2009)
  3475 --------------------------------
  3476 
  3477 *** General ***
  3478 
  3479 * Simplified main Isabelle executables, with less surprises on
  3480 case-insensitive file-systems (such as Mac OS).
  3481 
  3482   - The main Isabelle tool wrapper is now called "isabelle" instead of
  3483     "isatool."
  3484 
  3485   - The former "isabelle" alias for "isabelle-process" has been
  3486     removed (should rarely occur to regular users).
  3487 
  3488   - The former "isabelle-interface" and its alias "Isabelle" have been
  3489     removed (interfaces are now regular Isabelle tools).
  3490 
  3491 Within scripts and make files, the Isabelle environment variables
  3492 ISABELLE_TOOL and ISABELLE_PROCESS replace old ISATOOL and ISABELLE,
  3493 respectively.  (The latter are still available as legacy feature.)
  3494 
  3495 The old isabelle-interface wrapper could react in confusing ways if
  3496 the interface was uninstalled or changed otherwise.  Individual
  3497 interface tool configuration is now more explicit, see also the
  3498 Isabelle system manual.  In particular, Proof General is now available
  3499 via "isabelle emacs".
  3500 
  3501 INCOMPATIBILITY, need to adapt derivative scripts.  Users may need to
  3502 purge installed copies of Isabelle executables and re-run "isabelle
  3503 install -p ...", or use symlinks.
  3504 
  3505 * The default for ISABELLE_HOME_USER is now ~/.isabelle instead of the
  3506 old ~/isabelle, which was slightly non-standard and apt to cause
  3507 surprises on case-insensitive file-systems (such as Mac OS).
  3508 
  3509 INCOMPATIBILITY, need to move existing ~/isabelle/etc,
  3510 ~/isabelle/heaps, ~/isabelle/browser_info to the new place.  Special
  3511 care is required when using older releases of Isabelle.  Note that
  3512 ISABELLE_HOME_USER can be changed in Isabelle/etc/settings of any
  3513 Isabelle distribution, in order to use the new ~/.isabelle uniformly.
  3514 
  3515 * Proofs of fully specified statements are run in parallel on
  3516 multi-core systems.  A speedup factor of 2.5 to 3.2 can be expected on
  3517 a regular 4-core machine, if the initial heap space is made reasonably
  3518 large (cf. Poly/ML option -H).  (Requires Poly/ML 5.2.1 or later.)
  3519 
  3520 * The main reference manuals ("isar-ref", "implementation", and
  3521 "system") have been updated and extended.  Formally checked references
  3522 as hyperlinks are now available uniformly.
  3523 
  3524 
  3525 *** Pure ***
  3526 
  3527 * Complete re-implementation of locales.  INCOMPATIBILITY in several
  3528 respects.  The most important changes are listed below.  See the
  3529 Tutorial on Locales ("locales" manual) for details.
  3530 
  3531 - In locale expressions, instantiation replaces renaming.  Parameters
  3532 must be declared in a for clause.  To aid compatibility with previous
  3533 parameter inheritance, in locale declarations, parameters that are not
  3534 'touched' (instantiation position "_" or omitted) are implicitly added
  3535 with their syntax at the beginning of the for clause.
  3536 
  3537 - Syntax from abbreviations and definitions in locales is available in
  3538 locale expressions and context elements.  The latter is particularly
  3539 useful in locale declarations.
  3540 
  3541 - More flexible mechanisms to qualify names generated by locale
  3542 expressions.  Qualifiers (prefixes) may be specified in locale
  3543 expressions, and can be marked as mandatory (syntax: "name!:") or
  3544 optional (syntax "name?:").  The default depends for plain "name:"
  3545 depends on the situation where a locale expression is used: in
  3546 commands 'locale' and 'sublocale' prefixes are optional, in
  3547 'interpretation' and 'interpret' prefixes are mandatory.  The old
  3548 implicit qualifiers derived from the parameter names of a locale are
  3549 no longer generated.
  3550 
  3551 - Command "sublocale l < e" replaces "interpretation l < e".  The
  3552 instantiation clause in "interpretation" and "interpret" (square
  3553 brackets) is no longer available.  Use locale expressions.
  3554 
  3555 - When converting proof scripts, mandatory qualifiers in
  3556 'interpretation' and 'interpret' should be retained by default, even
  3557 if this is an INCOMPATIBILITY compared to former behavior.  In the
  3558 worst case, use the "name?:" form for non-mandatory ones.  Qualifiers
  3559 in locale expressions range over a single locale instance only.
  3560 
  3561 - Dropped locale element "includes".  This is a major INCOMPATIBILITY.
  3562 In existing theorem specifications replace the includes element by the
  3563 respective context elements of the included locale, omitting those
  3564 that are already present in the theorem specification.  Multiple
  3565 assume elements of a locale should be replaced by a single one
  3566 involving the locale predicate.  In the proof body, declarations (most
  3567 notably theorems) may be regained by interpreting the respective
  3568 locales in the proof context as required (command "interpret").
  3569 
  3570 If using "includes" in replacement of a target solely because the
  3571 parameter types in the theorem are not as general as in the target,
  3572 consider declaring a new locale with additional type constraints on
  3573 the parameters (context element "constrains").
  3574 
  3575 - Discontinued "locale (open)".  INCOMPATIBILITY.
  3576 
  3577 - Locale interpretation commands no longer attempt to simplify goal.
  3578 INCOMPATIBILITY: in rare situations the generated goal differs.  Use
  3579 methods intro_locales and unfold_locales to clarify.
  3580 
  3581 - Locale interpretation commands no longer accept interpretation
  3582 attributes.  INCOMPATIBILITY.
  3583 
  3584 * Class declaration: so-called "base sort" must not be given in import
  3585 list any longer, but is inferred from the specification.  Particularly
  3586 in HOL, write
  3587 
  3588     class foo = ...
  3589 
  3590 instead of
  3591 
  3592     class foo = type + ...
  3593 
  3594 * Class target: global versions of theorems stemming do not carry a
  3595 parameter prefix any longer.  INCOMPATIBILITY.
  3596 
  3597 * Class 'instance' command no longer accepts attached definitions.
  3598 INCOMPATIBILITY, use proper 'instantiation' target instead.
  3599 
  3600 * Recovered hiding of consts, which was accidentally broken in
  3601 Isabelle2007.  Potential INCOMPATIBILITY, ``hide const c'' really
  3602 makes c inaccessible; consider using ``hide (open) const c'' instead.
  3603 
  3604 * Slightly more coherent Pure syntax, with updated documentation in
  3605 isar-ref manual.  Removed locales meta_term_syntax and
  3606 meta_conjunction_syntax: TERM and &&& (formerly &&) are now permanent,
  3607 INCOMPATIBILITY in rare situations.  Note that &&& should not be used
  3608 directly in regular applications.
  3609 
  3610 * There is a new syntactic category "float_const" for signed decimal
  3611 fractions (e.g. 123.45 or -123.45).
  3612 
  3613 * Removed exotic 'token_translation' command.  INCOMPATIBILITY, use ML
  3614 interface with 'setup' command instead.
  3615 
  3616 * Command 'local_setup' is similar to 'setup', but operates on a local
  3617 theory context.
  3618 
  3619 * The 'axiomatization' command now only works within a global theory
  3620 context.  INCOMPATIBILITY.
  3621 
  3622 * Goal-directed proof now enforces strict proof irrelevance wrt. sort
  3623 hypotheses.  Sorts required in the course of reasoning need to be
  3624 covered by the constraints in the initial statement, completed by the
  3625 type instance information of the background theory.  Non-trivial sort
  3626 hypotheses, which rarely occur in practice, may be specified via
  3627 vacuous propositions of the form SORT_CONSTRAINT('a::c).  For example:
  3628 
  3629   lemma assumes "SORT_CONSTRAINT('a::empty)" shows False ...
  3630 
  3631 The result contains an implicit sort hypotheses as before --
  3632 SORT_CONSTRAINT premises are eliminated as part of the canonical rule
  3633 normalization.
  3634 
  3635 * Generalized Isar history, with support for linear undo, direct state
  3636 addressing etc.
  3637 
  3638 * Changed defaults for unify configuration options:
  3639 
  3640   unify_trace_bound = 50 (formerly 25)
  3641   unify_search_bound = 60 (formerly 30)
  3642 
  3643 * Different bookkeeping for code equations (INCOMPATIBILITY):
  3644 
  3645   a) On theory merge, the last set of code equations for a particular
  3646      constant is taken (in accordance with the policy applied by other
  3647      parts of the code generator framework).
  3648 
  3649   b) Code equations stemming from explicit declarations (e.g. code
  3650      attribute) gain priority over default code equations stemming
  3651      from definition, primrec, fun etc.
  3652 
  3653 * Keyword 'code_exception' now named 'code_abort'.  INCOMPATIBILITY.
  3654 
  3655 * Unified theorem tables for both code generators.  Thus [code
  3656 func] has disappeared and only [code] remains.  INCOMPATIBILITY.
  3657 
  3658 * Command 'find_consts' searches for constants based on type and name
  3659 patterns, e.g.
  3660 
  3661     find_consts "_ => bool"
  3662 
  3663 By default, matching is against subtypes, but it may be restricted to
  3664 the whole type.  Searching by name is possible.  Multiple queries are
  3665 conjunctive and queries may be negated by prefixing them with a
  3666 hyphen:
  3667 
  3668     find_consts strict: "_ => bool" name: "Int" -"int => int"
  3669 
  3670 * New 'find_theorems' criterion "solves" matches theorems that
  3671 directly solve the current goal (modulo higher-order unification).
  3672 
  3673 * Auto solve feature for main theorem statements: whenever a new goal
  3674 is stated, "find_theorems solves" is called; any theorems that could
  3675 solve the lemma directly are listed as part of the goal state.
  3676 Cf. associated options in Proof General Isabelle settings menu,
  3677 enabled by default, with reasonable timeout for pathological cases of
  3678 higher-order unification.
  3679 
  3680 
  3681 *** Document preparation ***
  3682 
  3683 * Antiquotation @{lemma} now imitates a regular terminal proof,
  3684 demanding keyword 'by' and supporting the full method expression
  3685 syntax just like the Isar command 'by'.
  3686 
  3687 
  3688 *** HOL ***
  3689 
  3690 * Integrated main parts of former image HOL-Complex with HOL.  Entry
  3691 points Main and Complex_Main remain as before.
  3692 
  3693 * Logic image HOL-Plain provides a minimal HOL with the most important
  3694 tools available (inductive, datatype, primrec, ...).  This facilitates
  3695 experimentation and tool development.  Note that user applications
  3696 (and library theories) should never refer to anything below theory
  3697 Main, as before.
  3698 
  3699 * Logic image HOL-Main stops at theory Main, and thus facilitates
  3700 experimentation due to shorter build times.
  3701 
  3702 * Logic image HOL-NSA contains theories of nonstandard analysis which
  3703 were previously part of former HOL-Complex.  Entry point Hyperreal
  3704 remains valid, but theories formerly using Complex_Main should now use
  3705 new entry point Hypercomplex.
  3706 
  3707 * Generic ATP manager for Sledgehammer, based on ML threads instead of
  3708 Posix processes.  Avoids potentially expensive forking of the ML
  3709 process.  New thread-based implementation also works on non-Unix
  3710 platforms (Cygwin).  Provers are no longer hardwired, but defined
  3711 within the theory via plain ML wrapper functions.  Basic Sledgehammer
  3712 commands are covered in the isar-ref manual.
  3713 
  3714 * Wrapper scripts for remote SystemOnTPTP service allows to use
  3715 sledgehammer without local ATP installation (Vampire etc.). Other
  3716 provers may be included via suitable ML wrappers, see also
  3717 src/HOL/ATP_Linkup.thy.
  3718 
  3719 * ATP selection (E/Vampire/Spass) is now via Proof General's settings
  3720 menu.
  3721 
  3722 * The metis method no longer fails because the theorem is too trivial
  3723 (contains the empty clause).
  3724 
  3725 * The metis method now fails in the usual manner, rather than raising
  3726 an exception, if it determines that it cannot prove the theorem.
  3727 
  3728 * Method "coherent" implements a prover for coherent logic (see also
  3729 src/Tools/coherent.ML).
  3730 
  3731 * Constants "undefined" and "default" replace "arbitrary".  Usually
  3732 "undefined" is the right choice to replace "arbitrary", though
  3733 logically there is no difference.  INCOMPATIBILITY.
  3734 
  3735 * Command "value" now integrates different evaluation mechanisms.  The
  3736 result of the first successful evaluation mechanism is printed.  In
  3737 square brackets a particular named evaluation mechanisms may be
  3738 specified (currently, [SML], [code] or [nbe]).  See further
  3739 src/HOL/ex/Eval_Examples.thy.
  3740 
  3741 * Normalization by evaluation now allows non-leftlinear equations.
  3742 Declare with attribute [code nbe].
  3743 
  3744 * Methods "case_tac" and "induct_tac" now refer to the very same rules
  3745 as the structured Isar versions "cases" and "induct", cf. the
  3746 corresponding "cases" and "induct" attributes.  Mutual induction rules
  3747 are now presented as a list of individual projections
  3748 (e.g. foo_bar.inducts for types foo and bar); the old format with
  3749 explicit HOL conjunction is no longer supported.  INCOMPATIBILITY, in
  3750 rare situations a different rule is selected --- notably nested tuple
  3751 elimination instead of former prod.exhaust: use explicit (case_tac t
  3752 rule: prod.exhaust) here.
  3753 
  3754 * Attributes "cases", "induct", "coinduct" support "del" option.
  3755 
  3756 * Removed fact "case_split_thm", which duplicates "case_split".
  3757 
  3758 * The option datatype has been moved to a new theory Option.  Renamed
  3759 option_map to Option.map, and o2s to Option.set, INCOMPATIBILITY.
  3760 
  3761 * New predicate "strict_mono" classifies strict functions on partial
  3762 orders.  With strict functions on linear orders, reasoning about
  3763 (in)equalities is facilitated by theorems "strict_mono_eq",
  3764 "strict_mono_less_eq" and "strict_mono_less".
  3765 
  3766 * Some set operations are now proper qualified constants with
  3767 authentic syntax.  INCOMPATIBILITY:
  3768 
  3769     op Int ~>   Set.Int
  3770     op Un ~>    Set.Un
  3771     INTER ~>    Set.INTER
  3772     UNION ~>    Set.UNION
  3773     Inter ~>    Set.Inter
  3774     Union ~>    Set.Union
  3775     {} ~>       Set.empty
  3776     UNIV ~>     Set.UNIV
  3777 
  3778 * Class complete_lattice with operations Inf, Sup, INFI, SUPR now in
  3779 theory Set.
  3780 
  3781 * Auxiliary class "itself" has disappeared -- classes without any
  3782 parameter are treated as expected by the 'class' command.
  3783 
  3784 * Leibnitz's Series for Pi and the arcus tangens and logarithm series.
  3785 
  3786 * Common decision procedures (Cooper, MIR, Ferrack, Approximation,
  3787 Dense_Linear_Order) are now in directory HOL/Decision_Procs.
  3788 
  3789 * Theory src/HOL/Decision_Procs/Approximation provides the new proof
  3790 method "approximation".  It proves formulas on real values by using
  3791 interval arithmetic.  In the formulas are also the transcendental
  3792 functions sin, cos, tan, atan, ln, exp and the constant pi are
  3793 allowed. For examples see
  3794 src/HOL/Descision_Procs/ex/Approximation_Ex.thy.
  3795 
  3796 * Theory "Reflection" now resides in HOL/Library.
  3797 
  3798 * Entry point to Word library now simply named "Word".
  3799 INCOMPATIBILITY.
  3800 
  3801 * Made source layout more coherent with logical distribution
  3802 structure:
  3803 
  3804     src/HOL/Library/RType.thy ~> src/HOL/Typerep.thy
  3805     src/HOL/Library/Code_Message.thy ~> src/HOL/
  3806     src/HOL/Library/GCD.thy ~> src/HOL/
  3807     src/HOL/Library/Order_Relation.thy ~> src/HOL/
  3808     src/HOL/Library/Parity.thy ~> src/HOL/
  3809     src/HOL/Library/Univ_Poly.thy ~> src/HOL/
  3810     src/HOL/Real/ContNotDenum.thy ~> src/HOL/Library/
  3811     src/HOL/Real/Lubs.thy ~> src/HOL/
  3812     src/HOL/Real/PReal.thy ~> src/HOL/
  3813     src/HOL/Real/Rational.thy ~> src/HOL/
  3814     src/HOL/Real/RComplete.thy ~> src/HOL/
  3815     src/HOL/Real/RealDef.thy ~> src/HOL/
  3816     src/HOL/Real/RealPow.thy ~> src/HOL/
  3817     src/HOL/Real/Real.thy ~> src/HOL/
  3818     src/HOL/Complex/Complex_Main.thy ~> src/HOL/
  3819     src/HOL/Complex/Complex.thy ~> src/HOL/
  3820     src/HOL/Complex/FrechetDeriv.thy ~> src/HOL/Library/
  3821     src/HOL/Complex/Fundamental_Theorem_Algebra.thy ~> src/HOL/Library/
  3822     src/HOL/Hyperreal/Deriv.thy ~> src/HOL/
  3823     src/HOL/Hyperreal/Fact.thy ~> src/HOL/
  3824     src/HOL/Hyperreal/Integration.thy ~> src/HOL/
  3825     src/HOL/Hyperreal/Lim.thy ~> src/HOL/
  3826     src/HOL/Hyperreal/Ln.thy ~> src/HOL/
  3827     src/HOL/Hyperreal/Log.thy ~> src/HOL/
  3828     src/HOL/Hyperreal/MacLaurin.thy ~> src/HOL/
  3829     src/HOL/Hyperreal/NthRoot.thy ~> src/HOL/
  3830     src/HOL/Hyperreal/Series.thy ~> src/HOL/
  3831     src/HOL/Hyperreal/SEQ.thy ~> src/HOL/
  3832     src/HOL/Hyperreal/Taylor.thy ~> src/HOL/
  3833     src/HOL/Hyperreal/Transcendental.thy ~> src/HOL/
  3834     src/HOL/Real/Float ~> src/HOL/Library/
  3835     src/HOL/Real/HahnBanach ~> src/HOL/HahnBanach
  3836     src/HOL/Real/RealVector.thy ~> src/HOL/
  3837 
  3838     src/HOL/arith_data.ML ~> src/HOL/Tools
  3839     src/HOL/hologic.ML ~> src/HOL/Tools
  3840     src/HOL/simpdata.ML ~> src/HOL/Tools
  3841     src/HOL/int_arith1.ML ~> src/HOL/Tools/int_arith.ML
  3842     src/HOL/int_factor_simprocs.ML ~> src/HOL/Tools
  3843     src/HOL/nat_simprocs.ML ~> src/HOL/Tools
  3844     src/HOL/Real/float_arith.ML ~> src/HOL/Tools
  3845     src/HOL/Real/float_syntax.ML ~> src/HOL/Tools
  3846     src/HOL/Real/rat_arith.ML ~> src/HOL/Tools
  3847     src/HOL/Real/real_arith.ML ~> src/HOL/Tools
  3848 
  3849     src/HOL/Library/Array.thy ~> src/HOL/Imperative_HOL
  3850     src/HOL/Library/Heap_Monad.thy ~> src/HOL/Imperative_HOL
  3851     src/HOL/Library/Heap.thy ~> src/HOL/Imperative_HOL
  3852     src/HOL/Library/Imperative_HOL.thy ~> src/HOL/Imperative_HOL
  3853     src/HOL/Library/Ref.thy ~> src/HOL/Imperative_HOL
  3854     src/HOL/Library/Relational.thy ~> src/HOL/Imperative_HOL
  3855 
  3856 * If methods "eval" and "evaluation" encounter a structured proof
  3857 state with !!/==>, only the conclusion is evaluated to True (if
  3858 possible), avoiding strange error messages.
  3859 
  3860 * Method "sizechange" automates termination proofs using (a
  3861 modification of) the size-change principle.  Requires SAT solver.  See
  3862 src/HOL/ex/Termination.thy for examples.
  3863 
  3864 * Simplifier: simproc for let expressions now unfolds if bound
  3865 variable occurs at most once in let expression body.  INCOMPATIBILITY.
  3866 
  3867 * Method "arith": Linear arithmetic now ignores all inequalities when
  3868 fast_arith_neq_limit is exceeded, instead of giving up entirely.
  3869 
  3870 * New attribute "arith" for facts that should always be used
  3871 automatically by arithmetic. It is intended to be used locally in
  3872 proofs, e.g.
  3873 
  3874   assumes [arith]: "x > 0"
  3875 
  3876 Global usage is discouraged because of possible performance impact.
  3877 
  3878 * New classes "top" and "bot" with corresponding operations "top" and
  3879 "bot" in theory Orderings; instantiation of class "complete_lattice"
  3880 requires instantiation of classes "top" and "bot".  INCOMPATIBILITY.
  3881 
  3882 * Changed definition lemma "less_fun_def" in order to provide an
  3883 instance for preorders on functions; use lemma "less_le" instead.
  3884 INCOMPATIBILITY.
  3885 
  3886 * Theory Orderings: class "wellorder" moved here, with explicit
  3887 induction rule "less_induct" as assumption.  For instantiation of
  3888 "wellorder" by means of predicate "wf", use rule wf_wellorderI.
  3889 INCOMPATIBILITY.
  3890 
  3891 * Theory Orderings: added class "preorder" as superclass of "order".
  3892 INCOMPATIBILITY: Instantiation proofs for order, linorder
  3893 etc. slightly changed.  Some theorems named order_class.* now named
  3894 preorder_class.*.
  3895 
  3896 * Theory Relation: renamed "refl" to "refl_on", "reflexive" to "refl,
  3897 "diag" to "Id_on".
  3898 
  3899 * Theory Finite_Set: added a new fold combinator of type
  3900 
  3901   ('a => 'b => 'b) => 'b => 'a set => 'b
  3902 
  3903 Occasionally this is more convenient than the old fold combinator
  3904 which is now defined in terms of the new one and renamed to
  3905 fold_image.
  3906 
  3907 * Theories Ring_and_Field and OrderedGroup: The lemmas "group_simps"
  3908 and "ring_simps" have been replaced by "algebra_simps" (which can be
  3909 extended with further lemmas!).  At the moment both still exist but
  3910 the former will disappear at some point.
  3911 
  3912 * Theory Power: Lemma power_Suc is now declared as a simp rule in
  3913 class recpower.  Type-specific simp rules for various recpower types
  3914 have been removed.  INCOMPATIBILITY, rename old lemmas as follows:
  3915 
  3916 rat_power_0    -> power_0
  3917 rat_power_Suc  -> power_Suc
  3918 realpow_0      -> power_0
  3919 realpow_Suc    -> power_Suc
  3920 complexpow_0   -> power_0
  3921 complexpow_Suc -> power_Suc
  3922 power_poly_0   -> power_0
  3923 power_poly_Suc -> power_Suc
  3924 
  3925 * Theories Ring_and_Field and Divides: Definition of "op dvd" has been
  3926 moved to separate class dvd in Ring_and_Field; a couple of lemmas on
  3927 dvd has been generalized to class comm_semiring_1.  Likewise a bunch
  3928 of lemmas from Divides has been generalized from nat to class
  3929 semiring_div.  INCOMPATIBILITY.  This involves the following theorem
  3930 renames resulting from duplicate elimination:
  3931 
  3932     dvd_def_mod ~>          dvd_eq_mod_eq_0
  3933     zero_dvd_iff ~>         dvd_0_left_iff
  3934     dvd_0 ~>                dvd_0_right
  3935     DIVISION_BY_ZERO_DIV ~> div_by_0
  3936     DIVISION_BY_ZERO_MOD ~> mod_by_0
  3937     mult_div ~>             div_mult_self2_is_id
  3938     mult_mod ~>             mod_mult_self2_is_0
  3939 
  3940 * Theory IntDiv: removed many lemmas that are instances of class-based
  3941 generalizations (from Divides and Ring_and_Field).  INCOMPATIBILITY,
  3942 rename old lemmas as follows:
  3943 
  3944 dvd_diff               -> nat_dvd_diff
  3945 dvd_zminus_iff         -> dvd_minus_iff
  3946 mod_add1_eq            -> mod_add_eq
  3947 mod_mult1_eq           -> mod_mult_right_eq
  3948 mod_mult1_eq'          -> mod_mult_left_eq
  3949 mod_mult_distrib_mod   -> mod_mult_eq
  3950 nat_mod_add_left_eq    -> mod_add_left_eq
  3951 nat_mod_add_right_eq   -> mod_add_right_eq
  3952 nat_mod_div_trivial    -> mod_div_trivial
  3953 nat_mod_mod_trivial    -> mod_mod_trivial
  3954 zdiv_zadd_self1        -> div_add_self1
  3955 zdiv_zadd_self2        -> div_add_self2
  3956 zdiv_zmult_self1       -> div_mult_self2_is_id
  3957 zdiv_zmult_self2       -> div_mult_self1_is_id
  3958 zdvd_triv_left         -> dvd_triv_left
  3959 zdvd_triv_right        -> dvd_triv_right
  3960 zdvd_zmult_cancel_disj -> dvd_mult_cancel_left
  3961 zmod_eq0_zdvd_iff      -> dvd_eq_mod_eq_0[symmetric]
  3962 zmod_zadd_left_eq      -> mod_add_left_eq
  3963 zmod_zadd_right_eq     -> mod_add_right_eq
  3964 zmod_zadd_self1        -> mod_add_self1
  3965 zmod_zadd_self2        -> mod_add_self2
  3966 zmod_zadd1_eq          -> mod_add_eq
  3967 zmod_zdiff1_eq         -> mod_diff_eq
  3968 zmod_zdvd_zmod         -> mod_mod_cancel
  3969 zmod_zmod_cancel       -> mod_mod_cancel
  3970 zmod_zmult_self1       -> mod_mult_self2_is_0
  3971 zmod_zmult_self2       -> mod_mult_self1_is_0
  3972 zmod_1                 -> mod_by_1
  3973 zdiv_1                 -> div_by_1
  3974 zdvd_abs1              -> abs_dvd_iff
  3975 zdvd_abs2              -> dvd_abs_iff
  3976 zdvd_refl              -> dvd_refl
  3977 zdvd_trans             -> dvd_trans
  3978 zdvd_zadd              -> dvd_add
  3979 zdvd_zdiff             -> dvd_diff
  3980 zdvd_zminus_iff        -> dvd_minus_iff
  3981 zdvd_zminus2_iff       -> minus_dvd_iff
  3982 zdvd_zmultD            -> dvd_mult_right
  3983 zdvd_zmultD2           -> dvd_mult_left
  3984 zdvd_zmult_mono        -> mult_dvd_mono
  3985 zdvd_0_right           -> dvd_0_right
  3986 zdvd_0_left            -> dvd_0_left_iff
  3987 zdvd_1_left            -> one_dvd
  3988 zminus_dvd_iff         -> minus_dvd_iff
  3989 
  3990 * Theory Rational: 'Fract k 0' now equals '0'.  INCOMPATIBILITY.
  3991 
  3992 * The real numbers offer decimal input syntax: 12.34 is translated
  3993 into 1234/10^2. This translation is not reversed upon output.
  3994 
  3995 * Theory Library/Polynomial defines an abstract type 'a poly of
  3996 univariate polynomials with coefficients of type 'a.  In addition to
  3997 the standard ring operations, it also supports div and mod.  Code
  3998 generation is also supported, using list-style constructors.
  3999 
  4000 * Theory Library/Inner_Product defines a class of real_inner for real
  4001 inner product spaces, with an overloaded operation inner :: 'a => 'a
  4002 => real.  Class real_inner is a subclass of real_normed_vector from
  4003 theory RealVector.
  4004 
  4005 * Theory Library/Product_Vector provides instances for the product
  4006 type 'a * 'b of several classes from RealVector and Inner_Product.
  4007 Definitions of addition, subtraction, scalar multiplication, norms,
  4008 and inner products are included.
  4009 
  4010 * Theory Library/Bit defines the field "bit" of integers modulo 2.  In
  4011 addition to the field operations, numerals and case syntax are also
  4012 supported.
  4013 
  4014 * Theory Library/Diagonalize provides constructive version of Cantor's
  4015 first diagonalization argument.
  4016 
  4017 * Theory Library/GCD: Curried operations gcd, lcm (for nat) and zgcd,
  4018 zlcm (for int); carried together from various gcd/lcm developements in
  4019 the HOL Distribution.  Constants zgcd and zlcm replace former igcd and
  4020 ilcm; corresponding theorems renamed accordingly.  INCOMPATIBILITY,
  4021 may recover tupled syntax as follows:
  4022 
  4023     hide (open) const gcd
  4024     abbreviation gcd where
  4025       "gcd == (%(a, b). GCD.gcd a b)"
  4026     notation (output)
  4027       GCD.gcd ("gcd '(_, _')")
  4028 
  4029 The same works for lcm, zgcd, zlcm.
  4030 
  4031 * Theory Library/Nat_Infinity: added addition, numeral syntax and more
  4032 instantiations for algebraic structures.  Removed some duplicate
  4033 theorems.  Changes in simp rules.  INCOMPATIBILITY.
  4034 
  4035 * ML antiquotation @{code} takes a constant as argument and generates
  4036 corresponding code in background and inserts name of the corresponding
  4037 resulting ML value/function/datatype constructor binding in place.
  4038 All occurrences of @{code} with a single ML block are generated
  4039 simultaneously.  Provides a generic and safe interface for
  4040 instrumentalizing code generation.  See
  4041 src/HOL/Decision_Procs/Ferrack.thy for a more ambitious application.
  4042 In future you ought to refrain from ad-hoc compiling generated SML
  4043 code on the ML toplevel.  Note that (for technical reasons) @{code}
  4044 cannot refer to constants for which user-defined serializations are
  4045 set.  Refer to the corresponding ML counterpart directly in that
  4046 cases.
  4047 
  4048 * Command 'rep_datatype': instead of theorem names the command now
  4049 takes a list of terms denoting the constructors of the type to be
  4050 represented as datatype.  The characteristic theorems have to be
  4051 proven.  INCOMPATIBILITY.  Also observe that the following theorems
  4052 have disappeared in favour of existing ones:
  4053 
  4054     unit_induct                 ~> unit.induct
  4055     prod_induct                 ~> prod.induct
  4056     sum_induct                  ~> sum.induct
  4057     Suc_Suc_eq                  ~> nat.inject
  4058     Suc_not_Zero Zero_not_Suc   ~> nat.distinct
  4059 
  4060 
  4061 *** HOL-Algebra ***
  4062 
  4063 * New locales for orders and lattices where the equivalence relation
  4064 is not restricted to equality.  INCOMPATIBILITY: all order and lattice
  4065 locales use a record structure with field eq for the equivalence.
  4066 
  4067 * New theory of factorial domains.
  4068 
  4069 * Units_l_inv and Units_r_inv are now simp rules by default.
  4070 INCOMPATIBILITY.  Simplifier proof that require deletion of l_inv
  4071 and/or r_inv will now also require deletion of these lemmas.
  4072 
  4073 * Renamed the following theorems, INCOMPATIBILITY:
  4074 
  4075 UpperD ~> Upper_memD
  4076 LowerD ~> Lower_memD
  4077 least_carrier ~> least_closed
  4078 greatest_carrier ~> greatest_closed
  4079 greatest_Lower_above ~> greatest_Lower_below
  4080 one_zero ~> carrier_one_zero
  4081 one_not_zero ~> carrier_one_not_zero  (collision with assumption)
  4082 
  4083 
  4084 *** HOL-Nominal ***
  4085 
  4086 * Nominal datatypes can now contain type-variables.
  4087 
  4088 * Commands 'nominal_inductive' and 'equivariance' work with local
  4089 theory targets.
  4090 
  4091 * Nominal primrec can now works with local theory targets and its
  4092 specification syntax now conforms to the general format as seen in
  4093 'inductive' etc.
  4094 
  4095 * Method "perm_simp" honours the standard simplifier attributes
  4096 (no_asm), (no_asm_use) etc.
  4097 
  4098 * The new predicate #* is defined like freshness, except that on the
  4099 left hand side can be a set or list of atoms.
  4100 
  4101 * Experimental command 'nominal_inductive2' derives strong induction
  4102 principles for inductive definitions.  In contrast to
  4103 'nominal_inductive', which can only deal with a fixed number of
  4104 binders, it can deal with arbitrary expressions standing for sets of
  4105 atoms to be avoided.  The only inductive definition we have at the
  4106 moment that needs this generalisation is the typing rule for Lets in
  4107 the algorithm W:
  4108 
  4109  Gamma |- t1 : T1   (x,close Gamma T1)::Gamma |- t2 : T2   x#Gamma
  4110  -----------------------------------------------------------------
  4111          Gamma |- Let x be t1 in t2 : T2
  4112 
  4113 In this rule one wants to avoid all the binders that are introduced by
  4114 "close Gamma T1".  We are looking for other examples where this
  4115 feature might be useful.  Please let us know.
  4116 
  4117 
  4118 *** HOLCF ***
  4119 
  4120 * Reimplemented the simplification procedure for proving continuity
  4121 subgoals.  The new simproc is extensible; users can declare additional
  4122 continuity introduction rules with the attribute [cont2cont].
  4123 
  4124 * The continuity simproc now uses a different introduction rule for
  4125 solving continuity subgoals on terms with lambda abstractions.  In
  4126 some rare cases the new simproc may fail to solve subgoals that the
  4127 old one could solve, and "simp add: cont2cont_LAM" may be necessary.
  4128 Potential INCOMPATIBILITY.
  4129 
  4130 * Command 'fixrec': specification syntax now conforms to the general
  4131 format as seen in 'inductive' etc.  See src/HOLCF/ex/Fixrec_ex.thy for
  4132 examples.  INCOMPATIBILITY.
  4133 
  4134 
  4135 *** ZF ***
  4136 
  4137 * Proof of Zorn's Lemma for partial orders.
  4138 
  4139 
  4140 *** ML ***
  4141 
  4142 * Multithreading for Poly/ML 5.1/5.2 is no longer supported, only for
  4143 Poly/ML 5.2.1 or later.  Important note: the TimeLimit facility
  4144 depends on multithreading, so timouts will not work before Poly/ML
  4145 5.2.1!
  4146 
  4147 * High-level support for concurrent ML programming, see
  4148 src/Pure/Cuncurrent.  The data-oriented model of "future values" is
  4149 particularly convenient to organize independent functional
  4150 computations.  The concept of "synchronized variables" provides a
  4151 higher-order interface for components with shared state, avoiding the
  4152 delicate details of mutexes and condition variables.  (Requires
  4153 Poly/ML 5.2.1 or later.)
  4154 
  4155 * ML bindings produced via Isar commands are stored within the Isar
  4156 context (theory or proof).  Consequently, commands like 'use' and 'ML'
  4157 become thread-safe and work with undo as expected (concerning
  4158 top-level bindings, not side-effects on global references).
  4159 INCOMPATIBILITY, need to provide proper Isar context when invoking the
  4160 compiler at runtime; really global bindings need to be given outside a
  4161 theory.  (Requires Poly/ML 5.2 or later.)
  4162 
  4163 * Command 'ML_prf' is analogous to 'ML' but works within a proof
  4164 context.  Top-level ML bindings are stored within the proof context in
  4165 a purely sequential fashion, disregarding the nested proof structure.
  4166 ML bindings introduced by 'ML_prf' are discarded at the end of the
  4167 proof.  (Requires Poly/ML 5.2 or later.)
  4168 
  4169 * Simplified ML attribute and method setup, cf. functions Attrib.setup
  4170 and Method.setup, as well as Isar commands 'attribute_setup' and
  4171 'method_setup'.  INCOMPATIBILITY for 'method_setup', need to simplify
  4172 existing code accordingly, or use plain 'setup' together with old
  4173 Method.add_method.
  4174 
  4175 * Simplified ML oracle interface Thm.add_oracle promotes 'a -> cterm
  4176 to 'a -> thm, while results are always tagged with an authentic oracle
  4177 name.  The Isar command 'oracle' is now polymorphic, no argument type
  4178 is specified.  INCOMPATIBILITY, need to simplify existing oracle code
  4179 accordingly.  Note that extra performance may be gained by producing
  4180 the cterm carefully, avoiding slow Thm.cterm_of.
  4181 
  4182 * Simplified interface for defining document antiquotations via
  4183 ThyOutput.antiquotation, ThyOutput.output, and optionally
  4184 ThyOutput.maybe_pretty_source.  INCOMPATIBILITY, need to simplify user
  4185 antiquotations accordingly, see src/Pure/Thy/thy_output.ML for common
  4186 examples.
  4187 
  4188 * More systematic treatment of long names, abstract name bindings, and
  4189 name space operations.  Basic operations on qualified names have been
  4190 move from structure NameSpace to Long_Name, e.g. Long_Name.base_name,
  4191 Long_Name.append.  Old type bstring has been mostly replaced by
  4192 abstract type binding (see structure Binding), which supports precise
  4193 qualification by packages and local theory targets, as well as proper
  4194 tracking of source positions.  INCOMPATIBILITY, need to wrap old
  4195 bstring values into Binding.name, or better pass through abstract
  4196 bindings everywhere.  See further src/Pure/General/long_name.ML,
  4197 src/Pure/General/binding.ML and src/Pure/General/name_space.ML
  4198 
  4199 * Result facts (from PureThy.note_thms, ProofContext.note_thms,
  4200 LocalTheory.note etc.) now refer to the *full* internal name, not the
  4201 bstring as before.  INCOMPATIBILITY, not detected by ML type-checking!
  4202 
  4203 * Disposed old type and term read functions (Sign.read_def_typ,
  4204 Sign.read_typ, Sign.read_def_terms, Sign.read_term,
  4205 Thm.read_def_cterms, Thm.read_cterm etc.).  INCOMPATIBILITY, should
  4206 use regular Syntax.read_typ, Syntax.read_term, Syntax.read_typ_global,
  4207 Syntax.read_term_global etc.; see also OldGoals.read_term as last
  4208 resort for legacy applications.
  4209 
  4210 * Disposed old declarations, tactics, tactic combinators that refer to
  4211 the simpset or claset of an implicit theory (such as Addsimps,
  4212 Simp_tac, SIMPSET).  INCOMPATIBILITY, should use @{simpset} etc. in
  4213 embedded ML text, or local_simpset_of with a proper context passed as
  4214 explicit runtime argument.
  4215 
  4216 * Rules and tactics that read instantiations (read_instantiate,
  4217 res_inst_tac, thin_tac, subgoal_tac etc.) now demand a proper proof
  4218 context, which is required for parsing and type-checking.  Moreover,
  4219 the variables are specified as plain indexnames, not string encodings
  4220 thereof.  INCOMPATIBILITY.
  4221 
  4222 * Generic Toplevel.add_hook interface allows to analyze the result of
  4223 transactions.  E.g. see src/Pure/ProofGeneral/proof_general_pgip.ML
  4224 for theorem dependency output of transactions resulting in a new
  4225 theory state.
  4226 
  4227 * ML antiquotations: block-structured compilation context indicated by
  4228 \<lbrace> ... \<rbrace>; additional antiquotation forms:
  4229 
  4230   @{binding name}                         - basic name binding
  4231   @{let ?pat = term}                      - term abbreviation (HO matching)
  4232   @{note name = fact}                     - fact abbreviation
  4233   @{thm fact}                             - singleton fact (with attributes)
  4234   @{thms fact}                            - general fact (with attributes)
  4235   @{lemma prop by method}                 - singleton goal
  4236   @{lemma prop by meth1 meth2}            - singleton goal
  4237   @{lemma prop1 ... propN by method}      - general goal
  4238   @{lemma prop1 ... propN by meth1 meth2} - general goal
  4239   @{lemma (open) ...}                     - open derivation
  4240 
  4241 
  4242 *** System ***
  4243 
  4244 * The Isabelle "emacs" tool provides a specific interface to invoke
  4245 Proof General / Emacs, with more explicit failure if that is not
  4246 installed (the old isabelle-interface script silently falls back on
  4247 isabelle-process).  The PROOFGENERAL_HOME setting determines the
  4248 installation location of the Proof General distribution.
  4249 
  4250 * Isabelle/lib/classes/Pure.jar provides basic support to integrate
  4251 the Isabelle process into a JVM/Scala application.  See
  4252 Isabelle/lib/jedit/plugin for a minimal example.  (The obsolete Java
  4253 process wrapper has been discontinued.)
  4254 
  4255 * Added homegrown Isabelle font with unicode layout, see lib/fonts.
  4256 
  4257 * Various status messages (with exact source position information) are
  4258 emitted, if proper markup print mode is enabled.  This allows
  4259 user-interface components to provide detailed feedback on internal
  4260 prover operations.
  4261 
  4262 
  4263 
  4264 New in Isabelle2008 (June 2008)
  4265 -------------------------------
  4266 
  4267 *** General ***
  4268 
  4269 * The Isabelle/Isar Reference Manual (isar-ref) has been reorganized
  4270 and updated, with formally checked references as hyperlinks.
  4271 
  4272 * Theory loader: use_thy (and similar operations) no longer set the
  4273 implicit ML context, which was occasionally hard to predict and in
  4274 conflict with concurrency.  INCOMPATIBILITY, use ML within Isar which
  4275 provides a proper context already.
  4276 
  4277 * Theory loader: old-style ML proof scripts being *attached* to a thy
  4278 file are no longer supported.  INCOMPATIBILITY, regular 'uses' and
  4279 'use' within a theory file will do the job.
  4280 
  4281 * Name space merge now observes canonical order, i.e. the second space
  4282 is inserted into the first one, while existing entries in the first
  4283 space take precedence.  INCOMPATIBILITY in rare situations, may try to
  4284 swap theory imports.
  4285 
  4286 * Syntax: symbol \<chi> is now considered a letter.  Potential
  4287 INCOMPATIBILITY in identifier syntax etc.
  4288 
  4289 * Outer syntax: string tokens no longer admit escaped white space,
  4290 which was an accidental (undocumented) feature.  INCOMPATIBILITY, use
  4291 white space without escapes.
  4292 
  4293 * Outer syntax: string tokens may contain arbitrary character codes
  4294 specified via 3 decimal digits (as in SML).  E.g. "foo\095bar" for
  4295 "foo_bar".
  4296 
  4297 
  4298 *** Pure ***
  4299 
  4300 * Context-dependent token translations.  Default setup reverts locally
  4301 fixed variables, and adds hilite markup for undeclared frees.
  4302 
  4303 * Unused theorems can be found using the new command 'unused_thms'.
  4304 There are three ways of invoking it:
  4305 
  4306 (1) unused_thms
  4307      Only finds unused theorems in the current theory.
  4308 
  4309 (2) unused_thms thy_1 ... thy_n -
  4310      Finds unused theorems in the current theory and all of its ancestors,
  4311      excluding the theories thy_1 ... thy_n and all of their ancestors.
  4312 
  4313 (3) unused_thms thy_1 ... thy_n - thy'_1 ... thy'_m
  4314      Finds unused theorems in the theories thy'_1 ... thy'_m and all of
  4315      their ancestors, excluding the theories thy_1 ... thy_n and all of
  4316      their ancestors.
  4317 
  4318 In order to increase the readability of the list produced by
  4319 unused_thms, theorems that have been created by a particular instance
  4320 of a theory command such as 'inductive' or 'function' are considered
  4321 to belong to the same "group", meaning that if at least one theorem in
  4322 this group is used, the other theorems in the same group are no longer
  4323 reported as unused.  Moreover, if all theorems in the group are
  4324 unused, only one theorem in the group is displayed.
  4325 
  4326 Note that proof objects have to be switched on in order for
  4327 unused_thms to work properly (i.e. !proofs must be >= 1, which is
  4328 usually the case when using Proof General with the default settings).
  4329 
  4330 * Authentic naming of facts disallows ad-hoc overwriting of previous
  4331 theorems within the same name space.  INCOMPATIBILITY, need to remove
  4332 duplicate fact bindings, or even accidental fact duplications.  Note
  4333 that tools may maintain dynamically scoped facts systematically, using
  4334 PureThy.add_thms_dynamic.
  4335 
  4336 * Command 'hide' now allows to hide from "fact" name space as well.
  4337 
  4338 * Eliminated destructive theorem database, simpset, claset, and
  4339 clasimpset.  Potential INCOMPATIBILITY, really need to observe linear
  4340 update of theories within ML code.
  4341 
  4342 * Eliminated theory ProtoPure and CPure, leaving just one Pure theory.
  4343 INCOMPATIBILITY, object-logics depending on former Pure require
  4344 additional setup PureThy.old_appl_syntax_setup; object-logics
  4345 depending on former CPure need to refer to Pure.
  4346 
  4347 * Commands 'use' and 'ML' are now purely functional, operating on
  4348 theory/local_theory.  Removed former 'ML_setup' (on theory), use 'ML'
  4349 instead.  Added 'ML_val' as mere diagnostic replacement for 'ML'.
  4350 INCOMPATIBILITY.
  4351 
  4352 * Command 'setup': discontinued implicit version with ML reference.
  4353 
  4354 * Instantiation target allows for simultaneous specification of class
  4355 instance operations together with an instantiation proof.
  4356 Type-checking phase allows to refer to class operations uniformly.
  4357 See src/HOL/Complex/Complex.thy for an Isar example and
  4358 src/HOL/Library/Eval.thy for an ML example.
  4359 
  4360 * Indexing of literal facts: be more serious about including only
  4361 facts from the visible specification/proof context, but not the
  4362 background context (locale etc.).  Affects `prop` notation and method
  4363 "fact".  INCOMPATIBILITY: need to name facts explicitly in rare
  4364 situations.
  4365 
  4366 * Method "cases", "induct", "coinduct": removed obsolete/undocumented
  4367 "(open)" option, which used to expose internal bound variables to the
  4368 proof text.
  4369 
  4370 * Isar statements: removed obsolete case "rule_context".
  4371 INCOMPATIBILITY, better use explicit fixes/assumes.
  4372 
  4373 * Locale proofs: default proof step now includes 'unfold_locales';
  4374 hence 'proof' without argument may be used to unfold locale
  4375 predicates.
  4376 
  4377 
  4378 *** Document preparation ***
  4379 
  4380 * Simplified pdfsetup.sty: color/hyperref is used unconditionally for
  4381 both pdf and dvi (hyperlinks usually work in xdvi as well); removed
  4382 obsolete thumbpdf setup (contemporary PDF viewers do this on the
  4383 spot); renamed link color from "darkblue" to "linkcolor" (default
  4384 value unchanged, can be redefined via \definecolor); no longer sets
  4385 "a4paper" option (unnecessary or even intrusive).
  4386 
  4387 * Antiquotation @{lemma A method} proves proposition A by the given
  4388 method (either a method name or a method name plus (optional) method
  4389 arguments in parentheses) and prints A just like @{prop A}.
  4390 
  4391 
  4392 *** HOL ***
  4393 
  4394 * New primrec package.  Specification syntax conforms in style to
  4395 definition/function/....  No separate induction rule is provided.  The
  4396 "primrec" command distinguishes old-style and new-style specifications
  4397 by syntax.  The former primrec package is now named OldPrimrecPackage.
  4398 When adjusting theories, beware: constants stemming from new-style
  4399 primrec specifications have authentic syntax.
  4400 
  4401 * Metis prover is now an order of magnitude faster, and also works
  4402 with multithreading.
  4403 
  4404 * Metis: the maximum number of clauses that can be produced from a
  4405 theorem is now given by the attribute max_clauses.  Theorems that
  4406 exceed this number are ignored, with a warning printed.
  4407 
  4408 * Sledgehammer no longer produces structured proofs by default. To
  4409 enable, declare [[sledgehammer_full = true]].  Attributes
  4410 reconstruction_modulus, reconstruction_sorts renamed
  4411 sledgehammer_modulus, sledgehammer_sorts.  INCOMPATIBILITY.
  4412 
  4413 * Method "induct_scheme" derives user-specified induction rules
  4414 from well-founded induction and completeness of patterns. This factors
  4415 out some operations that are done internally by the function package
  4416 and makes them available separately.  See
  4417 src/HOL/ex/Induction_Scheme.thy for examples.
  4418 
  4419 * More flexible generation of measure functions for termination
  4420 proofs: Measure functions can be declared by proving a rule of the
  4421 form "is_measure f" and giving it the [measure_function] attribute.
  4422 The "is_measure" predicate is logically meaningless (always true), and
  4423 just guides the heuristic.  To find suitable measure functions, the
  4424 termination prover sets up the goal "is_measure ?f" of the appropriate
  4425 type and generates all solutions by prolog-style backwards proof using
  4426 the declared rules.
  4427 
  4428 This setup also deals with rules like 
  4429 
  4430   "is_measure f ==> is_measure (list_size f)"
  4431 
  4432 which accommodates nested datatypes that recurse through lists.
  4433 Similar rules are predeclared for products and option types.
  4434 
  4435 * Turned the type of sets "'a set" into an abbreviation for "'a => bool"
  4436 
  4437   INCOMPATIBILITIES:
  4438 
  4439   - Definitions of overloaded constants on sets have to be replaced by
  4440     definitions on => and bool.
  4441 
  4442   - Some definitions of overloaded operators on sets can now be proved
  4443     using the definitions of the operators on => and bool.  Therefore,
  4444     the following theorems have been renamed:
  4445 
  4446       subset_def   -> subset_eq
  4447       psubset_def  -> psubset_eq
  4448       set_diff_def -> set_diff_eq
  4449       Compl_def    -> Compl_eq
  4450       Sup_set_def  -> Sup_set_eq
  4451       Inf_set_def  -> Inf_set_eq
  4452       sup_set_def  -> sup_set_eq
  4453       inf_set_def  -> inf_set_eq
  4454 
  4455   - Due to the incompleteness of the HO unification algorithm, some
  4456     rules such as subst may require manual instantiation, if some of
  4457     the unknowns in the rule is a set.
  4458 
  4459   - Higher order unification and forward proofs:
  4460     The proof pattern
  4461 
  4462       have "P (S::'a set)" <...>
  4463       then have "EX S. P S" ..
  4464 
  4465     no longer works (due to the incompleteness of the HO unification
  4466     algorithm) and must be replaced by the pattern
  4467 
  4468       have "EX S. P S"
  4469       proof
  4470         show "P S" <...>
  4471       qed
  4472 
  4473   - Calculational reasoning with subst (or similar rules):
  4474     The proof pattern
  4475 
  4476       have "P (S::'a set)" <...>
  4477       also have "S = T" <...>
  4478       finally have "P T" .
  4479 
  4480     no longer works (for similar reasons as the previous example) and
  4481     must be replaced by something like
  4482 
  4483       have "P (S::'a set)" <...>
  4484       moreover have "S = T" <...>
  4485       ultimately have "P T" by simp
  4486 
  4487   - Tactics or packages written in ML code:
  4488     Code performing pattern matching on types via
  4489 
  4490       Type ("set", [T]) => ...
  4491 
  4492     must be rewritten. Moreover, functions like strip_type or
  4493     binder_types no longer return the right value when applied to a
  4494     type of the form
  4495 
  4496       T1 => ... => Tn => U => bool
  4497 
  4498     rather than
  4499 
  4500       T1 => ... => Tn => U set
  4501 
  4502 * Merged theories Wellfounded_Recursion, Accessible_Part and
  4503 Wellfounded_Relations to theory Wellfounded.
  4504 
  4505 * Explicit class "eq" for executable equality.  INCOMPATIBILITY.
  4506 
  4507 * Class finite no longer treats UNIV as class parameter.  Use class
  4508 enum from theory Library/Enum instead to achieve a similar effect.
  4509 INCOMPATIBILITY.
  4510 
  4511 * Theory List: rule list_induct2 now has explicitly named cases "Nil"
  4512 and "Cons".  INCOMPATIBILITY.
  4513 
  4514 * HOL (and FOL): renamed variables in rules imp_elim and swap.
  4515 Potential INCOMPATIBILITY.
  4516 
  4517 * Theory Product_Type: duplicated lemmas split_Pair_apply and
  4518 injective_fst_snd removed, use split_eta and prod_eqI instead.
  4519 Renamed upd_fst to apfst and upd_snd to apsnd.  INCOMPATIBILITY.
  4520 
  4521 * Theory Nat: removed redundant lemmas that merely duplicate lemmas of
  4522 the same name in theory Orderings:
  4523 
  4524   less_trans
  4525   less_linear
  4526   le_imp_less_or_eq
  4527   le_less_trans
  4528   less_le_trans
  4529   less_not_sym
  4530   less_asym
  4531 
  4532 Renamed less_imp_le to less_imp_le_nat, and less_irrefl to
  4533 less_irrefl_nat.  Potential INCOMPATIBILITY due to more general types
  4534 and different variable names.
  4535 
  4536 * Library/Option_ord.thy: Canonical order on option type.
  4537 
  4538 * Library/RBT.thy: Red-black trees, an efficient implementation of
  4539 finite maps.
  4540 
  4541 * Library/Countable.thy: Type class for countable types.
  4542 
  4543 * Theory Int: The representation of numerals has changed.  The infix
  4544 operator BIT and the bit datatype with constructors B0 and B1 have
  4545 disappeared.  INCOMPATIBILITY, use "Int.Bit0 x" and "Int.Bit1 y" in
  4546 place of "x BIT bit.B0" and "y BIT bit.B1", respectively.  Theorems
  4547 involving BIT, B0, or B1 have been renamed with "Bit0" or "Bit1"
  4548 accordingly.
  4549 
  4550 * Theory Nat: definition of <= and < on natural numbers no longer
  4551 depend on well-founded relations.  INCOMPATIBILITY.  Definitions
  4552 le_def and less_def have disappeared.  Consider lemmas not_less
  4553 [symmetric, where ?'a = nat] and less_eq [symmetric] instead.
  4554 
  4555 * Theory Finite_Set: locales ACf, ACe, ACIf, ACIfSL and ACIfSLlin
  4556 (whose purpose mainly is for various fold_set functionals) have been
  4557 abandoned in favor of the existing algebraic classes
  4558 ab_semigroup_mult, comm_monoid_mult, ab_semigroup_idem_mult,
  4559 lower_semilattice (resp. upper_semilattice) and linorder.
  4560 INCOMPATIBILITY.
  4561 
  4562 * Theory Transitive_Closure: induct and cases rules now declare proper
  4563 case_names ("base" and "step").  INCOMPATIBILITY.
  4564 
  4565 * Theorem Inductive.lfp_ordinal_induct generalized to complete
  4566 lattices.  The form set-specific version is available as
  4567 Inductive.lfp_ordinal_induct_set.
  4568 
  4569 * Renamed theorems "power.simps" to "power_int.simps".
  4570 INCOMPATIBILITY.
  4571 
  4572 * Class semiring_div provides basic abstract properties of semirings
  4573 with division and modulo operations.  Subsumes former class dvd_mod.
  4574 
  4575 * Merged theories IntDef, Numeral and IntArith into unified theory
  4576 Int.  INCOMPATIBILITY.
  4577 
  4578 * Theory Library/Code_Index: type "index" now represents natural
  4579 numbers rather than integers.  INCOMPATIBILITY.
  4580 
  4581 * New class "uminus" with operation "uminus" (split of from class
  4582 "minus" which now only has operation "minus", binary).
  4583 INCOMPATIBILITY.
  4584 
  4585 * Constants "card", "internal_split", "option_map" now with authentic
  4586 syntax.  INCOMPATIBILITY.
  4587 
  4588 * Definitions subset_def, psubset_def, set_diff_def, Compl_def,
  4589 le_bool_def, less_bool_def, le_fun_def, less_fun_def, inf_bool_def,
  4590 sup_bool_def, Inf_bool_def, Sup_bool_def, inf_fun_def, sup_fun_def,
  4591 Inf_fun_def, Sup_fun_def, inf_set_def, sup_set_def, Inf_set_def,
  4592 Sup_set_def, le_def, less_def, option_map_def now with object
  4593 equality.  INCOMPATIBILITY.
  4594 
  4595 * Records. Removed K_record, and replaced it by pure lambda term
  4596 %x. c. The simplifier setup is now more robust against eta expansion.
  4597 INCOMPATIBILITY: in cases explicitly referring to K_record.
  4598 
  4599 * Library/Multiset: {#a, b, c#} abbreviates {#a#} + {#b#} + {#c#}.
  4600 
  4601 * Library/ListVector: new theory of arithmetic vector operations.
  4602 
  4603 * Library/Order_Relation: new theory of various orderings as sets of
  4604 pairs.  Defines preorders, partial orders, linear orders and
  4605 well-orders on sets and on types.
  4606 
  4607 
  4608 *** ZF ***
  4609 
  4610 * Renamed some theories to allow to loading both ZF and HOL in the
  4611 same session:
  4612 
  4613   Datatype  -> Datatype_ZF
  4614   Inductive -> Inductive_ZF
  4615   Int       -> Int_ZF
  4616   IntDiv    -> IntDiv_ZF
  4617   Nat       -> Nat_ZF
  4618   List      -> List_ZF
  4619   Main      -> Main_ZF
  4620 
  4621 INCOMPATIBILITY: ZF theories that import individual theories below
  4622 Main might need to be adapted.  Regular theory Main is still
  4623 available, as trivial extension of Main_ZF.
  4624 
  4625 
  4626 *** ML ***
  4627 
  4628 * ML within Isar: antiquotation @{const name} or @{const
  4629 name(typargs)} produces statically-checked Const term.
  4630 
  4631 * Functor NamedThmsFun: data is available to the user as dynamic fact
  4632 (of the same name).  Removed obsolete print command.
  4633 
  4634 * Removed obsolete "use_legacy_bindings" function.
  4635 
  4636 * The ``print mode'' is now a thread-local value derived from a global
  4637 template (the former print_mode reference), thus access becomes
  4638 non-critical.  The global print_mode reference is for session
  4639 management only; user-code should use print_mode_value,
  4640 print_mode_active, PrintMode.setmp etc.  INCOMPATIBILITY.
  4641 
  4642 * Functions system/system_out provide a robust way to invoke external
  4643 shell commands, with propagation of interrupts (requires Poly/ML
  4644 5.2.1).  Do not use OS.Process.system etc. from the basis library!
  4645 
  4646 
  4647 *** System ***
  4648 
  4649 * Default settings: PROOFGENERAL_OPTIONS no longer impose xemacs ---
  4650 in accordance with Proof General 3.7, which prefers GNU emacs.
  4651 
  4652 * isatool tty runs Isabelle process with plain tty interaction;
  4653 optional line editor may be specified via ISABELLE_LINE_EDITOR
  4654 setting, the default settings attempt to locate "ledit" and "rlwrap".
  4655 
  4656 * isatool browser now works with Cygwin as well, using general
  4657 "javapath" function defined in Isabelle process environment.
  4658 
  4659 * YXML notation provides a simple and efficient alternative to
  4660 standard XML transfer syntax.  See src/Pure/General/yxml.ML and
  4661 isatool yxml as described in the Isabelle system manual.
  4662 
  4663 * JVM class isabelle.IsabelleProcess (located in Isabelle/lib/classes)
  4664 provides general wrapper for managing an Isabelle process in a robust
  4665 fashion, with ``cooked'' output from stdin/stderr.
  4666 
  4667 * Rudimentary Isabelle plugin for jEdit (see Isabelle/lib/jedit),
  4668 based on Isabelle/JVM process wrapper (see Isabelle/lib/classes).
  4669 
  4670 * Removed obsolete THIS_IS_ISABELLE_BUILD feature.  NB: the documented
  4671 way of changing the user's settings is via
  4672 ISABELLE_HOME_USER/etc/settings, which is a fully featured bash
  4673 script.
  4674 
  4675 * Multithreading.max_threads := 0 refers to the number of actual CPU
  4676 cores of the underlying machine, which is a good starting point for
  4677 optimal performance tuning.  The corresponding usedir option -M allows
  4678 "max" as an alias for "0".  WARNING: does not work on certain versions
  4679 of Mac OS (with Poly/ML 5.1).
  4680 
  4681 * isabelle-process: non-ML sessions are run with "nice", to reduce the
  4682 adverse effect of Isabelle flooding interactive front-ends (notably
  4683 ProofGeneral / XEmacs).
  4684 
  4685 
  4686 
  4687 New in Isabelle2007 (November 2007)
  4688 -----------------------------------
  4689 
  4690 *** General ***
  4691 
  4692 * More uniform information about legacy features, notably a
  4693 warning/error of "Legacy feature: ...", depending on the state of the
  4694 tolerate_legacy_features flag (default true). FUTURE INCOMPATIBILITY:
  4695 legacy features will disappear eventually.
  4696 
  4697 * Theory syntax: the header format ``theory A = B + C:'' has been
  4698 discontinued in favour of ``theory A imports B C begin''.  Use isatool
  4699 fixheaders to convert existing theory files.  INCOMPATIBILITY.
  4700 
  4701 * Theory syntax: the old non-Isar theory file format has been
  4702 discontinued altogether.  Note that ML proof scripts may still be used
  4703 with Isar theories; migration is usually quite simple with the ML
  4704 function use_legacy_bindings.  INCOMPATIBILITY.
  4705 
  4706 * Theory syntax: some popular names (e.g. 'class', 'declaration',
  4707 'fun', 'help', 'if') are now keywords.  INCOMPATIBILITY, use double
  4708 quotes.
  4709 
  4710 * Theory loader: be more serious about observing the static theory
  4711 header specifications (including optional directories), but not the
  4712 accidental file locations of previously successful loads.  The strict
  4713 update policy of former update_thy is now already performed by
  4714 use_thy, so the former has been removed; use_thys updates several
  4715 theories simultaneously, just as 'imports' within a theory header
  4716 specification, but without merging the results.  Potential
  4717 INCOMPATIBILITY: may need to refine theory headers and commands
  4718 ROOT.ML which depend on load order.
  4719 
  4720 * Theory loader: optional support for content-based file
  4721 identification, instead of the traditional scheme of full physical
  4722 path plus date stamp; configured by the ISABELLE_FILE_IDENT setting
  4723 (cf. the system manual).  The new scheme allows to work with
  4724 non-finished theories in persistent session images, such that source
  4725 files may be moved later on without requiring reloads.
  4726 
  4727 * Theory loader: old-style ML proof scripts being *attached* to a thy
  4728 file (with the same base name as the theory) are considered a legacy
  4729 feature, which will disappear eventually. Even now, the theory loader
  4730 no longer maintains dependencies on such files.
  4731 
  4732 * Syntax: the scope for resolving ambiguities via type-inference is
  4733 now limited to individual terms, instead of whole simultaneous
  4734 specifications as before. This greatly reduces the complexity of the
  4735 syntax module and improves flexibility by separating parsing and
  4736 type-checking. INCOMPATIBILITY: additional type-constraints (explicit
  4737 'fixes' etc.) are required in rare situations.
  4738 
  4739 * Syntax: constants introduced by new-style packages ('definition',
  4740 'abbreviation' etc.) are passed through the syntax module in
  4741 ``authentic mode''. This means that associated mixfix annotations
  4742 really stick to such constants, independently of potential name space
  4743 ambiguities introduced later on. INCOMPATIBILITY: constants in parse
  4744 trees are represented slightly differently, may need to adapt syntax
  4745 translations accordingly. Use CONST marker in 'translations' and
  4746 @{const_syntax} antiquotation in 'parse_translation' etc.
  4747 
  4748 * Legacy goal package: reduced interface to the bare minimum required
  4749 to keep existing proof scripts running.  Most other user-level
  4750 functions are now part of the OldGoals structure, which is *not* open
  4751 by default (consider isatool expandshort before open OldGoals).
  4752 Removed top_sg, prin, printyp, pprint_term/typ altogether, because
  4753 these tend to cause confusion about the actual goal (!) context being
  4754 used here, which is not necessarily the same as the_context().
  4755 
  4756 * Command 'find_theorems': supports "*" wild-card in "name:"
  4757 criterion; "with_dups" option.  Certain ProofGeneral versions might
  4758 support a specific search form (see ProofGeneral/CHANGES).
  4759 
  4760 * The ``prems limit'' option (cf. ProofContext.prems_limit) is now -1
  4761 by default, which means that "prems" (and also "fixed variables") are
  4762 suppressed from proof state output.  Note that the ProofGeneral
  4763 settings mechanism allows to change and save options persistently, but
  4764 older versions of Isabelle will fail to start up if a negative prems
  4765 limit is imposed.
  4766 
  4767 * Local theory targets may be specified by non-nested blocks of
  4768 ``context/locale/class ... begin'' followed by ``end''.  The body may
  4769 contain definitions, theorems etc., including any derived mechanism
  4770 that has been implemented on top of these primitives.  This concept
  4771 generalizes the existing ``theorem (in ...)'' towards more versatility
  4772 and scalability.
  4773 
  4774 * Proof General interface: proper undo of final 'end' command;
  4775 discontinued Isabelle/classic mode (ML proof scripts).
  4776 
  4777 
  4778 *** Document preparation ***
  4779 
  4780 * Added antiquotation @{theory name} which prints the given name,
  4781 after checking that it refers to a valid ancestor theory in the
  4782 current context.
  4783 
  4784 * Added antiquotations @{ML_type text} and @{ML_struct text} which
  4785 check the given source text as ML type/structure, printing verbatim.
  4786 
  4787 * Added antiquotation @{abbrev "c args"} which prints the abbreviation
  4788 "c args == rhs" given in the current context.  (Any number of
  4789 arguments may be given on the LHS.)
  4790 
  4791 
  4792 *** Pure ***
  4793 
  4794 * The 'class' package offers a combination of axclass and locale to
  4795 achieve Haskell-like type classes in Isabelle.  Definitions and
  4796 theorems within a class context produce both relative results (with
  4797 implicit parameters according to the locale context), and polymorphic
  4798 constants with qualified polymorphism (according to the class
  4799 context).  Within the body context of a 'class' target, a separate
  4800 syntax layer ("user space type system") takes care of converting
  4801 between global polymorphic consts and internal locale representation.
  4802 See src/HOL/ex/Classpackage.thy for examples (as well as main HOL).
  4803 "isatool doc classes" provides a tutorial.
  4804 
  4805 * Generic code generator framework allows to generate executable
  4806 code for ML and Haskell (including Isabelle classes).  A short usage
  4807 sketch:
  4808 
  4809     internal compilation:
  4810         export_code <list of constants (term syntax)> in SML
  4811     writing SML code to a file:
  4812         export_code <list of constants (term syntax)> in SML <filename>
  4813     writing OCaml code to a file:
  4814         export_code <list of constants (term syntax)> in OCaml <filename>
  4815     writing Haskell code to a bunch of files:
  4816         export_code <list of constants (term syntax)> in Haskell <filename>
  4817 
  4818     evaluating closed propositions to True/False using code generation:
  4819         method ``eval''
  4820 
  4821 Reasonable default setup of framework in HOL.
  4822 
  4823 Theorem attributs for selecting and transforming function equations theorems:
  4824 
  4825     [code fun]:        select a theorem as function equation for a specific constant
  4826     [code fun del]:    deselect a theorem as function equation for a specific constant
  4827     [code inline]:     select an equation theorem for unfolding (inlining) in place
  4828     [code inline del]: deselect an equation theorem for unfolding (inlining) in place
  4829 
  4830 User-defined serializations (target in {SML, OCaml, Haskell}):
  4831 
  4832     code_const <and-list of constants (term syntax)>
  4833       {(target) <and-list of const target syntax>}+
  4834 
  4835     code_type <and-list of type constructors>
  4836       {(target) <and-list of type target syntax>}+
  4837 
  4838     code_instance <and-list of instances>
  4839       {(target)}+
  4840         where instance ::= <type constructor> :: <class>
  4841 
  4842     code_class <and_list of classes>
  4843       {(target) <and-list of class target syntax>}+
  4844         where class target syntax ::= <class name> {where {<classop> == <target syntax>}+}?
  4845 
  4846 code_instance and code_class only are effective to target Haskell.
  4847 
  4848 For example usage see src/HOL/ex/Codegenerator.thy and
  4849 src/HOL/ex/Codegenerator_Pretty.thy.  A separate tutorial on code
  4850 generation from Isabelle/HOL theories is available via "isatool doc
  4851 codegen".
  4852 
  4853 * Code generator: consts in 'consts_code' Isar commands are now
  4854 referred to by usual term syntax (including optional type
  4855 annotations).
  4856 
  4857 * Command 'no_translations' removes translation rules from theory
  4858 syntax.
  4859 
  4860 * Overloaded definitions are now actually checked for acyclic
  4861 dependencies.  The overloading scheme is slightly more general than
  4862 that of Haskell98, although Isabelle does not demand an exact
  4863 correspondence to type class and instance declarations.
  4864 INCOMPATIBILITY, use ``defs (unchecked overloaded)'' to admit more
  4865 exotic versions of overloading -- at the discretion of the user!
  4866 
  4867 Polymorphic constants are represented via type arguments, i.e. the
  4868 instantiation that matches an instance against the most general
  4869 declaration given in the signature.  For example, with the declaration
  4870 c :: 'a => 'a => 'a, an instance c :: nat => nat => nat is represented
  4871 as c(nat).  Overloading is essentially simultaneous structural
  4872 recursion over such type arguments.  Incomplete specification patterns
  4873 impose global constraints on all occurrences, e.g. c('a * 'a) on the
  4874 LHS means that more general c('a * 'b) will be disallowed on any RHS.
  4875 Command 'print_theory' outputs the normalized system of recursive
  4876 equations, see section "definitions".
  4877 
  4878 * Configuration options are maintained within the theory or proof
  4879 context (with name and type bool/int/string), providing a very simple
  4880 interface to a poor-man's version of general context data.  Tools may
  4881 declare options in ML (e.g. using Attrib.config_int) and then refer to
  4882 these values using Config.get etc.  Users may change options via an
  4883 associated attribute of the same name.  This form of context
  4884 declaration works particularly well with commands 'declare' or
  4885 'using', for example ``declare [[foo = 42]]''.  Thus it has become
  4886 very easy to avoid global references, which would not observe Isar
  4887 toplevel undo/redo and fail to work with multithreading.
  4888 
  4889 Various global ML references of Pure and HOL have been turned into
  4890 configuration options:
  4891 
  4892   Unify.search_bound		unify_search_bound
  4893   Unify.trace_bound		unify_trace_bound
  4894   Unify.trace_simp		unify_trace_simp
  4895   Unify.trace_types		unify_trace_types
  4896   Simplifier.simp_depth_limit	simp_depth_limit
  4897   Blast.depth_limit		blast_depth_limit
  4898   DatatypeProp.dtK		datatype_distinctness_limit
  4899   fast_arith_neq_limit  	fast_arith_neq_limit
  4900   fast_arith_split_limit	fast_arith_split_limit
  4901 
  4902 * Named collections of theorems may be easily installed as context
  4903 data using the functor NamedThmsFun (see also
  4904 src/Pure/Tools/named_thms.ML).  The user may add or delete facts via
  4905 attributes; there is also a toplevel print command.  This facility is
  4906 just a common case of general context data, which is the preferred way
  4907 for anything more complex than just a list of facts in canonical
  4908 order.
  4909 
  4910 * Isar: command 'declaration' augments a local theory by generic
  4911 declaration functions written in ML.  This enables arbitrary content
  4912 being added to the context, depending on a morphism that tells the
  4913 difference of the original declaration context wrt. the application
  4914 context encountered later on.
  4915 
  4916 * Isar: proper interfaces for simplification procedures.  Command
  4917 'simproc_setup' declares named simprocs (with match patterns, and body
  4918 text in ML).  Attribute "simproc" adds/deletes simprocs in the current
  4919 context.  ML antiquotation @{simproc name} retrieves named simprocs.
  4920 
  4921 * Isar: an extra pair of brackets around attribute declarations
  4922 abbreviates a theorem reference involving an internal dummy fact,
  4923 which will be ignored later --- only the effect of the attribute on
  4924 the background context will persist.  This form of in-place
  4925 declarations is particularly useful with commands like 'declare' and
  4926 'using', for example ``have A using [[simproc a]] by simp''.
  4927 
  4928 * Isar: method "assumption" (and implicit closing of subproofs) now
  4929 takes simple non-atomic goal assumptions into account: after applying
  4930 an assumption as a rule the resulting subgoals are solved by atomic
  4931 assumption steps.  This is particularly useful to finish 'obtain'
  4932 goals, such as "!!x. (!!x. P x ==> thesis) ==> P x ==> thesis",
  4933 without referring to the original premise "!!x. P x ==> thesis" in the
  4934 Isar proof context.  POTENTIAL INCOMPATIBILITY: method "assumption" is
  4935 more permissive.
  4936 
  4937 * Isar: implicit use of prems from the Isar proof context is
  4938 considered a legacy feature.  Common applications like ``have A .''
  4939 may be replaced by ``have A by fact'' or ``note `A`''.  In general,
  4940 referencing facts explicitly here improves readability and
  4941 maintainability of proof texts.
  4942 
  4943 * Isar: improper proof element 'guess' is like 'obtain', but derives
  4944 the obtained context from the course of reasoning!  For example:
  4945 
  4946   assume "EX x y. A x & B y"   -- "any previous fact"
  4947   then guess x and y by clarify
  4948 
  4949 This technique is potentially adventurous, depending on the facts and
  4950 proof tools being involved here.
  4951 
  4952 * Isar: known facts from the proof context may be specified as literal
  4953 propositions, using ASCII back-quote syntax.  This works wherever
  4954 named facts used to be allowed so far, in proof commands, proof
  4955 methods, attributes etc.  Literal facts are retrieved from the context
  4956 according to unification of type and term parameters.  For example,
  4957 provided that "A" and "A ==> B" and "!!x. P x ==> Q x" are known
  4958 theorems in the current context, then these are valid literal facts:
  4959 `A` and `A ==> B` and `!!x. P x ==> Q x" as well as `P a ==> Q a` etc.
  4960 
  4961 There is also a proof method "fact" which does the same composition
  4962 for explicit goal states, e.g. the following proof texts coincide with
  4963 certain special cases of literal facts:
  4964 
  4965   have "A" by fact                 ==  note `A`
  4966   have "A ==> B" by fact           ==  note `A ==> B`
  4967   have "!!x. P x ==> Q x" by fact  ==  note `!!x. P x ==> Q x`
  4968   have "P a ==> Q a" by fact       ==  note `P a ==> Q a`
  4969 
  4970 * Isar: ":" (colon) is no longer a symbolic identifier character in
  4971 outer syntax.  Thus symbolic identifiers may be used without
  4972 additional white space in declarations like this: ``assume *: A''.
  4973 
  4974 * Isar: 'print_facts' prints all local facts of the current context,
  4975 both named and unnamed ones.
  4976 
  4977 * Isar: 'def' now admits simultaneous definitions, e.g.:
  4978 
  4979   def x == "t" and y == "u"
  4980 
  4981 * Isar: added command 'unfolding', which is structurally similar to
  4982 'using', but affects both the goal state and facts by unfolding given
  4983 rewrite rules.  Thus many occurrences of the 'unfold' method or
  4984 'unfolded' attribute may be replaced by first-class proof text.
  4985 
  4986 * Isar: methods 'unfold' / 'fold', attributes 'unfolded' / 'folded',
  4987 and command 'unfolding' now all support object-level equalities
  4988 (potentially conditional).  The underlying notion of rewrite rule is
  4989 analogous to the 'rule_format' attribute, but *not* that of the
  4990 Simplifier (which is usually more generous).
  4991 
  4992 * Isar: the new attribute [rotated n] (default n = 1) rotates the
  4993 premises of a theorem by n. Useful in conjunction with drule.
  4994 
  4995 * Isar: the goal restriction operator [N] (default N = 1) evaluates a
  4996 method expression within a sandbox consisting of the first N
  4997 sub-goals, which need to exist.  For example, ``simp_all [3]''
  4998 simplifies the first three sub-goals, while (rule foo, simp_all)[]
  4999 simplifies all new goals that emerge from applying rule foo to the
  5000 originally first one.
  5001 
  5002 * Isar: schematic goals are no longer restricted to higher-order
  5003 patterns; e.g. ``lemma "?P(?x)" by (rule TrueI)'' now works as
  5004 expected.
  5005 
  5006 * Isar: the conclusion of a long theorem statement is now either
  5007 'shows' (a simultaneous conjunction, as before), or 'obtains'
  5008 (essentially a disjunction of cases with local parameters and
  5009 assumptions).  The latter allows to express general elimination rules
  5010 adequately; in this notation common elimination rules look like this:
  5011 
  5012   lemma exE:    -- "EX x. P x ==> (!!x. P x ==> thesis) ==> thesis"
  5013     assumes "EX x. P x"
  5014     obtains x where "P x"
  5015 
  5016   lemma conjE:  -- "A & B ==> (A ==> B ==> thesis) ==> thesis"
  5017     assumes "A & B"
  5018     obtains A and B
  5019 
  5020   lemma disjE:  -- "A | B ==> (A ==> thesis) ==> (B ==> thesis) ==> thesis"
  5021     assumes "A | B"
  5022     obtains
  5023       A
  5024     | B
  5025 
  5026 The subsequent classical rules even refer to the formal "thesis"
  5027 explicitly:
  5028 
  5029   lemma classical:     -- "(~ thesis ==> thesis) ==> thesis"
  5030     obtains "~ thesis"
  5031 
  5032   lemma Peirce's_Law:  -- "((thesis ==> something) ==> thesis) ==> thesis"
  5033     obtains "thesis ==> something"
  5034 
  5035 The actual proof of an 'obtains' statement is analogous to that of the
  5036 Isar proof element 'obtain', only that there may be several cases.
  5037 Optional case names may be specified in parentheses; these will be
  5038 available both in the present proof and as annotations in the
  5039 resulting rule, for later use with the 'cases' method (cf. attribute
  5040 case_names).
  5041 
  5042 * Isar: the assumptions of a long theorem statement are available as
  5043 "assms" fact in the proof context.  This is more appropriate than the
  5044 (historical) "prems", which refers to all assumptions of the current
  5045 context, including those from the target locale, proof body etc.
  5046 
  5047 * Isar: 'print_statement' prints theorems from the current theory or
  5048 proof context in long statement form, according to the syntax of a
  5049 top-level lemma.
  5050 
  5051 * Isar: 'obtain' takes an optional case name for the local context
  5052 introduction rule (default "that").
  5053 
  5054 * Isar: removed obsolete 'concl is' patterns.  INCOMPATIBILITY, use
  5055 explicit (is "_ ==> ?foo") in the rare cases where this still happens
  5056 to occur.
  5057 
  5058 * Pure: syntax "CONST name" produces a fully internalized constant
  5059 according to the current context.  This is particularly useful for
  5060 syntax translations that should refer to internal constant
  5061 representations independently of name spaces.
  5062 
  5063 * Pure: syntax constant for foo (binder "FOO ") is called "foo_binder"
  5064 instead of "FOO ". This allows multiple binder declarations to coexist
  5065 in the same context.  INCOMPATIBILITY.
  5066 
  5067 * Isar/locales: 'notation' provides a robust interface to the 'syntax'
  5068 primitive that also works in a locale context (both for constants and
  5069 fixed variables). Type declaration and internal syntactic representation
  5070 of given constants retrieved from the context. Likewise, the
  5071 'no_notation' command allows to remove given syntax annotations from the
  5072 current context.
  5073 
  5074 * Isar/locales: new derived specification elements 'axiomatization',
  5075 'definition', 'abbreviation', which support type-inference, admit
  5076 object-level specifications (equality, equivalence).  See also the
  5077 isar-ref manual.  Examples:
  5078 
  5079   axiomatization
  5080     eq  (infix "===" 50) where
  5081     eq_refl: "x === x" and eq_subst: "x === y ==> P x ==> P y"
  5082 
  5083   definition "f x y = x + y + 1"
  5084   definition g where "g x = f x x"
  5085 
  5086   abbreviation
  5087     neq  (infix "=!=" 50) where
  5088     "x =!= y == ~ (x === y)"
  5089 
  5090 These specifications may be also used in a locale context.  Then the
  5091 constants being introduced depend on certain fixed parameters, and the
  5092 constant name is qualified by the locale base name.  An internal
  5093 abbreviation takes care for convenient input and output, making the
  5094 parameters implicit and using the original short name.  See also
  5095 src/HOL/ex/Abstract_NAT.thy for an example of deriving polymorphic
  5096 entities from a monomorphic theory.
  5097 
  5098 Presently, abbreviations are only available 'in' a target locale, but
  5099 not inherited by general import expressions.  Also note that
  5100 'abbreviation' may be used as a type-safe replacement for 'syntax' +
  5101 'translations' in common applications.  The "no_abbrevs" print mode
  5102 prevents folding of abbreviations in term output.
  5103 
  5104 Concrete syntax is attached to specified constants in internal form,
  5105 independently of name spaces.  The parse tree representation is
  5106 slightly different -- use 'notation' instead of raw 'syntax', and
  5107 'translations' with explicit "CONST" markup to accommodate this.
  5108 
  5109 * Pure/Isar: unified syntax for new-style specification mechanisms
  5110 (e.g.  'definition', 'abbreviation', or 'inductive' in HOL) admits
  5111 full type inference and dummy patterns ("_").  For example:
  5112 
  5113   definition "K x _ = x"
  5114 
  5115   inductive conj for A B
  5116   where "A ==> B ==> conj A B"
  5117 
  5118 * Pure: command 'print_abbrevs' prints all constant abbreviations of
  5119 the current context.  Print mode "no_abbrevs" prevents inversion of
  5120 abbreviations on output.
  5121 
  5122 * Isar/locales: improved parameter handling: use of locales "var" and
  5123 "struct" no longer necessary; - parameter renamings are no longer
  5124 required to be injective.  For example, this allows to define
  5125 endomorphisms as locale endom = homom mult mult h.
  5126 
  5127 * Isar/locales: changed the way locales with predicates are defined.
  5128 Instead of accumulating the specification, the imported expression is
  5129 now an interpretation.  INCOMPATIBILITY: different normal form of
  5130 locale expressions.  In particular, in interpretations of locales with
  5131 predicates, goals repesenting already interpreted fragments are not
  5132 removed automatically.  Use methods `intro_locales' and
  5133 `unfold_locales'; see below.
  5134 
  5135 * Isar/locales: new methods `intro_locales' and `unfold_locales'
  5136 provide backward reasoning on locales predicates.  The methods are
  5137 aware of interpretations and discharge corresponding goals.
  5138 `intro_locales' is less aggressive then `unfold_locales' and does not
  5139 unfold predicates to assumptions.
  5140 
  5141 * Isar/locales: the order in which locale fragments are accumulated
  5142 has changed.  This enables to override declarations from fragments due
  5143 to interpretations -- for example, unwanted simp rules.
  5144 
  5145 * Isar/locales: interpretation in theories and proof contexts has been
  5146 extended.  One may now specify (and prove) equations, which are
  5147 unfolded in interpreted theorems.  This is useful for replacing
  5148 defined concepts (constants depending on locale parameters) by
  5149 concepts already existing in the target context.  Example:
  5150 
  5151   interpretation partial_order ["op <= :: [int, int] => bool"]
  5152     where "partial_order.less (op <=) (x::int) y = (x < y)"
  5153 
  5154 Typically, the constant `partial_order.less' is created by a
  5155 definition specification element in the context of locale
  5156 partial_order.
  5157 
  5158 * Method "induct": improved internal context management to support
  5159 local fixes and defines on-the-fly. Thus explicit meta-level
  5160 connectives !!  and ==> are rarely required anymore in inductive goals
  5161 (using object-logic connectives for this purpose has been long
  5162 obsolete anyway). Common proof patterns are explained in
  5163 src/HOL/Induct/Common_Patterns.thy, see also
  5164 src/HOL/Isar_examples/Puzzle.thy and src/HOL/Lambda for realistic
  5165 examples.
  5166 
  5167 * Method "induct": improved handling of simultaneous goals. Instead of
  5168 introducing object-level conjunction, the statement is now split into
  5169 several conclusions, while the corresponding symbolic cases are nested
  5170 accordingly. INCOMPATIBILITY, proofs need to be structured explicitly,
  5171 see src/HOL/Induct/Common_Patterns.thy, for example.
  5172 
  5173 * Method "induct": mutual induction rules are now specified as a list
  5174 of rule sharing the same induction cases. HOL packages usually provide
  5175 foo_bar.inducts for mutually defined items foo and bar (e.g. inductive
  5176 predicates/sets or datatypes). INCOMPATIBILITY, users need to specify
  5177 mutual induction rules differently, i.e. like this:
  5178 
  5179   (induct rule: foo_bar.inducts)
  5180   (induct set: foo bar)
  5181   (induct pred: foo bar)
  5182   (induct type: foo bar)
  5183 
  5184 The ML function ProjectRule.projections turns old-style rules into the
  5185 new format.
  5186 
  5187 * Method "coinduct": dual of induction, see
  5188 src/HOL/Library/Coinductive_List.thy for various examples.
  5189 
  5190 * Method "cases", "induct", "coinduct": the ``(open)'' option is
  5191 considered a legacy feature.
  5192 
  5193 * Attribute "symmetric" produces result with standardized schematic
  5194 variables (index 0).  Potential INCOMPATIBILITY.
  5195 
  5196 * Simplifier: by default the simplifier trace only shows top level
  5197 rewrites now. That is, trace_simp_depth_limit is set to 1 by
  5198 default. Thus there is less danger of being flooded by the trace. The
  5199 trace indicates where parts have been suppressed.
  5200   
  5201 * Provers/classical: removed obsolete classical version of elim_format
  5202 attribute; classical elim/dest rules are now treated uniformly when
  5203 manipulating the claset.
  5204 
  5205 * Provers/classical: stricter checks to ensure that supplied intro,
  5206 dest and elim rules are well-formed; dest and elim rules must have at
  5207 least one premise.
  5208 
  5209 * Provers/classical: attributes dest/elim/intro take an optional
  5210 weight argument for the rule (just as the Pure versions).  Weights are
  5211 ignored by automated tools, but determine the search order of single
  5212 rule steps.
  5213 
  5214 * Syntax: input syntax now supports dummy variable binding "%_. b",
  5215 where the body does not mention the bound variable.  Note that dummy
  5216 patterns implicitly depend on their context of bounds, which makes
  5217 "{_. _}" match any set comprehension as expected.  Potential
  5218 INCOMPATIBILITY -- parse translations need to cope with syntactic
  5219 constant "_idtdummy" in the binding position.
  5220 
  5221 * Syntax: removed obsolete syntactic constant "_K" and its associated
  5222 parse translation.  INCOMPATIBILITY -- use dummy abstraction instead,
  5223 for example "A -> B" => "Pi A (%_. B)".
  5224 
  5225 * Pure: 'class_deps' command visualizes the subclass relation, using
  5226 the graph browser tool.
  5227 
  5228 * Pure: 'print_theory' now suppresses certain internal declarations by
  5229 default; use '!' option for full details.
  5230 
  5231 
  5232 *** HOL ***
  5233 
  5234 * Method "metis" proves goals by applying the Metis general-purpose
  5235 resolution prover (see also http://gilith.com/software/metis/).
  5236 Examples are in the directory MetisExamples.  WARNING: the
  5237 Isabelle/HOL-Metis integration does not yet work properly with
  5238 multi-threading.
  5239   
  5240 * Command 'sledgehammer' invokes external automatic theorem provers as
  5241 background processes.  It generates calls to the "metis" method if
  5242 successful. These can be pasted into the proof.  Users do not have to
  5243 wait for the automatic provers to return.  WARNING: does not really
  5244 work with multi-threading.
  5245 
  5246 * New "auto_quickcheck" feature tests outermost goal statements for
  5247 potential counter-examples.  Controlled by ML references
  5248 auto_quickcheck (default true) and auto_quickcheck_time_limit (default
  5249 5000 milliseconds).  Fails silently if statements is outside of
  5250 executable fragment, or any other codgenerator problem occurs.
  5251 
  5252 * New constant "undefined" with axiom "undefined x = undefined".
  5253 
  5254 * Added class "HOL.eq", allowing for code generation with polymorphic
  5255 equality.
  5256 
  5257 * Some renaming of class constants due to canonical name prefixing in
  5258 the new 'class' package:
  5259 
  5260     HOL.abs ~> HOL.abs_class.abs
  5261     HOL.divide ~> HOL.divide_class.divide
  5262     0 ~> HOL.zero_class.zero
  5263     1 ~> HOL.one_class.one
  5264     op + ~> HOL.plus_class.plus
  5265     op - ~> HOL.minus_class.minus
  5266     uminus ~> HOL.minus_class.uminus
  5267     op * ~> HOL.times_class.times
  5268     op < ~> HOL.ord_class.less
  5269     op <= > HOL.ord_class.less_eq
  5270     Nat.power ~> Power.power_class.power
  5271     Nat.size ~> Nat.size_class.size
  5272     Numeral.number_of ~> Numeral.number_class.number_of
  5273     FixedPoint.Inf ~> Lattices.complete_lattice_class.Inf
  5274     FixedPoint.Sup ~> Lattices.complete_lattice_class.Sup
  5275     Orderings.min ~> Orderings.ord_class.min
  5276     Orderings.max ~> Orderings.ord_class.max
  5277     Divides.op div ~> Divides.div_class.div
  5278     Divides.op mod ~> Divides.div_class.mod
  5279     Divides.op dvd ~> Divides.div_class.dvd
  5280 
  5281 INCOMPATIBILITY.  Adaptions may be required in the following cases:
  5282 
  5283 a) User-defined constants using any of the names "plus", "minus",
  5284 "times", "less" or "less_eq". The standard syntax translations for
  5285 "+", "-" and "*" may go wrong.  INCOMPATIBILITY: use more specific
  5286 names.
  5287 
  5288 b) Variables named "plus", "minus", "times", "less", "less_eq"
  5289 INCOMPATIBILITY: use more specific names.
  5290 
  5291 c) Permutative equations (e.g. "a + b = b + a")
  5292 Since the change of names also changes the order of terms, permutative
  5293 rewrite rules may get applied in a different order. Experience shows
  5294 that this is rarely the case (only two adaptions in the whole Isabelle
  5295 distribution).  INCOMPATIBILITY: rewrite proofs
  5296 
  5297 d) ML code directly refering to constant names
  5298 This in general only affects hand-written proof tactics, simprocs and
  5299 so on.  INCOMPATIBILITY: grep your sourcecode and replace names.
  5300 Consider using @{const_name} antiquotation.
  5301 
  5302 * New class "default" with associated constant "default".
  5303 
  5304 * Function "sgn" is now overloaded and available on int, real, complex
  5305 (and other numeric types), using class "sgn".  Two possible defs of
  5306 sgn are given as equational assumptions in the classes sgn_if and
  5307 sgn_div_norm; ordered_idom now also inherits from sgn_if.
  5308 INCOMPATIBILITY.
  5309 
  5310 * Locale "partial_order" now unified with class "order" (cf. theory
  5311 Orderings), added parameter "less".  INCOMPATIBILITY.
  5312 
  5313 * Renamings in classes "order" and "linorder": facts "refl", "trans" and
  5314 "cases" to "order_refl", "order_trans" and "linorder_cases", to avoid
  5315 clashes with HOL "refl" and "trans".  INCOMPATIBILITY.
  5316 
  5317 * Classes "order" and "linorder": potential INCOMPATIBILITY due to
  5318 changed order of proof goals in instance proofs.
  5319 
  5320 * The transitivity reasoner for partial and linear orders is set up
  5321 for classes "order" and "linorder".  Instances of the reasoner are available
  5322 in all contexts importing or interpreting the corresponding locales.
  5323 Method "order" invokes the reasoner separately; the reasoner
  5324 is also integrated with the Simplifier as a solver.  Diagnostic
  5325 command 'print_orders' shows the available instances of the reasoner
  5326 in the current context.
  5327 
  5328 * Localized monotonicity predicate in theory "Orderings"; integrated
  5329 lemmas max_of_mono and min_of_mono with this predicate.
  5330 INCOMPATIBILITY.
  5331 
  5332 * Formulation of theorem "dense" changed slightly due to integration
  5333 with new class dense_linear_order.
  5334 
  5335 * Uniform lattice theory development in HOL.
  5336 
  5337     constants "meet" and "join" now named "inf" and "sup"
  5338     constant "Meet" now named "Inf"
  5339 
  5340     classes "meet_semilorder" and "join_semilorder" now named
  5341       "lower_semilattice" and "upper_semilattice"
  5342     class "lorder" now named "lattice"
  5343     class "comp_lat" now named "complete_lattice"
  5344 
  5345     Instantiation of lattice classes allows explicit definitions
  5346     for "inf" and "sup" operations (or "Inf" and "Sup" for complete lattices).
  5347 
  5348   INCOMPATIBILITY.  Theorem renames:
  5349 
  5350     meet_left_le            ~> inf_le1
  5351     meet_right_le           ~> inf_le2
  5352     join_left_le            ~> sup_ge1
  5353     join_right_le           ~> sup_ge2
  5354     meet_join_le            ~> inf_sup_ord
  5355     le_meetI                ~> le_infI
  5356     join_leI                ~> le_supI
  5357     le_meet                 ~> le_inf_iff
  5358     le_join                 ~> ge_sup_conv
  5359     meet_idempotent         ~> inf_idem
  5360     join_idempotent         ~> sup_idem
  5361     meet_comm               ~> inf_commute
  5362     join_comm               ~> sup_commute
  5363     meet_leI1               ~> le_infI1
  5364     meet_leI2               ~> le_infI2
  5365     le_joinI1               ~> le_supI1
  5366     le_joinI2               ~> le_supI2
  5367     meet_assoc              ~> inf_assoc
  5368     join_assoc              ~> sup_assoc
  5369     meet_left_comm          ~> inf_left_commute
  5370     meet_left_idempotent    ~> inf_left_idem
  5371     join_left_comm          ~> sup_left_commute
  5372     join_left_idempotent    ~> sup_left_idem
  5373     meet_aci                ~> inf_aci
  5374     join_aci                ~> sup_aci
  5375     le_def_meet             ~> le_iff_inf
  5376     le_def_join             ~> le_iff_sup
  5377     join_absorp2            ~> sup_absorb2
  5378     join_absorp1            ~> sup_absorb1
  5379     meet_absorp1            ~> inf_absorb1
  5380     meet_absorp2            ~> inf_absorb2
  5381     meet_join_absorp        ~> inf_sup_absorb
  5382     join_meet_absorp        ~> sup_inf_absorb
  5383     distrib_join_le         ~> distrib_sup_le
  5384     distrib_meet_le         ~> distrib_inf_le
  5385 
  5386     add_meet_distrib_left   ~> add_inf_distrib_left
  5387     add_join_distrib_left   ~> add_sup_distrib_left
  5388     is_join_neg_meet        ~> is_join_neg_inf
  5389     is_meet_neg_join        ~> is_meet_neg_sup
  5390     add_meet_distrib_right  ~> add_inf_distrib_right
  5391     add_join_distrib_right  ~> add_sup_distrib_right
  5392     add_meet_join_distribs  ~> add_sup_inf_distribs
  5393     join_eq_neg_meet        ~> sup_eq_neg_inf
  5394     meet_eq_neg_join        ~> inf_eq_neg_sup
  5395     add_eq_meet_join        ~> add_eq_inf_sup
  5396     meet_0_imp_0            ~> inf_0_imp_0
  5397     join_0_imp_0            ~> sup_0_imp_0
  5398     meet_0_eq_0             ~> inf_0_eq_0
  5399     join_0_eq_0             ~> sup_0_eq_0
  5400     neg_meet_eq_join        ~> neg_inf_eq_sup
  5401     neg_join_eq_meet        ~> neg_sup_eq_inf
  5402     join_eq_if              ~> sup_eq_if
  5403 
  5404     mono_meet               ~> mono_inf
  5405     mono_join               ~> mono_sup
  5406     meet_bool_eq            ~> inf_bool_eq
  5407     join_bool_eq            ~> sup_bool_eq
  5408     meet_fun_eq             ~> inf_fun_eq
  5409     join_fun_eq             ~> sup_fun_eq
  5410     meet_set_eq             ~> inf_set_eq
  5411     join_set_eq             ~> sup_set_eq
  5412     meet1_iff               ~> inf1_iff
  5413     meet2_iff               ~> inf2_iff
  5414     meet1I                  ~> inf1I
  5415     meet2I                  ~> inf2I
  5416     meet1D1                 ~> inf1D1
  5417     meet2D1                 ~> inf2D1
  5418     meet1D2                 ~> inf1D2
  5419     meet2D2                 ~> inf2D2
  5420     meet1E                  ~> inf1E
  5421     meet2E                  ~> inf2E
  5422     join1_iff               ~> sup1_iff
  5423     join2_iff               ~> sup2_iff
  5424     join1I1                 ~> sup1I1
  5425     join2I1                 ~> sup2I1
  5426     join1I1                 ~> sup1I1
  5427     join2I2                 ~> sup1I2
  5428     join1CI                 ~> sup1CI
  5429     join2CI                 ~> sup2CI
  5430     join1E                  ~> sup1E
  5431     join2E                  ~> sup2E
  5432 
  5433     is_meet_Meet            ~> is_meet_Inf
  5434     Meet_bool_def           ~> Inf_bool_def
  5435     Meet_fun_def            ~> Inf_fun_def
  5436     Meet_greatest           ~> Inf_greatest
  5437     Meet_lower              ~> Inf_lower
  5438     Meet_set_def            ~> Inf_set_def
  5439 
  5440     Sup_def                 ~> Sup_Inf
  5441     Sup_bool_eq             ~> Sup_bool_def
  5442     Sup_fun_eq              ~> Sup_fun_def
  5443     Sup_set_eq              ~> Sup_set_def
  5444 
  5445     listsp_meetI            ~> listsp_infI
  5446     listsp_meet_eq          ~> listsp_inf_eq
  5447 
  5448     meet_min                ~> inf_min
  5449     join_max                ~> sup_max
  5450 
  5451 * Added syntactic class "size"; overloaded constant "size" now has
  5452 type "'a::size ==> bool"
  5453 
  5454 * Internal reorganisation of `size' of datatypes: size theorems
  5455 "foo.size" are no longer subsumed by "foo.simps" (but are still
  5456 simplification rules by default!); theorems "prod.size" now named
  5457 "*.size".
  5458 
  5459 * Class "div" now inherits from class "times" rather than "type".
  5460 INCOMPATIBILITY.
  5461 
  5462 * HOL/Finite_Set: "name-space" locales Lattice, Distrib_lattice,
  5463 Linorder etc.  have disappeared; operations defined in terms of
  5464 fold_set now are named Inf_fin, Sup_fin.  INCOMPATIBILITY.
  5465 
  5466 * HOL/Nat: neq0_conv no longer declared as iff.  INCOMPATIBILITY.
  5467 
  5468 * HOL-Word: New extensive library and type for generic, fixed size
  5469 machine words, with arithmetic, bit-wise, shifting and rotating
  5470 operations, reflection into int, nat, and bool lists, automation for
  5471 linear arithmetic (by automatic reflection into nat or int), including
  5472 lemmas on overflow and monotonicity.  Instantiated to all appropriate
  5473 arithmetic type classes, supporting automatic simplification of
  5474 numerals on all operations.
  5475 
  5476 * Library/Boolean_Algebra: locales for abstract boolean algebras.
  5477 
  5478 * Library/Numeral_Type: numbers as types, e.g. TYPE(32).
  5479 
  5480 * Code generator library theories:
  5481   - Code_Integer represents HOL integers by big integer literals in target
  5482     languages.
  5483   - Code_Char represents HOL characters by character literals in target
  5484     languages.
  5485   - Code_Char_chr like Code_Char, but also offers treatment of character
  5486     codes; includes Code_Integer.
  5487   - Executable_Set allows to generate code for finite sets using lists.
  5488   - Executable_Rat implements rational numbers as triples (sign, enumerator,
  5489     denominator).
  5490   - Executable_Real implements a subset of real numbers, namly those
  5491     representable by rational numbers.
  5492   - Efficient_Nat implements natural numbers by integers, which in general will
  5493     result in higher efficency; pattern matching with 0/Suc is eliminated;
  5494     includes Code_Integer.
  5495   - Code_Index provides an additional datatype index which is mapped to
  5496     target-language built-in integers.
  5497   - Code_Message provides an additional datatype message_string which is isomorphic to
  5498     strings; messages are mapped to target-language strings.
  5499 
  5500 * New package for inductive predicates
  5501 
  5502   An n-ary predicate p with m parameters z_1, ..., z_m can now be defined via
  5503 
  5504     inductive
  5505       p :: "U_1 => ... => U_m => T_1 => ... => T_n => bool"
  5506       for z_1 :: U_1 and ... and z_n :: U_m
  5507     where
  5508       rule_1: "... ==> p z_1 ... z_m t_1_1 ... t_1_n"
  5509     | ...
  5510 
  5511   with full support for type-inference, rather than
  5512 
  5513     consts s :: "U_1 => ... => U_m => (T_1 * ... * T_n) set"
  5514 
  5515     abbreviation p :: "U_1 => ... => U_m => T_1 => ... => T_n => bool"
  5516     where "p z_1 ... z_m x_1 ... x_n == (x_1, ..., x_n) : s z_1 ... z_m"
  5517 
  5518     inductive "s z_1 ... z_m"
  5519     intros
  5520       rule_1: "... ==> (t_1_1, ..., t_1_n) : s z_1 ... z_m"
  5521       ...
  5522 
  5523   For backward compatibility, there is a wrapper allowing inductive
  5524   sets to be defined with the new package via
  5525 
  5526     inductive_set
  5527       s :: "U_1 => ... => U_m => (T_1 * ... * T_n) set"
  5528       for z_1 :: U_1 and ... and z_n :: U_m
  5529     where
  5530       rule_1: "... ==> (t_1_1, ..., t_1_n) : s z_1 ... z_m"
  5531     | ...
  5532 
  5533   or
  5534 
  5535     inductive_set
  5536       s :: "U_1 => ... => U_m => (T_1 * ... * T_n) set"
  5537       and p :: "U_1 => ... => U_m => T_1 => ... => T_n => bool"
  5538       for z_1 :: U_1 and ... and z_n :: U_m
  5539     where
  5540       "p z_1 ... z_m x_1 ... x_n == (x_1, ..., x_n) : s z_1 ... z_m"
  5541     | rule_1: "... ==> p z_1 ... z_m t_1_1 ... t_1_n"
  5542     | ...
  5543 
  5544   if the additional syntax "p ..." is required.
  5545 
  5546   Numerous examples can be found in the subdirectories src/HOL/Auth,
  5547   src/HOL/Bali, src/HOL/Induct, and src/HOL/MicroJava.
  5548 
  5549   INCOMPATIBILITIES:
  5550 
  5551   - Since declaration and definition of inductive sets or predicates
  5552     is no longer separated, abbreviations involving the newly
  5553     introduced sets or predicates must be specified together with the
  5554     introduction rules after the 'where' keyword (see above), rather
  5555     than before the actual inductive definition.
  5556 
  5557   - The variables in induction and elimination rules are now
  5558     quantified in the order of their occurrence in the introduction
  5559     rules, rather than in alphabetical order. Since this may break
  5560     some proofs, these proofs either have to be repaired, e.g. by
  5561     reordering the variables a_i_1 ... a_i_{k_i} in Isar 'case'
  5562     statements of the form
  5563 
  5564       case (rule_i a_i_1 ... a_i_{k_i})
  5565 
  5566     or the old order of quantification has to be restored by explicitly adding
  5567     meta-level quantifiers in the introduction rules, i.e.
  5568 
  5569       | rule_i: "!!a_i_1 ... a_i_{k_i}. ... ==> p z_1 ... z_m t_i_1 ... t_i_n"
  5570 
  5571   - The format of the elimination rules is now
  5572 
  5573       p z_1 ... z_m x_1 ... x_n ==>
  5574         (!!a_1_1 ... a_1_{k_1}. x_1 = t_1_1 ==> ... ==> x_n = t_1_n ==> ... ==> P)
  5575         ==> ... ==> P
  5576 
  5577     for predicates and
  5578 
  5579       (x_1, ..., x_n) : s z_1 ... z_m ==>
  5580         (!!a_1_1 ... a_1_{k_1}. x_1 = t_1_1 ==> ... ==> x_n = t_1_n ==> ... ==> P)
  5581         ==> ... ==> P
  5582 
  5583     for sets rather than
  5584 
  5585       x : s z_1 ... z_m ==>
  5586         (!!a_1_1 ... a_1_{k_1}. x = (t_1_1, ..., t_1_n) ==> ... ==> P)
  5587         ==> ... ==> P
  5588 
  5589     This may require terms in goals to be expanded to n-tuples
  5590     (e.g. using case_tac or simplification with the split_paired_all
  5591     rule) before the above elimination rule is applicable.
  5592 
  5593   - The elimination or case analysis rules for (mutually) inductive
  5594     sets or predicates are now called "p_1.cases" ... "p_k.cases". The
  5595     list of rules "p_1_..._p_k.elims" is no longer available.
  5596 
  5597 * New package "function"/"fun" for general recursive functions,
  5598 supporting mutual and nested recursion, definitions in local contexts,
  5599 more general pattern matching and partiality. See HOL/ex/Fundefs.thy
  5600 for small examples, and the separate tutorial on the function
  5601 package. The old recdef "package" is still available as before, but
  5602 users are encouraged to use the new package.
  5603 
  5604 * Method "lexicographic_order" automatically synthesizes termination
  5605 relations as lexicographic combinations of size measures. 
  5606 
  5607 * Case-expressions allow arbitrary constructor-patterns (including
  5608 "_") and take their order into account, like in functional
  5609 programming.  Internally, this is translated into nested
  5610 case-expressions; missing cases are added and mapped to the predefined
  5611 constant "undefined". In complicated cases printing may no longer show
  5612 the original input but the internal form. Lambda-abstractions allow
  5613 the same form of pattern matching: "% pat1 => e1 | ..." is an
  5614 abbreviation for "%x. case x of pat1 => e1 | ..." where x is a new
  5615 variable.
  5616 
  5617 * IntDef: The constant "int :: nat => int" has been removed; now "int"
  5618 is an abbreviation for "of_nat :: nat => int". The simplification
  5619 rules for "of_nat" have been changed to work like "int" did
  5620 previously.  Potential INCOMPATIBILITY:
  5621   - "of_nat (Suc m)" simplifies to "1 + of_nat m" instead of "of_nat m + 1"
  5622   - of_nat_diff and of_nat_mult are no longer default simp rules
  5623 
  5624 * Method "algebra" solves polynomial equations over (semi)rings using
  5625 Groebner bases. The (semi)ring structure is defined by locales and the
  5626 tool setup depends on that generic context. Installing the method for
  5627 a specific type involves instantiating the locale and possibly adding
  5628 declarations for computation on the coefficients.  The method is
  5629 already instantiated for natural numbers and for the axiomatic class
  5630 of idoms with numerals.  See also the paper by Chaieb and Wenzel at
  5631 CALCULEMUS 2007 for the general principles underlying this
  5632 architecture of context-aware proof-tools.
  5633 
  5634 * Method "ferrack" implements quantifier elimination over
  5635 special-purpose dense linear orders using locales (analogous to
  5636 "algebra"). The method is already installed for class
  5637 {ordered_field,recpower,number_ring} which subsumes real, hyperreal,
  5638 rat, etc.
  5639 
  5640 * Former constant "List.op @" now named "List.append".  Use ML
  5641 antiquotations @{const_name List.append} or @{term " ... @ ... "} to
  5642 circumvent possible incompatibilities when working on ML level.
  5643 
  5644 * primrec: missing cases mapped to "undefined" instead of "arbitrary".
  5645 
  5646 * New function listsum :: 'a list => 'a for arbitrary monoids.
  5647 Special syntax: "SUM x <- xs. f x" (and latex variants)
  5648 
  5649 * New syntax for Haskell-like list comprehension (input only), eg.
  5650 [(x,y). x <- xs, y <- ys, x ~= y], see also src/HOL/List.thy.
  5651 
  5652 * The special syntax for function "filter" has changed from [x :
  5653 xs. P] to [x <- xs. P] to avoid an ambiguity caused by list
  5654 comprehension syntax, and for uniformity.  INCOMPATIBILITY.
  5655 
  5656 * [a..b] is now defined for arbitrary linear orders.  It used to be
  5657 defined on nat only, as an abbreviation for [a..<Suc b]
  5658 INCOMPATIBILITY.
  5659 
  5660 * Renamed lemma "set_take_whileD"  to "set_takeWhileD".
  5661 
  5662 * New functions "sorted" and "sort" in src/HOL/List.thy.
  5663 
  5664 * New lemma collection field_simps (an extension of ring_simps) for
  5665 manipulating (in)equations involving division. Multiplies with all
  5666 denominators that can be proved to be non-zero (in equations) or
  5667 positive/negative (in inequations).
  5668 
  5669 * Lemma collections ring_eq_simps, group_eq_simps and ring_distrib
  5670 have been improved and renamed to ring_simps, group_simps and
  5671 ring_distribs.  Removed lemmas field_xyz in theory Ring_and_Field
  5672 because they were subsumed by lemmas xyz.  INCOMPATIBILITY.
  5673 
  5674 * Theory Library/Commutative_Ring: switched from recdef to function
  5675 package; constants add, mul, pow now curried.  Infix syntax for
  5676 algebraic operations.
  5677 
  5678 * Dropped redundant lemma def_imp_eq in favor of meta_eq_to_obj_eq.
  5679 INCOMPATIBILITY.
  5680 
  5681 * Dropped redundant lemma if_def2 in favor of if_bool_eq_conj.
  5682 INCOMPATIBILITY.
  5683 
  5684 * HOL/records: generalised field-update to take a function on the
  5685 field rather than the new value: r(|A := x|) is translated to A_update
  5686 (K x) r The K-combinator that is internally used is called K_record.
  5687 INCOMPATIBILITY: Usage of the plain update functions has to be
  5688 adapted.
  5689  
  5690 * Class "semiring_0" now contains annihilation axioms x * 0 = 0 and 0
  5691 * x = 0, which are required for a semiring.  Richer structures do not
  5692 inherit from semiring_0 anymore, because this property is a theorem
  5693 there, not an axiom.  INCOMPATIBILITY: In instances of semiring_0,
  5694 there is more to prove, but this is mostly trivial.
  5695 
  5696 * Class "recpower" is generalized to arbitrary monoids, not just
  5697 commutative semirings.  INCOMPATIBILITY: may need to incorporate
  5698 commutativity or semiring properties additionally.
  5699 
  5700 * Constant "List.list_all2" in List.thy now uses authentic syntax.
  5701 INCOMPATIBILITY: translations containing list_all2 may go wrong,
  5702 better use 'abbreviation'.
  5703 
  5704 * Renamed constant "List.op mem" to "List.member".  INCOMPATIBILITY.
  5705 
  5706 * Numeral syntax: type 'bin' which was a mere type copy of 'int' has
  5707 been abandoned in favour of plain 'int'.  INCOMPATIBILITY --
  5708 significant changes for setting up numeral syntax for types:
  5709   - New constants Numeral.pred and Numeral.succ instead
  5710       of former Numeral.bin_pred and Numeral.bin_succ.
  5711   - Use integer operations instead of bin_add, bin_mult and so on.
  5712   - Numeral simplification theorems named Numeral.numeral_simps instead of Bin_simps.
  5713   - ML structure Bin_Simprocs now named Int_Numeral_Base_Simprocs.
  5714 
  5715 See src/HOL/Integ/IntArith.thy for an example setup.
  5716 
  5717 * Command 'normal_form' computes the normal form of a term that may
  5718 contain free variables.  For example ``normal_form "rev [a, b, c]"''
  5719 produces ``[b, c, a]'' (without proof).  This command is suitable for
  5720 heavy-duty computations because the functions are compiled to ML
  5721 first.  Correspondingly, a method "normalization" is provided.  See
  5722 further src/HOL/ex/NormalForm.thy and src/Tools/nbe.ML.
  5723 
  5724 * Alternative iff syntax "A <-> B" for equality on bool (with priority
  5725 25 like -->); output depends on the "iff" print_mode, the default is
  5726 "A = B" (with priority 50).
  5727 
  5728 * Relations less (<) and less_eq (<=) are also available on type bool.
  5729 Modified syntax to disallow nesting without explicit parentheses,
  5730 e.g. "(x < y) < z" or "x < (y < z)", but NOT "x < y < z".  Potential
  5731 INCOMPATIBILITY.
  5732 
  5733 * "LEAST x:A. P" expands to "LEAST x. x:A & P" (input only).
  5734 
  5735 * Relation composition operator "op O" now has precedence 75 and binds
  5736 stronger than union and intersection. INCOMPATIBILITY.
  5737 
  5738 * The old set interval syntax "{m..n(}" (and relatives) has been
  5739 removed.  Use "{m..<n}" (and relatives) instead.
  5740 
  5741 * In the context of the assumption "~(s = t)" the Simplifier rewrites
  5742 "t = s" to False (by simproc "neq").  INCOMPATIBILITY, consider using
  5743 ``declare [[simproc del: neq]]''.
  5744 
  5745 * Simplifier: "m dvd n" where m and n are numbers is evaluated to
  5746 True/False.
  5747 
  5748 * Theorem Cons_eq_map_conv no longer declared as "simp".
  5749 
  5750 * Theorem setsum_mult renamed to setsum_right_distrib.
  5751 
  5752 * Prefer ex1I over ex_ex1I in single-step reasoning, e.g. by the
  5753 ``rule'' method.
  5754 
  5755 * Reimplemented methods "sat" and "satx", with several improvements:
  5756 goals no longer need to be stated as "<prems> ==> False", equivalences
  5757 (i.e. "=" on type bool) are handled, variable names of the form
  5758 "lit_<n>" are no longer reserved, significant speedup.
  5759 
  5760 * Methods "sat" and "satx" can now replay MiniSat proof traces.
  5761 zChaff is still supported as well.
  5762 
  5763 * 'inductive' and 'datatype': provide projections of mutual rules,
  5764 bundled as foo_bar.inducts;
  5765 
  5766 * Library: moved theories Parity, GCD, Binomial, Infinite_Set to
  5767 Library.
  5768 
  5769 * Library: moved theory Accessible_Part to main HOL.
  5770 
  5771 * Library: added theory Coinductive_List of potentially infinite lists
  5772 as greatest fixed-point.
  5773 
  5774 * Library: added theory AssocList which implements (finite) maps as
  5775 association lists.
  5776 
  5777 * Method "evaluation" solves goals (i.e. a boolean expression)
  5778 efficiently by compiling it to ML.  The goal is "proved" (via an
  5779 oracle) if it evaluates to True.
  5780 
  5781 * Linear arithmetic now splits certain operators (e.g. min, max, abs)
  5782 also when invoked by the simplifier.  This results in the Simplifier
  5783 being more powerful on arithmetic goals.  INCOMPATIBILITY.
  5784 Configuration option fast_arith_split_limit=0 recovers the old
  5785 behavior.
  5786 
  5787 * Support for hex (0x20) and binary (0b1001) numerals.
  5788 
  5789 * New method: reify eqs (t), where eqs are equations for an
  5790 interpretation I :: 'a list => 'b => 'c and t::'c is an optional
  5791 parameter, computes a term s::'b and a list xs::'a list and proves the
  5792 theorem I xs s = t. This is also known as reification or quoting. The
  5793 resulting theorem is applied to the subgoal to substitute t with I xs
  5794 s.  If t is omitted, the subgoal itself is reified.
  5795 
  5796 * New method: reflection corr_thm eqs (t). The parameters eqs and (t)
  5797 are as explained above. corr_thm is a theorem for I vs (f t) = I vs t,
  5798 where f is supposed to be a computable function (in the sense of code
  5799 generattion). The method uses reify to compute s and xs as above then
  5800 applies corr_thm and uses normalization by evaluation to "prove" f s =
  5801 r and finally gets the theorem t = r, which is again applied to the
  5802 subgoal. An Example is available in src/HOL/ex/ReflectionEx.thy.
  5803 
  5804 * Reflection: Automatic reification now handels binding, an example is
  5805 available in src/HOL/ex/ReflectionEx.thy
  5806 
  5807 * HOL-Statespace: ``State Spaces: The Locale Way'' introduces a
  5808 command 'statespace' that is similar to 'record', but introduces an
  5809 abstract specification based on the locale infrastructure instead of
  5810 HOL types.  This leads to extra flexibility in composing state spaces,
  5811 in particular multiple inheritance and renaming of components.
  5812 
  5813 
  5814 *** HOL-Complex ***
  5815 
  5816 * Hyperreal: Functions root and sqrt are now defined on negative real
  5817 inputs so that root n (- x) = - root n x and sqrt (- x) = - sqrt x.
  5818 Nonnegativity side conditions have been removed from many lemmas, so
  5819 that more subgoals may now be solved by simplification; potential
  5820 INCOMPATIBILITY.
  5821 
  5822 * Real: new type classes formalize real normed vector spaces and
  5823 algebras, using new overloaded constants scaleR :: real => 'a => 'a
  5824 and norm :: 'a => real.
  5825 
  5826 * Real: constant of_real :: real => 'a::real_algebra_1 injects from
  5827 reals into other types. The overloaded constant Reals :: 'a set is now
  5828 defined as range of_real; potential INCOMPATIBILITY.
  5829 
  5830 * Real: proper support for ML code generation, including 'quickcheck'.
  5831 Reals are implemented as arbitrary precision rationals.
  5832 
  5833 * Hyperreal: Several constants that previously worked only for the
  5834 reals have been generalized, so they now work over arbitrary vector
  5835 spaces. Type annotations may need to be added in some cases; potential
  5836 INCOMPATIBILITY.
  5837 
  5838   Infinitesimal  :: ('a::real_normed_vector) star set
  5839   HFinite        :: ('a::real_normed_vector) star set
  5840   HInfinite      :: ('a::real_normed_vector) star set
  5841   approx         :: ('a::real_normed_vector) star => 'a star => bool
  5842   monad          :: ('a::real_normed_vector) star => 'a star set
  5843   galaxy         :: ('a::real_normed_vector) star => 'a star set
  5844   (NS)LIMSEQ     :: [nat => 'a::real_normed_vector, 'a] => bool
  5845   (NS)convergent :: (nat => 'a::real_normed_vector) => bool
  5846   (NS)Bseq       :: (nat => 'a::real_normed_vector) => bool
  5847   (NS)Cauchy     :: (nat => 'a::real_normed_vector) => bool
  5848   (NS)LIM        :: ['a::real_normed_vector => 'b::real_normed_vector, 'a, 'b] => bool
  5849   is(NS)Cont     :: ['a::real_normed_vector => 'b::real_normed_vector, 'a] => bool
  5850   deriv          :: ['a::real_normed_field => 'a, 'a, 'a] => bool
  5851   sgn            :: 'a::real_normed_vector => 'a
  5852   exp            :: 'a::{recpower,real_normed_field,banach} => 'a
  5853 
  5854 * Complex: Some complex-specific constants are now abbreviations for
  5855 overloaded ones: complex_of_real = of_real, cmod = norm, hcmod =
  5856 hnorm.  Other constants have been entirely removed in favor of the
  5857 polymorphic versions (INCOMPATIBILITY):
  5858 
  5859   approx        <-- capprox
  5860   HFinite       <-- CFinite
  5861   HInfinite     <-- CInfinite
  5862   Infinitesimal <-- CInfinitesimal
  5863   monad         <-- cmonad
  5864   galaxy        <-- cgalaxy
  5865   (NS)LIM       <-- (NS)CLIM, (NS)CRLIM
  5866   is(NS)Cont    <-- is(NS)Contc, is(NS)contCR
  5867   (ns)deriv     <-- (ns)cderiv
  5868 
  5869 
  5870 *** HOL-Algebra ***
  5871 
  5872 * Formalisation of ideals and the quotient construction over rings.
  5873 
  5874 * Order and lattice theory no longer based on records.
  5875 INCOMPATIBILITY.
  5876 
  5877 * Renamed lemmas least_carrier -> least_closed and greatest_carrier ->
  5878 greatest_closed.  INCOMPATIBILITY.
  5879 
  5880 * Method algebra is now set up via an attribute.  For examples see
  5881 Ring.thy.  INCOMPATIBILITY: the method is now weaker on combinations
  5882 of algebraic structures.
  5883 
  5884 * Renamed theory CRing to Ring.
  5885 
  5886 
  5887 *** HOL-Nominal ***
  5888 
  5889 * Substantial, yet incomplete support for nominal datatypes (binding
  5890 structures) based on HOL-Nominal logic.  See src/HOL/Nominal and
  5891 src/HOL/Nominal/Examples.  Prospective users should consult
  5892 http://isabelle.in.tum.de/nominal/
  5893 
  5894 
  5895 *** ML ***
  5896 
  5897 * ML basics: just one true type int, which coincides with IntInf.int
  5898 (even on SML/NJ).
  5899 
  5900 * ML within Isar: antiquotations allow to embed statically-checked
  5901 formal entities in the source, referring to the context available at
  5902 compile-time.  For example:
  5903 
  5904 ML {* @{sort "{zero,one}"} *}
  5905 ML {* @{typ "'a => 'b"} *}
  5906 ML {* @{term "%x. x"} *}
  5907 ML {* @{prop "x == y"} *}
  5908 ML {* @{ctyp "'a => 'b"} *}
  5909 ML {* @{cterm "%x. x"} *}
  5910 ML {* @{cprop "x == y"} *}
  5911 ML {* @{thm asm_rl} *}
  5912 ML {* @{thms asm_rl} *}
  5913 ML {* @{type_name c} *}
  5914 ML {* @{type_syntax c} *}
  5915 ML {* @{const_name c} *}
  5916 ML {* @{const_syntax c} *}
  5917 ML {* @{context} *}
  5918 ML {* @{theory} *}
  5919 ML {* @{theory Pure} *}
  5920 ML {* @{theory_ref} *}
  5921 ML {* @{theory_ref Pure} *}
  5922 ML {* @{simpset} *}
  5923 ML {* @{claset} *}
  5924 ML {* @{clasimpset} *}
  5925 
  5926 The same works for sources being ``used'' within an Isar context.
  5927 
  5928 * ML in Isar: improved error reporting; extra verbosity with
  5929 ML_Context.trace enabled.
  5930 
  5931 * Pure/General/table.ML: the join operations now works via exceptions
  5932 DUP/SAME instead of type option. This is simpler in simple cases, and
  5933 admits slightly more efficient complex applications.
  5934 
  5935 * Pure: 'advanced' translation functions (parse_translation etc.) now
  5936 use Context.generic instead of just theory.
  5937 
  5938 * Pure: datatype Context.generic joins theory/Proof.context and
  5939 provides some facilities for code that works in either kind of
  5940 context, notably GenericDataFun for uniform theory and proof data.
  5941 
  5942 * Pure: simplified internal attribute type, which is now always
  5943 Context.generic * thm -> Context.generic * thm. Global (theory) vs.
  5944 local (Proof.context) attributes have been discontinued, while
  5945 minimizing code duplication. Thm.rule_attribute and
  5946 Thm.declaration_attribute build canonical attributes; see also structure
  5947 Context for further operations on Context.generic, notably
  5948 GenericDataFun. INCOMPATIBILITY, need to adapt attribute type
  5949 declarations and definitions.
  5950 
  5951 * Context data interfaces (Theory/Proof/GenericDataFun): removed
  5952 name/print, uninitialized data defaults to ad-hoc copy of empty value,
  5953 init only required for impure data. INCOMPATIBILITY: empty really need
  5954 to be empty (no dependencies on theory content!)
  5955 
  5956 * Pure/kernel: consts certification ignores sort constraints given in
  5957 signature declarations. (This information is not relevant to the
  5958 logic, but only for type inference.) SIGNIFICANT INTERNAL CHANGE,
  5959 potential INCOMPATIBILITY.
  5960 
  5961 * Pure: axiomatic type classes are now purely definitional, with
  5962 explicit proofs of class axioms and super class relations performed
  5963 internally. See Pure/axclass.ML for the main internal interfaces --
  5964 notably AxClass.define_class supercedes AxClass.add_axclass, and
  5965 AxClass.axiomatize_class/classrel/arity supersede
  5966 Sign.add_classes/classrel/arities.
  5967 
  5968 * Pure/Isar: Args/Attrib parsers operate on Context.generic --
  5969 global/local versions on theory vs. Proof.context have been
  5970 discontinued; Attrib.syntax and Method.syntax have been adapted
  5971 accordingly.  INCOMPATIBILITY, need to adapt parser expressions for
  5972 attributes, methods, etc.
  5973 
  5974 * Pure: several functions of signature "... -> theory -> theory * ..."
  5975 have been reoriented to "... -> theory -> ... * theory" in order to
  5976 allow natural usage in combination with the ||>, ||>>, |-> and
  5977 fold_map combinators.
  5978 
  5979 * Pure: official theorem names (closed derivations) and additional
  5980 comments (tags) are now strictly separate.  Name hints -- which are
  5981 maintained as tags -- may be attached any time without affecting the
  5982 derivation.
  5983 
  5984 * Pure: primitive rule lift_rule now takes goal cterm instead of an
  5985 actual goal state (thm).  Use Thm.lift_rule (Thm.cprem_of st i) to
  5986 achieve the old behaviour.
  5987 
  5988 * Pure: the "Goal" constant is now called "prop", supporting a
  5989 slightly more general idea of ``protecting'' meta-level rule
  5990 statements.
  5991 
  5992 * Pure: Logic.(un)varify only works in a global context, which is now
  5993 enforced instead of silently assumed.  INCOMPATIBILITY, may use
  5994 Logic.legacy_(un)varify as temporary workaround.
  5995 
  5996 * Pure: structure Name provides scalable operations for generating
  5997 internal variable names, notably Name.variants etc.  This replaces
  5998 some popular functions from term.ML:
  5999 
  6000   Term.variant		->  Name.variant
  6001   Term.variantlist	->  Name.variant_list
  6002   Term.invent_names	->  Name.invent_list
  6003 
  6004 Note that low-level renaming rarely occurs in new code -- operations
  6005 from structure Variable are used instead (see below).
  6006 
  6007 * Pure: structure Variable provides fundamental operations for proper
  6008 treatment of fixed/schematic variables in a context.  For example,
  6009 Variable.import introduces fixes for schematics of given facts and
  6010 Variable.export reverses the effect (up to renaming) -- this replaces
  6011 various freeze_thaw operations.
  6012 
  6013 * Pure: structure Goal provides simple interfaces for
  6014 init/conclude/finish and tactical prove operations (replacing former
  6015 Tactic.prove).  Goal.prove is the canonical way to prove results
  6016 within a given context; Goal.prove_global is a degraded version for
  6017 theory level goals, including a global Drule.standard.  Note that
  6018 OldGoals.prove_goalw_cterm has long been obsolete, since it is
  6019 ill-behaved in a local proof context (e.g. with local fixes/assumes or
  6020 in a locale context).
  6021 
  6022 * Pure/Syntax: generic interfaces for parsing (Syntax.parse_term etc.)
  6023 and type checking (Syntax.check_term etc.), with common combinations
  6024 (Syntax.read_term etc.). These supersede former Sign.read_term etc.
  6025 which are considered legacy and await removal.
  6026 
  6027 * Pure/Syntax: generic interfaces for type unchecking
  6028 (Syntax.uncheck_terms etc.) and unparsing (Syntax.unparse_term etc.),
  6029 with common combinations (Syntax.pretty_term, Syntax.string_of_term
  6030 etc.).  Former Sign.pretty_term, Sign.string_of_term etc. are still
  6031 available for convenience, but refer to the very same operations using
  6032 a mere theory instead of a full context.
  6033 
  6034 * Isar: simplified treatment of user-level errors, using exception
  6035 ERROR of string uniformly.  Function error now merely raises ERROR,
  6036 without any side effect on output channels.  The Isar toplevel takes
  6037 care of proper display of ERROR exceptions.  ML code may use plain
  6038 handle/can/try; cat_error may be used to concatenate errors like this:
  6039 
  6040   ... handle ERROR msg => cat_error msg "..."
  6041 
  6042 Toplevel ML code (run directly or through the Isar toplevel) may be
  6043 embedded into the Isar toplevel with exception display/debug like
  6044 this:
  6045 
  6046   Isar.toplevel (fn () => ...)
  6047 
  6048 INCOMPATIBILITY, removed special transform_error facilities, removed
  6049 obsolete variants of user-level exceptions (ERROR_MESSAGE,
  6050 Context.PROOF, ProofContext.CONTEXT, Proof.STATE, ProofHistory.FAIL)
  6051 -- use plain ERROR instead.
  6052 
  6053 * Isar: theory setup now has type (theory -> theory), instead of a
  6054 list.  INCOMPATIBILITY, may use #> to compose setup functions.
  6055 
  6056 * Isar: ML toplevel pretty printer for type Proof.context, subject to
  6057 ProofContext.debug/verbose flags.
  6058 
  6059 * Isar: Toplevel.theory_to_proof admits transactions that modify the
  6060 theory before entering a proof state.  Transactions now always see a
  6061 quasi-functional intermediate checkpoint, both in interactive and
  6062 batch mode.
  6063 
  6064 * Isar: simplified interfaces for outer syntax.  Renamed
  6065 OuterSyntax.add_keywords to OuterSyntax.keywords.  Removed
  6066 OuterSyntax.add_parsers -- this functionality is now included in
  6067 OuterSyntax.command etc.  INCOMPATIBILITY.
  6068 
  6069 * Simplifier: the simpset of a running simplification process now
  6070 contains a proof context (cf. Simplifier.the_context), which is the
  6071 very context that the initial simpset has been retrieved from (by
  6072 simpset_of/local_simpset_of).  Consequently, all plug-in components
  6073 (solver, looper etc.) may depend on arbitrary proof data.
  6074 
  6075 * Simplifier.inherit_context inherits the proof context (plus the
  6076 local bounds) of the current simplification process; any simproc
  6077 etc. that calls the Simplifier recursively should do this!  Removed
  6078 former Simplifier.inherit_bounds, which is already included here --
  6079 INCOMPATIBILITY.  Tools based on low-level rewriting may even have to
  6080 specify an explicit context using Simplifier.context/theory_context.
  6081 
  6082 * Simplifier/Classical Reasoner: more abstract interfaces
  6083 change_simpset/claset for modifying the simpset/claset reference of a
  6084 theory; raw versions simpset/claset_ref etc. have been discontinued --
  6085 INCOMPATIBILITY.
  6086 
  6087 * Provers: more generic wrt. syntax of object-logics, avoid hardwired
  6088 "Trueprop" etc.
  6089 
  6090 
  6091 *** System ***
  6092 
  6093 * settings: the default heap location within ISABELLE_HOME_USER now
  6094 includes ISABELLE_IDENTIFIER.  This simplifies use of multiple
  6095 Isabelle installations.
  6096 
  6097 * isabelle-process: option -S (secure mode) disables some critical
  6098 operations, notably runtime compilation and evaluation of ML source
  6099 code.
  6100 
  6101 * Basic Isabelle mode for jEdit, see Isabelle/lib/jedit/.
  6102 
  6103 * Support for parallel execution, using native multicore support of
  6104 Poly/ML 5.1.  The theory loader exploits parallelism when processing
  6105 independent theories, according to the given theory header
  6106 specifications. The maximum number of worker threads is specified via
  6107 usedir option -M or the "max-threads" setting in Proof General. A
  6108 speedup factor of 1.5--3.5 can be expected on a 4-core machine, and up
  6109 to 6 on a 8-core machine.  User-code needs to observe certain
  6110 guidelines for thread-safe programming, see appendix A in the Isar
  6111 Implementation manual.
  6112 
  6113 
  6114 
  6115 New in Isabelle2005 (October 2005)
  6116 ----------------------------------
  6117 
  6118 *** General ***
  6119 
  6120 * Theory headers: the new header syntax for Isar theories is
  6121 
  6122   theory <name>
  6123   imports <theory1> ... <theoryN>
  6124   uses <file1> ... <fileM>
  6125   begin
  6126 
  6127 where the 'uses' part is optional.  The previous syntax
  6128 
  6129   theory <name> = <theory1> + ... + <theoryN>:
  6130 
  6131 will disappear in the next release.  Use isatool fixheaders to convert
  6132 existing theory files.  Note that there is no change in ancient
  6133 non-Isar theories now, but these will disappear soon.
  6134 
  6135 * Theory loader: parent theories can now also be referred to via
  6136 relative and absolute paths.
  6137 
  6138 * Command 'find_theorems' searches for a list of criteria instead of a
  6139 list of constants. Known criteria are: intro, elim, dest, name:string,
  6140 simp:term, and any term. Criteria can be preceded by '-' to select
  6141 theorems that do not match. Intro, elim, dest select theorems that
  6142 match the current goal, name:s selects theorems whose fully qualified
  6143 name contain s, and simp:term selects all simplification rules whose
  6144 lhs match term.  Any other term is interpreted as pattern and selects
  6145 all theorems matching the pattern. Available in ProofGeneral under
  6146 'ProofGeneral -> Find Theorems' or C-c C-f.  Example:
  6147 
  6148   C-c C-f (100) "(_::nat) + _ + _" intro -name: "HOL."
  6149 
  6150 prints the last 100 theorems matching the pattern "(_::nat) + _ + _",
  6151 matching the current goal as introduction rule and not having "HOL."
  6152 in their name (i.e. not being defined in theory HOL).
  6153 
  6154 * Command 'thms_containing' has been discontinued in favour of
  6155 'find_theorems'; INCOMPATIBILITY.
  6156 
  6157 * Communication with Proof General is now 8bit clean, which means that
  6158 Unicode text in UTF-8 encoding may be used within theory texts (both
  6159 formal and informal parts).  Cf. option -U of the Isabelle Proof
  6160 General interface.  Here are some simple examples (cf. src/HOL/ex):
  6161 
  6162   http://isabelle.in.tum.de/library/HOL/ex/Hebrew.html
  6163   http://isabelle.in.tum.de/library/HOL/ex/Chinese.html
  6164 
  6165 * Improved efficiency of the Simplifier and, to a lesser degree, the
  6166 Classical Reasoner.  Typical big applications run around 2 times
  6167 faster.
  6168 
  6169 
  6170 *** Document preparation ***
  6171 
  6172 * Commands 'display_drafts' and 'print_drafts' perform simple output
  6173 of raw sources.  Only those symbols that do not require additional
  6174 LaTeX packages (depending on comments in isabellesym.sty) are
  6175 displayed properly, everything else is left verbatim.  isatool display
  6176 and isatool print are used as front ends (these are subject to the
  6177 DVI/PDF_VIEWER and PRINT_COMMAND settings, respectively).
  6178 
  6179 * Command tags control specific markup of certain regions of text,
  6180 notably folding and hiding.  Predefined tags include "theory" (for
  6181 theory begin and end), "proof" for proof commands, and "ML" for
  6182 commands involving ML code; the additional tags "visible" and
  6183 "invisible" are unused by default.  Users may give explicit tag
  6184 specifications in the text, e.g. ''by %invisible (auto)''.  The
  6185 interpretation of tags is determined by the LaTeX job during document
  6186 preparation: see option -V of isatool usedir, or options -n and -t of
  6187 isatool document, or even the LaTeX macros \isakeeptag, \isafoldtag,
  6188 \isadroptag.
  6189 
  6190 Several document versions may be produced at the same time via isatool
  6191 usedir (the generated index.html will link all of them).  Typical
  6192 specifications include ''-V document=theory,proof,ML'' to present
  6193 theory/proof/ML parts faithfully, ''-V outline=/proof,/ML'' to fold
  6194 proof and ML commands, and ''-V mutilated=-theory,-proof,-ML'' to omit
  6195 these parts without any formal replacement text.  The Isabelle site
  6196 default settings produce ''document'' and ''outline'' versions as
  6197 specified above.
  6198 
  6199 * Several new antiquotations:
  6200 
  6201   @{term_type term} prints a term with its type annotated;
  6202 
  6203   @{typeof term} prints the type of a term;
  6204 
  6205   @{const const} is the same as @{term const}, but checks that the
  6206   argument is a known logical constant;
  6207 
  6208   @{term_style style term} and @{thm_style style thm} print a term or
  6209   theorem applying a "style" to it
  6210 
  6211   @{ML text}
  6212 
  6213 Predefined styles are 'lhs' and 'rhs' printing the lhs/rhs of
  6214 definitions, equations, inequations etc., 'concl' printing only the
  6215 conclusion of a meta-logical statement theorem, and 'prem1' .. 'prem19'
  6216 to print the specified premise.  TermStyle.add_style provides an ML
  6217 interface for introducing further styles.  See also the "LaTeX Sugar"
  6218 document practical applications.  The ML antiquotation prints
  6219 type-checked ML expressions verbatim.
  6220 
  6221 * Markup commands 'chapter', 'section', 'subsection', 'subsubsection',
  6222 and 'text' support optional locale specification '(in loc)', which
  6223 specifies the default context for interpreting antiquotations.  For
  6224 example: 'text (in lattice) {* @{thm inf_assoc}*}'.
  6225 
  6226 * Option 'locale=NAME' of antiquotations specifies an alternative
  6227 context interpreting the subsequent argument.  For example: @{thm
  6228 [locale=lattice] inf_assoc}.
  6229 
  6230 * Proper output of proof terms (@{prf ...} and @{full_prf ...}) within
  6231 a proof context.
  6232 
  6233 * Proper output of antiquotations for theory commands involving a
  6234 proof context (such as 'locale' or 'theorem (in loc) ...').
  6235 
  6236 * Delimiters of outer tokens (string etc.) now produce separate LaTeX
  6237 macros (\