NEWS
author kuncar
Thu May 24 14:20:23 2012 +0200 (2012-05-24)
changeset 47982 7aa35601ff65
parent 47967 c422128d3889
child 47979 59ec72d3d0b9
permissions -rw-r--r--
prove reflexivity also for the quotient composition relation; reflp_preserve renamed to reflexivity_rule
     1 Isabelle NEWS -- history user-relevant changes
     2 ==============================================
     3 
     4 New in this Isabelle version
     5 ----------------------------
     6 
     7 *** General ***
     8 
     9 * Discontinued obsolete method fastsimp / tactic fast_simp_tac, which
    10 is called fastforce / fast_force_tac already since Isabelle2011-1.
    11 
    12 
    13 
    14 New in Isabelle2012 (May 2012)
    15 ------------------------------
    16 
    17 *** General ***
    18 
    19 * Prover IDE (PIDE) improvements:
    20 
    21   - more robust Sledgehammer integration (as before the sledgehammer
    22     command-line needs to be typed into the source buffer)
    23   - markup for bound variables
    24   - markup for types of term variables (displayed as tooltips)
    25   - support for user-defined Isar commands within the running session
    26   - improved support for Unicode outside original 16bit range
    27     e.g. glyph for \<A> (thanks to jEdit 4.5.1)
    28 
    29 * Forward declaration of outer syntax keywords within the theory
    30 header -- minor INCOMPATIBILITY for user-defined commands.  Allow new
    31 commands to be used in the same theory where defined.
    32 
    33 * Auxiliary contexts indicate block structure for specifications with
    34 additional parameters and assumptions.  Such unnamed contexts may be
    35 nested within other targets, like 'theory', 'locale', 'class',
    36 'instantiation' etc.  Results from the local context are generalized
    37 accordingly and applied to the enclosing target context.  Example:
    38 
    39   context
    40     fixes x y z :: 'a
    41     assumes xy: "x = y" and yz: "y = z"
    42   begin
    43 
    44   lemma my_trans: "x = z" using xy yz by simp
    45 
    46   end
    47 
    48   thm my_trans
    49 
    50 The most basic application is to factor-out context elements of
    51 several fixes/assumes/shows theorem statements, e.g. see
    52 ~~/src/HOL/Isar_Examples/Group_Context.thy
    53 
    54 Any other local theory specification element works within the "context
    55 ... begin ... end" block as well.
    56 
    57 * Bundled declarations associate attributed fact expressions with a
    58 given name in the context.  These may be later included in other
    59 contexts.  This allows to manage context extensions casually, without
    60 the logical dependencies of locales and locale interpretation.  See
    61 commands 'bundle', 'include', 'including' etc. in the isar-ref manual.
    62 
    63 * Commands 'lemmas' and 'theorems' allow local variables using 'for'
    64 declaration, and results are standardized before being stored.  Thus
    65 old-style "standard" after instantiation or composition of facts
    66 becomes obsolete.  Minor INCOMPATIBILITY, due to potential change of
    67 indices of schematic variables.
    68 
    69 * Rule attributes in local theory declarations (e.g. locale or class)
    70 are now statically evaluated: the resulting theorem is stored instead
    71 of the original expression.  INCOMPATIBILITY in rare situations, where
    72 the historic accident of dynamic re-evaluation in interpretations
    73 etc. was exploited.
    74 
    75 * New tutorial "Programming and Proving in Isabelle/HOL"
    76 ("prog-prove").  It completely supersedes "A Tutorial Introduction to
    77 Structured Isar Proofs" ("isar-overview"), which has been removed.  It
    78 also supersedes "Isabelle/HOL, A Proof Assistant for Higher-Order
    79 Logic" as the recommended beginners tutorial, but does not cover all
    80 of the material of that old tutorial.
    81 
    82 * Updated and extended reference manuals: "isar-ref",
    83 "implementation", "system"; reduced remaining material in old "ref"
    84 manual.
    85 
    86 
    87 *** Pure ***
    88 
    89 * Command 'definition' no longer exports the foundational "raw_def"
    90 into the user context.  Minor INCOMPATIBILITY, may use the regular
    91 "def" result with attribute "abs_def" to imitate the old version.
    92 
    93 * Attribute "abs_def" turns an equation of the form "f x y == t" into
    94 "f == %x y. t", which ensures that "simp" or "unfold" steps always
    95 expand it.  This also works for object-logic equality.  (Formerly
    96 undocumented feature.)
    97 
    98 * Sort constraints are now propagated in simultaneous statements, just
    99 like type constraints.  INCOMPATIBILITY in rare situations, where
   100 distinct sorts used to be assigned accidentally.  For example:
   101 
   102   lemma "P (x::'a::foo)" and "Q (y::'a::bar)"  -- "now illegal"
   103 
   104   lemma "P (x::'a)" and "Q (y::'a::bar)"
   105     -- "now uniform 'a::bar instead of default sort for first occurrence (!)"
   106 
   107 * Rule composition via attribute "OF" (or ML functions OF/MRS) is more
   108 tolerant against multiple unifiers, as long as the final result is
   109 unique.  (As before, rules are composed in canonical right-to-left
   110 order to accommodate newly introduced premises.)
   111 
   112 * Renamed some inner syntax categories:
   113 
   114     num ~> num_token
   115     xnum ~> xnum_token
   116     xstr ~> str_token
   117 
   118 Minor INCOMPATIBILITY.  Note that in practice "num_const" or
   119 "num_position" etc. are mainly used instead (which also include
   120 position information via constraints).
   121 
   122 * Simplified configuration options for syntax ambiguity: see
   123 "syntax_ambiguity_warning" and "syntax_ambiguity_limit" in isar-ref
   124 manual.  Minor INCOMPATIBILITY.
   125 
   126 * Discontinued configuration option "syntax_positions": atomic terms
   127 in parse trees are always annotated by position constraints.
   128 
   129 * Old code generator for SML and its commands 'code_module',
   130 'code_library', 'consts_code', 'types_code' have been discontinued.
   131 Use commands of the generic code generator instead.  INCOMPATIBILITY.
   132 
   133 * Redundant attribute "code_inline" has been discontinued. Use
   134 "code_unfold" instead.  INCOMPATIBILITY.
   135 
   136 * Dropped attribute "code_unfold_post" in favor of the its dual
   137 "code_abbrev", which yields a common pattern in definitions like
   138 
   139   definition [code_abbrev]: "f = t"
   140 
   141 INCOMPATIBILITY.
   142 
   143 * Obsolete 'types' command has been discontinued.  Use 'type_synonym'
   144 instead.  INCOMPATIBILITY.
   145 
   146 * Discontinued old "prems" fact, which used to refer to the accidental
   147 collection of foundational premises in the context (already marked as
   148 legacy since Isabelle2011).
   149 
   150 
   151 *** HOL ***
   152 
   153 * Type 'a set is now a proper type constructor (just as before
   154 Isabelle2008).  Definitions mem_def and Collect_def have disappeared.
   155 Non-trivial INCOMPATIBILITY.  For developments keeping predicates and
   156 sets separate, it is often sufficient to rephrase some set S that has
   157 been accidentally used as predicates by "%x. x : S", and some
   158 predicate P that has been accidentally used as set by "{x. P x}".
   159 Corresponding proofs in a first step should be pruned from any
   160 tinkering with former theorems mem_def and Collect_def as far as
   161 possible.
   162 
   163 For developments which deliberately mix predicates and sets, a
   164 planning step is necessary to determine what should become a predicate
   165 and what a set.  It can be helpful to carry out that step in
   166 Isabelle2011-1 before jumping right into the current release.
   167 
   168 * Code generation by default implements sets as container type rather
   169 than predicates.  INCOMPATIBILITY.
   170 
   171 * New type synonym 'a rel = ('a * 'a) set
   172 
   173 * The representation of numerals has changed.  Datatype "num"
   174 represents strictly positive binary numerals, along with functions
   175 "numeral :: num => 'a" and "neg_numeral :: num => 'a" to represent
   176 positive and negated numeric literals, respectively.  See also
   177 definitions in ~~/src/HOL/Num.thy.  Potential INCOMPATIBILITY, some
   178 user theories may require adaptations as follows:
   179 
   180   - Theorems with number_ring or number_semiring constraints: These
   181     classes are gone; use comm_ring_1 or comm_semiring_1 instead.
   182 
   183   - Theories defining numeric types: Remove number, number_semiring,
   184     and number_ring instances. Defer all theorems about numerals until
   185     after classes one and semigroup_add have been instantiated.
   186 
   187   - Numeral-only simp rules: Replace each rule having a "number_of v"
   188     pattern with two copies, one for numeral and one for neg_numeral.
   189 
   190   - Theorems about subclasses of semiring_1 or ring_1: These classes
   191     automatically support numerals now, so more simp rules and
   192     simprocs may now apply within the proof.
   193 
   194   - Definitions and theorems using old constructors Pls/Min/Bit0/Bit1:
   195     Redefine using other integer operations.
   196 
   197 * Transfer: New package intended to generalize the existing
   198 "descending" method and related theorem attributes from the Quotient
   199 package.  (Not all functionality is implemented yet, but future
   200 development will focus on Transfer as an eventual replacement for the
   201 corresponding parts of the Quotient package.)
   202 
   203   - transfer_rule attribute: Maintains a collection of transfer rules,
   204     which relate constants at two different types. Transfer rules may
   205     relate different type instances of the same polymorphic constant,
   206     or they may relate an operation on a raw type to a corresponding
   207     operation on an abstract type (quotient or subtype). For example:
   208 
   209     ((A ===> B) ===> list_all2 A ===> list_all2 B) map map
   210     (cr_int ===> cr_int ===> cr_int) (%(x,y) (u,v). (x+u, y+v)) plus_int
   211 
   212   - transfer method: Replaces a subgoal on abstract types with an
   213     equivalent subgoal on the corresponding raw types. Constants are
   214     replaced with corresponding ones according to the transfer rules.
   215     Goals are generalized over all free variables by default; this is
   216     necessary for variables whose types change, but can be overridden
   217     for specific variables with e.g. "transfer fixing: x y z".  The
   218     variant transfer' method allows replacing a subgoal with one that
   219     is logically stronger (rather than equivalent).
   220 
   221   - relator_eq attribute: Collects identity laws for relators of
   222     various type constructors, e.g. "list_all2 (op =) = (op =)".  The
   223     transfer method uses these lemmas to infer transfer rules for
   224     non-polymorphic constants on the fly.
   225 
   226   - transfer_prover method: Assists with proving a transfer rule for a
   227     new constant, provided the constant is defined in terms of other
   228     constants that already have transfer rules. It should be applied
   229     after unfolding the constant definitions.
   230 
   231   - HOL/ex/Transfer_Int_Nat.thy: Example theory demonstrating transfer
   232     from type nat to type int.
   233 
   234 * Lifting: New package intended to generalize the quotient_definition
   235 facility of the Quotient package; designed to work with Transfer.
   236 
   237   - lift_definition command: Defines operations on an abstract type in
   238     terms of a corresponding operation on a representation
   239     type.  Example syntax:
   240 
   241     lift_definition dlist_insert :: "'a => 'a dlist => 'a dlist"
   242       is List.insert
   243 
   244     Users must discharge a respectfulness proof obligation when each
   245     constant is defined. (For a type copy, i.e. a typedef with UNIV,
   246     the proof is discharged automatically.) The obligation is
   247     presented in a user-friendly, readable form; a respectfulness
   248     theorem in the standard format and a transfer rule are generated
   249     by the package.
   250 
   251   - Integration with code_abstype: For typedefs (e.g. subtypes
   252     corresponding to a datatype invariant, such as dlist),
   253     lift_definition generates a code certificate theorem and sets up
   254     code generation for each constant.
   255 
   256   - setup_lifting command: Sets up the Lifting package to work with a
   257     user-defined type. The user must provide either a quotient theorem
   258     or a type_definition theorem.  The package configures transfer
   259     rules for equality and quantifiers on the type, and sets up the
   260     lift_definition command to work with the type.
   261 
   262   - Usage examples: See Quotient_Examples/Lift_DList.thy,
   263     Quotient_Examples/Lift_RBT.thy, Quotient_Examples/Lift_FSet.thy,
   264     Word/Word.thy and Library/Float.thy.
   265 
   266 * Quotient package:
   267 
   268   - The 'quotient_type' command now supports a 'morphisms' option with
   269     rep and abs functions, similar to typedef.
   270 
   271   - 'quotient_type' sets up new types to work with the Lifting and
   272     Transfer packages, as with 'setup_lifting'.
   273 
   274   - The 'quotient_definition' command now requires the user to prove a
   275     respectfulness property at the point where the constant is
   276     defined, similar to lift_definition; INCOMPATIBILITY.
   277 
   278   - Renamed predicate 'Quotient' to 'Quotient3', and renamed theorems
   279     accordingly, INCOMPATIBILITY.
   280 
   281 * New diagnostic command 'find_unused_assms' to find potentially
   282 superfluous assumptions in theorems using Quickcheck.
   283 
   284 * Quickcheck:
   285 
   286   - Quickcheck returns variable assignments as counterexamples, which
   287     allows to reveal the underspecification of functions under test.
   288     For example, refuting "hd xs = x", it presents the variable
   289     assignment xs = [] and x = a1 as a counterexample, assuming that
   290     any property is false whenever "hd []" occurs in it.
   291 
   292     These counterexample are marked as potentially spurious, as
   293     Quickcheck also returns "xs = []" as a counterexample to the
   294     obvious theorem "hd xs = hd xs".
   295 
   296     After finding a potentially spurious counterexample, Quickcheck
   297     continues searching for genuine ones.
   298 
   299     By default, Quickcheck shows potentially spurious and genuine
   300     counterexamples. The option "genuine_only" sets quickcheck to only
   301     show genuine counterexamples.
   302 
   303   - The command 'quickcheck_generator' creates random and exhaustive
   304     value generators for a given type and operations.
   305 
   306     It generates values by using the operations as if they were
   307     constructors of that type.
   308 
   309   - Support for multisets.
   310 
   311   - Added "use_subtype" options.
   312 
   313   - Added "quickcheck_locale" configuration to specify how to process
   314     conjectures in a locale context.
   315 
   316 * Nitpick: Fixed infinite loop caused by the 'peephole_optim' option
   317 and affecting 'rat' and 'real'.
   318 
   319 * Sledgehammer:
   320   - Integrated more tightly with SPASS, as described in the ITP 2012
   321     paper "More SPASS with Isabelle".
   322   - Made it try "smt" as a fallback if "metis" fails or times out.
   323   - Added support for the following provers: Alt-Ergo (via Why3 and
   324     TFF1), iProver, iProver-Eq.
   325   - Replaced remote E-SInE with remote Satallax in the default setup.
   326   - Sped up the minimizer.
   327   - Added "lam_trans", "uncurry_aliases", and "minimize" options.
   328   - Renamed "slicing" ("no_slicing") option to "slice" ("dont_slice").
   329   - Renamed "sound" option to "strict".
   330 
   331 * Metis: Added possibility to specify lambda translations scheme as a
   332 parenthesized argument (e.g., "by (metis (lifting) ...)").
   333 
   334 * SMT: Renamed "smt_fixed" option to "smt_read_only_certificates".
   335 
   336 * Command 'try0': Renamed from 'try_methods'. INCOMPATIBILITY.
   337 
   338 * New "case_product" attribute to generate a case rule doing multiple
   339 case distinctions at the same time.  E.g.
   340 
   341   list.exhaust [case_product nat.exhaust]
   342 
   343 produces a rule which can be used to perform case distinction on both
   344 a list and a nat.
   345 
   346 * New "eventually_elim" method as a generalized variant of the
   347 eventually_elim* rules.  Supports structured proofs.
   348 
   349 * Typedef with implicit set definition is considered legacy.  Use
   350 "typedef (open)" form instead, which will eventually become the
   351 default.
   352 
   353 * Record: code generation can be switched off manually with
   354 
   355   declare [[record_coden = false]]  -- "default true"
   356 
   357 * Datatype: type parameters allow explicit sort constraints.
   358 
   359 * Concrete syntax for case expressions includes constraints for source
   360 positions, and thus produces Prover IDE markup for its bindings.
   361 INCOMPATIBILITY for old-style syntax translations that augment the
   362 pattern notation; e.g. see src/HOL/HOLCF/One.thy for translations of
   363 one_case.
   364 
   365 * Clarified attribute "mono_set": pure declaration without modifying
   366 the result of the fact expression.
   367 
   368 * More default pred/set conversions on a couple of relation operations
   369 and predicates.  Added powers of predicate relations.  Consolidation
   370 of some relation theorems:
   371 
   372   converse_def ~> converse_unfold
   373   rel_comp_def ~> relcomp_unfold
   374   symp_def ~> (modified, use symp_def and sym_def instead)
   375   transp_def ~> transp_trans
   376   Domain_def ~> Domain_unfold
   377   Range_def ~> Domain_converse [symmetric]
   378 
   379 Generalized theorems INF_INT_eq, INF_INT_eq2, SUP_UN_eq, SUP_UN_eq2.
   380 
   381 See theory "Relation" for examples for making use of pred/set
   382 conversions by means of attributes "to_set" and "to_pred".
   383 
   384 INCOMPATIBILITY.
   385 
   386 * Renamed facts about the power operation on relations, i.e., relpow
   387 to match the constant's name:
   388 
   389   rel_pow_1 ~> relpow_1
   390   rel_pow_0_I ~> relpow_0_I
   391   rel_pow_Suc_I ~> relpow_Suc_I
   392   rel_pow_Suc_I2 ~> relpow_Suc_I2
   393   rel_pow_0_E ~> relpow_0_E
   394   rel_pow_Suc_E ~> relpow_Suc_E
   395   rel_pow_E ~> relpow_E
   396   rel_pow_Suc_D2 ~> relpow_Suc_D2
   397   rel_pow_Suc_E2 ~> relpow_Suc_E2
   398   rel_pow_Suc_D2' ~> relpow_Suc_D2'
   399   rel_pow_E2 ~> relpow_E2
   400   rel_pow_add ~> relpow_add
   401   rel_pow_commute ~> relpow
   402   rel_pow_empty ~> relpow_empty:
   403   rtrancl_imp_UN_rel_pow ~> rtrancl_imp_UN_relpow
   404   rel_pow_imp_rtrancl ~> relpow_imp_rtrancl
   405   rtrancl_is_UN_rel_pow ~> rtrancl_is_UN_relpow
   406   rtrancl_imp_rel_pow ~> rtrancl_imp_relpow
   407   rel_pow_fun_conv ~> relpow_fun_conv
   408   rel_pow_finite_bounded1 ~> relpow_finite_bounded1
   409   rel_pow_finite_bounded ~> relpow_finite_bounded
   410   rtrancl_finite_eq_rel_pow ~> rtrancl_finite_eq_relpow
   411   trancl_finite_eq_rel_pow ~> trancl_finite_eq_relpow
   412   single_valued_rel_pow ~> single_valued_relpow
   413 
   414 INCOMPATIBILITY.
   415 
   416 * Theory Relation: Consolidated constant name for relation composition
   417 and corresponding theorem names:
   418 
   419   - Renamed constant rel_comp to relcomp.
   420 
   421   - Dropped abbreviation pred_comp. Use relcompp instead.
   422 
   423   - Renamed theorems:
   424 
   425     rel_compI ~> relcompI
   426     rel_compEpair ~> relcompEpair
   427     rel_compE ~> relcompE
   428     pred_comp_rel_comp_eq ~> relcompp_relcomp_eq
   429     rel_comp_empty1 ~> relcomp_empty1
   430     rel_comp_mono ~> relcomp_mono
   431     rel_comp_subset_Sigma ~> relcomp_subset_Sigma
   432     rel_comp_distrib ~> relcomp_distrib
   433     rel_comp_distrib2 ~> relcomp_distrib2
   434     rel_comp_UNION_distrib ~> relcomp_UNION_distrib
   435     rel_comp_UNION_distrib2 ~> relcomp_UNION_distrib2
   436     single_valued_rel_comp ~> single_valued_relcomp
   437     rel_comp_def ~> relcomp_unfold
   438     converse_rel_comp ~> converse_relcomp
   439     pred_compI ~> relcomppI
   440     pred_compE ~> relcomppE
   441     pred_comp_bot1 ~> relcompp_bot1
   442     pred_comp_bot2 ~> relcompp_bot2
   443     transp_pred_comp_less_eq ~> transp_relcompp_less_eq
   444     pred_comp_mono ~> relcompp_mono
   445     pred_comp_distrib ~> relcompp_distrib
   446     pred_comp_distrib2 ~> relcompp_distrib2
   447     converse_pred_comp ~> converse_relcompp
   448 
   449     finite_rel_comp ~> finite_relcomp
   450 
   451     set_rel_comp ~> set_relcomp
   452 
   453 INCOMPATIBILITY.
   454 
   455 * Theory Divides: Discontinued redundant theorems about div and mod.
   456 INCOMPATIBILITY, use the corresponding generic theorems instead.
   457 
   458   DIVISION_BY_ZERO ~> div_by_0, mod_by_0
   459   zdiv_self ~> div_self
   460   zmod_self ~> mod_self
   461   zdiv_zero ~> div_0
   462   zmod_zero ~> mod_0
   463   zdiv_zmod_equality ~> div_mod_equality2
   464   zdiv_zmod_equality2 ~> div_mod_equality
   465   zmod_zdiv_trivial ~> mod_div_trivial
   466   zdiv_zminus_zminus ~> div_minus_minus
   467   zmod_zminus_zminus ~> mod_minus_minus
   468   zdiv_zminus2 ~> div_minus_right
   469   zmod_zminus2 ~> mod_minus_right
   470   zdiv_minus1_right ~> div_minus1_right
   471   zmod_minus1_right ~> mod_minus1_right
   472   zdvd_mult_div_cancel ~> dvd_mult_div_cancel
   473   zmod_zmult1_eq ~> mod_mult_right_eq
   474   zpower_zmod ~> power_mod
   475   zdvd_zmod ~> dvd_mod
   476   zdvd_zmod_imp_zdvd ~> dvd_mod_imp_dvd
   477   mod_mult_distrib ~> mult_mod_left
   478   mod_mult_distrib2 ~> mult_mod_right
   479 
   480 * Removed redundant theorems nat_mult_2 and nat_mult_2_right; use
   481 generic mult_2 and mult_2_right instead. INCOMPATIBILITY.
   482 
   483 * Finite_Set.fold now qualified.  INCOMPATIBILITY.
   484 
   485 * Consolidated theorem names concerning fold combinators:
   486 
   487   inf_INFI_fold_inf ~> inf_INF_fold_inf
   488   sup_SUPR_fold_sup ~> sup_SUP_fold_sup
   489   INFI_fold_inf ~> INF_fold_inf
   490   SUPR_fold_sup ~> SUP_fold_sup
   491   union_set ~> union_set_fold
   492   minus_set ~> minus_set_fold
   493   INFI_set_fold ~> INF_set_fold
   494   SUPR_set_fold ~> SUP_set_fold
   495   INF_code ~> INF_set_foldr
   496   SUP_code ~> SUP_set_foldr
   497   foldr.simps ~> foldr.simps (in point-free formulation)
   498   foldr_fold_rev ~> foldr_conv_fold
   499   foldl_fold ~> foldl_conv_fold
   500   foldr_foldr ~> foldr_conv_foldl
   501   foldl_foldr ~> foldl_conv_foldr
   502   fold_set_remdups ~> fold_set_fold_remdups
   503   fold_set ~> fold_set_fold
   504   fold1_set ~> fold1_set_fold
   505 
   506 INCOMPATIBILITY.
   507 
   508 * Dropped rarely useful theorems concerning fold combinators:
   509 foldl_apply, foldl_fun_comm, foldl_rev, fold_weak_invariant,
   510 rev_foldl_cons, fold_set_remdups, fold_set, fold_set1,
   511 concat_conv_foldl, foldl_weak_invariant, foldl_invariant,
   512 foldr_invariant, foldl_absorb0, foldl_foldr1_lemma, foldl_foldr1,
   513 listsum_conv_fold, listsum_foldl, sort_foldl_insort, foldl_assoc,
   514 foldr_conv_foldl, start_le_sum, elem_le_sum, sum_eq_0_conv.
   515 INCOMPATIBILITY.  For the common phrases "%xs. List.foldr plus xs 0"
   516 and "List.foldl plus 0", prefer "List.listsum".  Otherwise it can be
   517 useful to boil down "List.foldr" and "List.foldl" to "List.fold" by
   518 unfolding "foldr_conv_fold" and "foldl_conv_fold".
   519 
   520 * Dropped lemmas minus_set_foldr, union_set_foldr, union_coset_foldr,
   521 inter_coset_foldr, Inf_fin_set_foldr, Sup_fin_set_foldr,
   522 Min_fin_set_foldr, Max_fin_set_foldr, Inf_set_foldr, Sup_set_foldr,
   523 INF_set_foldr, SUP_set_foldr.  INCOMPATIBILITY.  Prefer corresponding
   524 lemmas over fold rather than foldr, or make use of lemmas
   525 fold_conv_foldr and fold_rev.
   526 
   527 * Congruence rules Option.map_cong and Option.bind_cong for recursion
   528 through option types.
   529 
   530 * "Transitive_Closure.ntrancl": bounded transitive closure on
   531 relations.
   532 
   533 * Constant "Set.not_member" now qualified.  INCOMPATIBILITY.
   534 
   535 * Theory Int: Discontinued many legacy theorems specific to type int.
   536 INCOMPATIBILITY, use the corresponding generic theorems instead.
   537 
   538   zminus_zminus ~> minus_minus
   539   zminus_0 ~> minus_zero
   540   zminus_zadd_distrib ~> minus_add_distrib
   541   zadd_commute ~> add_commute
   542   zadd_assoc ~> add_assoc
   543   zadd_left_commute ~> add_left_commute
   544   zadd_ac ~> add_ac
   545   zmult_ac ~> mult_ac
   546   zadd_0 ~> add_0_left
   547   zadd_0_right ~> add_0_right
   548   zadd_zminus_inverse2 ~> left_minus
   549   zmult_zminus ~> mult_minus_left
   550   zmult_commute ~> mult_commute
   551   zmult_assoc ~> mult_assoc
   552   zadd_zmult_distrib ~> left_distrib
   553   zadd_zmult_distrib2 ~> right_distrib
   554   zdiff_zmult_distrib ~> left_diff_distrib
   555   zdiff_zmult_distrib2 ~> right_diff_distrib
   556   zmult_1 ~> mult_1_left
   557   zmult_1_right ~> mult_1_right
   558   zle_refl ~> order_refl
   559   zle_trans ~> order_trans
   560   zle_antisym ~> order_antisym
   561   zle_linear ~> linorder_linear
   562   zless_linear ~> linorder_less_linear
   563   zadd_left_mono ~> add_left_mono
   564   zadd_strict_right_mono ~> add_strict_right_mono
   565   zadd_zless_mono ~> add_less_le_mono
   566   int_0_less_1 ~> zero_less_one
   567   int_0_neq_1 ~> zero_neq_one
   568   zless_le ~> less_le
   569   zpower_zadd_distrib ~> power_add
   570   zero_less_zpower_abs_iff ~> zero_less_power_abs_iff
   571   zero_le_zpower_abs ~> zero_le_power_abs
   572 
   573 * Theory Deriv: Renamed
   574 
   575   DERIV_nonneg_imp_nonincreasing ~> DERIV_nonneg_imp_nondecreasing
   576 
   577 * Theory Library/Multiset: Improved code generation of multisets.
   578 
   579 * Theory HOL/Library/Set_Algebras: Addition and multiplication on sets
   580 are expressed via type classes again. The special syntax
   581 \<oplus>/\<otimes> has been replaced by plain +/*. Removed constant
   582 setsum_set, which is now subsumed by Big_Operators.setsum.
   583 INCOMPATIBILITY.
   584 
   585 * Theory HOL/Library/Diagonalize has been removed. INCOMPATIBILITY,
   586 use theory HOL/Library/Nat_Bijection instead.
   587 
   588 * Theory HOL/Library/RBT_Impl: Backing implementation of red-black
   589 trees is now inside a type class context.  Names of affected
   590 operations and lemmas have been prefixed by rbt_.  INCOMPATIBILITY for
   591 theories working directly with raw red-black trees, adapt the names as
   592 follows:
   593 
   594   Operations:
   595   bulkload -> rbt_bulkload
   596   del_from_left -> rbt_del_from_left
   597   del_from_right -> rbt_del_from_right
   598   del -> rbt_del
   599   delete -> rbt_delete
   600   ins -> rbt_ins
   601   insert -> rbt_insert
   602   insertw -> rbt_insert_with
   603   insert_with_key -> rbt_insert_with_key
   604   map_entry -> rbt_map_entry
   605   lookup -> rbt_lookup
   606   sorted -> rbt_sorted
   607   tree_greater -> rbt_greater
   608   tree_less -> rbt_less
   609   tree_less_symbol -> rbt_less_symbol
   610   union -> rbt_union
   611   union_with -> rbt_union_with
   612   union_with_key -> rbt_union_with_key
   613 
   614   Lemmas:
   615   balance_left_sorted -> balance_left_rbt_sorted
   616   balance_left_tree_greater -> balance_left_rbt_greater
   617   balance_left_tree_less -> balance_left_rbt_less
   618   balance_right_sorted -> balance_right_rbt_sorted
   619   balance_right_tree_greater -> balance_right_rbt_greater
   620   balance_right_tree_less -> balance_right_rbt_less
   621   balance_sorted -> balance_rbt_sorted
   622   balance_tree_greater -> balance_rbt_greater
   623   balance_tree_less -> balance_rbt_less
   624   bulkload_is_rbt -> rbt_bulkload_is_rbt
   625   combine_sorted -> combine_rbt_sorted
   626   combine_tree_greater -> combine_rbt_greater
   627   combine_tree_less -> combine_rbt_less
   628   delete_in_tree -> rbt_delete_in_tree
   629   delete_is_rbt -> rbt_delete_is_rbt
   630   del_from_left_tree_greater -> rbt_del_from_left_rbt_greater
   631   del_from_left_tree_less -> rbt_del_from_left_rbt_less
   632   del_from_right_tree_greater -> rbt_del_from_right_rbt_greater
   633   del_from_right_tree_less -> rbt_del_from_right_rbt_less
   634   del_in_tree -> rbt_del_in_tree
   635   del_inv1_inv2 -> rbt_del_inv1_inv2
   636   del_sorted -> rbt_del_rbt_sorted
   637   del_tree_greater -> rbt_del_rbt_greater
   638   del_tree_less -> rbt_del_rbt_less
   639   dom_lookup_Branch -> dom_rbt_lookup_Branch
   640   entries_lookup -> entries_rbt_lookup
   641   finite_dom_lookup -> finite_dom_rbt_lookup
   642   insert_sorted -> rbt_insert_rbt_sorted
   643   insertw_is_rbt -> rbt_insertw_is_rbt
   644   insertwk_is_rbt -> rbt_insertwk_is_rbt
   645   insertwk_sorted -> rbt_insertwk_rbt_sorted
   646   insertw_sorted -> rbt_insertw_rbt_sorted
   647   ins_sorted -> ins_rbt_sorted
   648   ins_tree_greater -> ins_rbt_greater
   649   ins_tree_less -> ins_rbt_less
   650   is_rbt_sorted -> is_rbt_rbt_sorted
   651   lookup_balance -> rbt_lookup_balance
   652   lookup_bulkload -> rbt_lookup_rbt_bulkload
   653   lookup_delete -> rbt_lookup_rbt_delete
   654   lookup_Empty -> rbt_lookup_Empty
   655   lookup_from_in_tree -> rbt_lookup_from_in_tree
   656   lookup_in_tree -> rbt_lookup_in_tree
   657   lookup_ins -> rbt_lookup_ins
   658   lookup_insert -> rbt_lookup_rbt_insert
   659   lookup_insertw -> rbt_lookup_rbt_insertw
   660   lookup_insertwk -> rbt_lookup_rbt_insertwk
   661   lookup_keys -> rbt_lookup_keys
   662   lookup_map -> rbt_lookup_map
   663   lookup_map_entry -> rbt_lookup_rbt_map_entry
   664   lookup_tree_greater -> rbt_lookup_rbt_greater
   665   lookup_tree_less -> rbt_lookup_rbt_less
   666   lookup_union -> rbt_lookup_rbt_union
   667   map_entry_color_of -> rbt_map_entry_color_of
   668   map_entry_inv1 -> rbt_map_entry_inv1
   669   map_entry_inv2 -> rbt_map_entry_inv2
   670   map_entry_is_rbt -> rbt_map_entry_is_rbt
   671   map_entry_sorted -> rbt_map_entry_rbt_sorted
   672   map_entry_tree_greater -> rbt_map_entry_rbt_greater
   673   map_entry_tree_less -> rbt_map_entry_rbt_less
   674   map_tree_greater -> map_rbt_greater
   675   map_tree_less -> map_rbt_less
   676   map_sorted -> map_rbt_sorted
   677   paint_sorted -> paint_rbt_sorted
   678   paint_lookup -> paint_rbt_lookup
   679   paint_tree_greater -> paint_rbt_greater
   680   paint_tree_less -> paint_rbt_less
   681   sorted_entries -> rbt_sorted_entries
   682   tree_greater_eq_trans -> rbt_greater_eq_trans
   683   tree_greater_nit -> rbt_greater_nit
   684   tree_greater_prop -> rbt_greater_prop
   685   tree_greater_simps -> rbt_greater_simps
   686   tree_greater_trans -> rbt_greater_trans
   687   tree_less_eq_trans -> rbt_less_eq_trans
   688   tree_less_nit -> rbt_less_nit
   689   tree_less_prop -> rbt_less_prop
   690   tree_less_simps -> rbt_less_simps
   691   tree_less_trans -> rbt_less_trans
   692   tree_ord_props -> rbt_ord_props
   693   union_Branch -> rbt_union_Branch
   694   union_is_rbt -> rbt_union_is_rbt
   695   unionw_is_rbt -> rbt_unionw_is_rbt
   696   unionwk_is_rbt -> rbt_unionwk_is_rbt
   697   unionwk_sorted -> rbt_unionwk_rbt_sorted
   698 
   699 * Theory HOL/Library/Float: Floating point numbers are now defined as
   700 a subset of the real numbers.  All operations are defined using the
   701 lifing-framework and proofs use the transfer method.  INCOMPATIBILITY.
   702 
   703   Changed Operations:
   704   float_abs -> abs
   705   float_nprt -> nprt
   706   float_pprt -> pprt
   707   pow2 -> use powr
   708   round_down -> float_round_down
   709   round_up -> float_round_up
   710   scale -> exponent
   711 
   712   Removed Operations:
   713   ceiling_fl, lb_mult, lb_mod, ub_mult, ub_mod
   714 
   715   Renamed Lemmas:
   716   abs_float_def -> Float.compute_float_abs
   717   bitlen_ge0 -> bitlen_nonneg
   718   bitlen.simps -> Float.compute_bitlen
   719   float_components -> Float_mantissa_exponent
   720   float_divl.simps -> Float.compute_float_divl
   721   float_divr.simps -> Float.compute_float_divr
   722   float_eq_odd -> mult_powr_eq_mult_powr_iff
   723   float_power -> real_of_float_power
   724   lapprox_posrat_def -> Float.compute_lapprox_posrat
   725   lapprox_rat.simps -> Float.compute_lapprox_rat
   726   le_float_def' -> Float.compute_float_le
   727   le_float_def -> less_eq_float.rep_eq
   728   less_float_def' -> Float.compute_float_less
   729   less_float_def -> less_float.rep_eq
   730   normfloat_def -> Float.compute_normfloat
   731   normfloat_imp_odd_or_zero -> mantissa_not_dvd and mantissa_noteq_0
   732   normfloat -> normfloat_def
   733   normfloat_unique -> use normfloat_def
   734   number_of_float_Float -> Float.compute_float_numeral, Float.compute_float_neg_numeral
   735   one_float_def -> Float.compute_float_one
   736   plus_float_def -> Float.compute_float_plus
   737   rapprox_posrat_def -> Float.compute_rapprox_posrat
   738   rapprox_rat.simps -> Float.compute_rapprox_rat
   739   real_of_float_0 -> zero_float.rep_eq
   740   real_of_float_1 -> one_float.rep_eq
   741   real_of_float_abs -> abs_float.rep_eq
   742   real_of_float_add -> plus_float.rep_eq
   743   real_of_float_minus -> uminus_float.rep_eq
   744   real_of_float_mult -> times_float.rep_eq
   745   real_of_float_simp -> Float.rep_eq
   746   real_of_float_sub -> minus_float.rep_eq
   747   round_down.simps -> Float.compute_float_round_down
   748   round_up.simps -> Float.compute_float_round_up
   749   times_float_def -> Float.compute_float_times
   750   uminus_float_def -> Float.compute_float_uminus
   751   zero_float_def -> Float.compute_float_zero
   752 
   753   Lemmas not necessary anymore, use the transfer method:
   754   bitlen_B0, bitlen_B1, bitlen_ge1, bitlen_Min, bitlen_Pls, float_divl,
   755   float_divr, float_le_simp, float_less1_mantissa_bound,
   756   float_less_simp, float_less_zero, float_le_zero,
   757   float_pos_less1_e_neg, float_pos_m_pos, float_split, float_split2,
   758   floor_pos_exp, lapprox_posrat, lapprox_posrat_bottom, lapprox_rat,
   759   lapprox_rat_bottom, normalized_float, rapprox_posrat,
   760   rapprox_posrat_le1, rapprox_rat, real_of_float_ge0_exp,
   761   real_of_float_neg_exp, real_of_float_nge0_exp, round_down floor_fl,
   762   round_up, zero_le_float, zero_less_float
   763 
   764 * New theory HOL/Library/DAList provides an abstract type for
   765 association lists with distinct keys.
   766 
   767 * Session HOL/IMP: Added new theory of abstract interpretation of
   768 annotated commands.
   769 
   770 * Session HOL-Import: Re-implementation from scratch is faster,
   771 simpler, and more scalable.  Requires a proof bundle, which is
   772 available as an external component.  Discontinued old (and mostly
   773 dead) Importer for HOL4 and HOL Light.  INCOMPATIBILITY.
   774 
   775 * Session HOL-Word: Discontinued many redundant theorems specific to
   776 type 'a word. INCOMPATIBILITY, use the corresponding generic theorems
   777 instead.
   778 
   779   word_sub_alt ~> word_sub_wi
   780   word_add_alt ~> word_add_def
   781   word_mult_alt ~> word_mult_def
   782   word_minus_alt ~> word_minus_def
   783   word_0_alt ~> word_0_wi
   784   word_1_alt ~> word_1_wi
   785   word_add_0 ~> add_0_left
   786   word_add_0_right ~> add_0_right
   787   word_mult_1 ~> mult_1_left
   788   word_mult_1_right ~> mult_1_right
   789   word_add_commute ~> add_commute
   790   word_add_assoc ~> add_assoc
   791   word_add_left_commute ~> add_left_commute
   792   word_mult_commute ~> mult_commute
   793   word_mult_assoc ~> mult_assoc
   794   word_mult_left_commute ~> mult_left_commute
   795   word_left_distrib ~> left_distrib
   796   word_right_distrib ~> right_distrib
   797   word_left_minus ~> left_minus
   798   word_diff_0_right ~> diff_0_right
   799   word_diff_self ~> diff_self
   800   word_sub_def ~> diff_minus
   801   word_diff_minus ~> diff_minus
   802   word_add_ac ~> add_ac
   803   word_mult_ac ~> mult_ac
   804   word_plus_ac0 ~> add_0_left add_0_right add_ac
   805   word_times_ac1 ~> mult_1_left mult_1_right mult_ac
   806   word_order_trans ~> order_trans
   807   word_order_refl ~> order_refl
   808   word_order_antisym ~> order_antisym
   809   word_order_linear ~> linorder_linear
   810   lenw1_zero_neq_one ~> zero_neq_one
   811   word_number_of_eq ~> number_of_eq
   812   word_of_int_add_hom ~> wi_hom_add
   813   word_of_int_sub_hom ~> wi_hom_sub
   814   word_of_int_mult_hom ~> wi_hom_mult
   815   word_of_int_minus_hom ~> wi_hom_neg
   816   word_of_int_succ_hom ~> wi_hom_succ
   817   word_of_int_pred_hom ~> wi_hom_pred
   818   word_of_int_0_hom ~> word_0_wi
   819   word_of_int_1_hom ~> word_1_wi
   820 
   821 * Session HOL-Word: New proof method "word_bitwise" for splitting
   822 machine word equalities and inequalities into logical circuits,
   823 defined in HOL/Word/WordBitwise.thy.  Supports addition, subtraction,
   824 multiplication, shifting by constants, bitwise operators and numeric
   825 constants.  Requires fixed-length word types, not 'a word.  Solves
   826 many standard word identities outright and converts more into first
   827 order problems amenable to blast or similar.  See also examples in
   828 HOL/Word/Examples/WordExamples.thy.
   829 
   830 * Session HOL-Probability: Introduced the type "'a measure" to
   831 represent measures, this replaces the records 'a algebra and 'a
   832 measure_space.  The locales based on subset_class now have two
   833 locale-parameters the space \<Omega> and the set of measurable sets M.
   834 The product of probability spaces uses now the same constant as the
   835 finite product of sigma-finite measure spaces "PiM :: ('i => 'a)
   836 measure".  Most constants are defined now outside of locales and gain
   837 an additional parameter, like null_sets, almost_eventually or \<mu>'.
   838 Measure space constructions for distributions and densities now got
   839 their own constants distr and density.  Instead of using locales to
   840 describe measure spaces with a finite space, the measure count_space
   841 and point_measure is introduced.  INCOMPATIBILITY.
   842 
   843   Renamed constants:
   844   measure -> emeasure
   845   finite_measure.\<mu>' -> measure
   846   product_algebra_generator -> prod_algebra
   847   product_prob_space.emb -> prod_emb
   848   product_prob_space.infprod_algebra -> PiM
   849 
   850   Removed locales:
   851   completeable_measure_space
   852   finite_measure_space
   853   finite_prob_space
   854   finite_product_finite_prob_space
   855   finite_product_sigma_algebra
   856   finite_sigma_algebra
   857   measure_space
   858   pair_finite_prob_space
   859   pair_finite_sigma_algebra
   860   pair_finite_space
   861   pair_sigma_algebra
   862   product_sigma_algebra
   863 
   864   Removed constants:
   865   conditional_space
   866   distribution -> use distr measure, or distributed predicate
   867   image_space
   868   joint_distribution -> use distr measure, or distributed predicate
   869   pair_measure_generator
   870   product_prob_space.infprod_algebra -> use PiM
   871   subvimage
   872 
   873   Replacement theorems:
   874   finite_additivity_sufficient -> ring_of_sets.countably_additiveI_finite
   875   finite_measure.empty_measure -> measure_empty
   876   finite_measure.finite_continuity_from_above -> finite_measure.finite_Lim_measure_decseq
   877   finite_measure.finite_continuity_from_below -> finite_measure.finite_Lim_measure_incseq
   878   finite_measure.finite_measure_countably_subadditive -> finite_measure.finite_measure_subadditive_countably
   879   finite_measure.finite_measure_eq -> finite_measure.emeasure_eq_measure
   880   finite_measure.finite_measure -> finite_measure.emeasure_finite
   881   finite_measure.finite_measure_finite_singleton -> finite_measure.finite_measure_eq_setsum_singleton
   882   finite_measure.positive_measure' -> measure_nonneg
   883   finite_measure.real_measure -> finite_measure.emeasure_real
   884   finite_product_prob_space.finite_measure_times -> finite_product_prob_space.finite_measure_PiM_emb
   885   finite_product_sigma_algebra.in_P -> sets_PiM_I_finite
   886   finite_product_sigma_algebra.P_empty -> space_PiM_empty, sets_PiM_empty
   887   information_space.conditional_entropy_eq -> information_space.conditional_entropy_simple_distributed
   888   information_space.conditional_entropy_positive -> information_space.conditional_entropy_nonneg_simple
   889   information_space.conditional_mutual_information_eq_mutual_information -> information_space.conditional_mutual_information_eq_mutual_information_simple
   890   information_space.conditional_mutual_information_generic_positive -> information_space.conditional_mutual_information_nonneg_simple
   891   information_space.conditional_mutual_information_positive -> information_space.conditional_mutual_information_nonneg_simple
   892   information_space.entropy_commute -> information_space.entropy_commute_simple
   893   information_space.entropy_eq -> information_space.entropy_simple_distributed
   894   information_space.entropy_generic_eq -> information_space.entropy_simple_distributed
   895   information_space.entropy_positive -> information_space.entropy_nonneg_simple
   896   information_space.entropy_uniform_max -> information_space.entropy_uniform
   897   information_space.KL_eq_0_imp -> information_space.KL_eq_0_iff_eq
   898   information_space.KL_eq_0 -> information_space.KL_same_eq_0
   899   information_space.KL_ge_0 -> information_space.KL_nonneg
   900   information_space.mutual_information_eq -> information_space.mutual_information_simple_distributed
   901   information_space.mutual_information_positive -> information_space.mutual_information_nonneg_simple
   902   Int_stable_cuboids -> Int_stable_atLeastAtMost
   903   Int_stable_product_algebra_generator -> positive_integral
   904   measure_preserving -> equality "distr M N f = N" "f : measurable M N"
   905   measure_space.additive -> emeasure_additive
   906   measure_space.AE_iff_null_set -> AE_iff_null
   907   measure_space.almost_everywhere_def -> eventually_ae_filter
   908   measure_space.almost_everywhere_vimage -> AE_distrD
   909   measure_space.continuity_from_above -> INF_emeasure_decseq
   910   measure_space.continuity_from_above_Lim -> Lim_emeasure_decseq
   911   measure_space.continuity_from_below_Lim -> Lim_emeasure_incseq
   912   measure_space.continuity_from_below -> SUP_emeasure_incseq
   913   measure_space_density -> emeasure_density
   914   measure_space.density_is_absolutely_continuous -> absolutely_continuousI_density
   915   measure_space.integrable_vimage -> integrable_distr
   916   measure_space.integral_translated_density -> integral_density
   917   measure_space.integral_vimage -> integral_distr
   918   measure_space.measure_additive -> plus_emeasure
   919   measure_space.measure_compl -> emeasure_compl
   920   measure_space.measure_countable_increasing -> emeasure_countable_increasing
   921   measure_space.measure_countably_subadditive -> emeasure_subadditive_countably
   922   measure_space.measure_decseq -> decseq_emeasure
   923   measure_space.measure_Diff -> emeasure_Diff
   924   measure_space.measure_Diff_null_set -> emeasure_Diff_null_set
   925   measure_space.measure_eq_0 -> emeasure_eq_0
   926   measure_space.measure_finitely_subadditive -> emeasure_subadditive_finite
   927   measure_space.measure_finite_singleton -> emeasure_eq_setsum_singleton
   928   measure_space.measure_incseq -> incseq_emeasure
   929   measure_space.measure_insert -> emeasure_insert
   930   measure_space.measure_mono -> emeasure_mono
   931   measure_space.measure_not_negative -> emeasure_not_MInf
   932   measure_space.measure_preserving_Int_stable -> measure_eqI_generator_eq
   933   measure_space.measure_setsum -> setsum_emeasure
   934   measure_space.measure_setsum_split -> setsum_emeasure_cover
   935   measure_space.measure_space_vimage -> emeasure_distr
   936   measure_space.measure_subadditive_finite -> emeasure_subadditive_finite
   937   measure_space.measure_subadditive -> subadditive
   938   measure_space.measure_top -> emeasure_space
   939   measure_space.measure_UN_eq_0 -> emeasure_UN_eq_0
   940   measure_space.measure_Un_null_set -> emeasure_Un_null_set
   941   measure_space.positive_integral_translated_density -> positive_integral_density
   942   measure_space.positive_integral_vimage -> positive_integral_distr
   943   measure_space.real_continuity_from_above -> Lim_measure_decseq
   944   measure_space.real_continuity_from_below -> Lim_measure_incseq
   945   measure_space.real_measure_countably_subadditive -> measure_subadditive_countably
   946   measure_space.real_measure_Diff -> measure_Diff
   947   measure_space.real_measure_finite_Union -> measure_finite_Union
   948   measure_space.real_measure_setsum_singleton -> measure_eq_setsum_singleton
   949   measure_space.real_measure_subadditive -> measure_subadditive
   950   measure_space.real_measure_Union -> measure_Union
   951   measure_space.real_measure_UNION -> measure_UNION
   952   measure_space.simple_function_vimage -> simple_function_comp
   953   measure_space.simple_integral_vimage -> simple_integral_distr
   954   measure_space.simple_integral_vimage -> simple_integral_distr
   955   measure_unique_Int_stable -> measure_eqI_generator_eq
   956   measure_unique_Int_stable_vimage -> measure_eqI_generator_eq
   957   pair_sigma_algebra.measurable_cut_fst -> sets_Pair1
   958   pair_sigma_algebra.measurable_cut_snd -> sets_Pair2
   959   pair_sigma_algebra.measurable_pair_image_fst -> measurable_Pair1
   960   pair_sigma_algebra.measurable_pair_image_snd -> measurable_Pair2
   961   pair_sigma_algebra.measurable_product_swap -> measurable_pair_swap_iff
   962   pair_sigma_algebra.pair_sigma_algebra_measurable -> measurable_pair_swap
   963   pair_sigma_algebra.pair_sigma_algebra_swap_measurable -> measurable_pair_swap'
   964   pair_sigma_algebra.sets_swap -> sets_pair_swap
   965   pair_sigma_finite.measure_cut_measurable_fst -> pair_sigma_finite.measurable_emeasure_Pair1
   966   pair_sigma_finite.measure_cut_measurable_snd -> pair_sigma_finite.measurable_emeasure_Pair2
   967   pair_sigma_finite.measure_preserving_swap -> pair_sigma_finite.distr_pair_swap
   968   pair_sigma_finite.pair_measure_alt2 -> pair_sigma_finite.emeasure_pair_measure_alt2
   969   pair_sigma_finite.pair_measure_alt -> pair_sigma_finite.emeasure_pair_measure_alt
   970   pair_sigma_finite.pair_measure_times -> pair_sigma_finite.emeasure_pair_measure_Times
   971   prob_space.indep_distribution_eq_measure -> prob_space.indep_vars_iff_distr_eq_PiM
   972   prob_space.indep_var_distributionD -> prob_space.indep_var_distribution_eq
   973   prob_space.measure_space_1 -> prob_space.emeasure_space_1
   974   prob_space.prob_space_vimage -> prob_space_distr
   975   prob_space.random_variable_restrict -> measurable_restrict
   976   prob_space_unique_Int_stable -> measure_eqI_prob_space
   977   product_algebraE -> prod_algebraE_all
   978   product_algebra_generator_der -> prod_algebra_eq_finite
   979   product_algebra_generator_into_space -> prod_algebra_sets_into_space
   980   product_algebraI -> sets_PiM_I_finite
   981   product_measure_exists -> product_sigma_finite.sigma_finite
   982   product_prob_space.finite_index_eq_finite_product -> product_prob_space.sets_PiM_generator
   983   product_prob_space.finite_measure_infprod_emb_Pi -> product_prob_space.measure_PiM_emb
   984   product_prob_space.infprod_spec -> product_prob_space.emeasure_PiM_emb_not_empty
   985   product_prob_space.measurable_component -> measurable_component_singleton
   986   product_prob_space.measurable_emb -> measurable_prod_emb
   987   product_prob_space.measurable_into_infprod_algebra -> measurable_PiM_single
   988   product_prob_space.measurable_singleton_infprod -> measurable_component_singleton
   989   product_prob_space.measure_emb -> emeasure_prod_emb
   990   product_prob_space.measure_preserving_restrict -> product_prob_space.distr_restrict
   991   product_sigma_algebra.product_algebra_into_space -> space_closed
   992   product_sigma_finite.measure_fold -> product_sigma_finite.distr_merge
   993   product_sigma_finite.measure_preserving_component_singelton -> product_sigma_finite.distr_singleton
   994   product_sigma_finite.measure_preserving_merge -> product_sigma_finite.distr_merge
   995   sequence_space.measure_infprod -> sequence_space.measure_PiM_countable
   996   sets_product_algebra -> sets_PiM
   997   sigma_algebra.measurable_sigma -> measurable_measure_of
   998   sigma_finite_measure.disjoint_sigma_finite -> sigma_finite_disjoint
   999   sigma_finite_measure.RN_deriv_vimage -> sigma_finite_measure.RN_deriv_distr
  1000   sigma_product_algebra_sigma_eq -> sigma_prod_algebra_sigma_eq
  1001   space_product_algebra -> space_PiM
  1002 
  1003 * Session HOL-TPTP: support to parse and import TPTP problems (all
  1004 languages) into Isabelle/HOL.
  1005 
  1006 
  1007 *** FOL ***
  1008 
  1009 * New "case_product" attribute (see HOL).
  1010 
  1011 
  1012 *** ZF ***
  1013 
  1014 * Greater support for structured proofs involving induction or case
  1015 analysis.
  1016 
  1017 * Much greater use of mathematical symbols.
  1018 
  1019 * Removal of many ML theorem bindings.  INCOMPATIBILITY.
  1020 
  1021 
  1022 *** ML ***
  1023 
  1024 * Antiquotation @{keyword "name"} produces a parser for outer syntax
  1025 from a minor keyword introduced via theory header declaration.
  1026 
  1027 * Antiquotation @{command_spec "name"} produces the
  1028 Outer_Syntax.command_spec from a major keyword introduced via theory
  1029 header declaration; it can be passed to Outer_Syntax.command etc.
  1030 
  1031 * Local_Theory.define no longer hard-wires default theorem name
  1032 "foo_def", but retains the binding as given.  If that is Binding.empty
  1033 / Attrib.empty_binding, the result is not registered as user-level
  1034 fact.  The Local_Theory.define_internal variant allows to specify a
  1035 non-empty name (used for the foundation in the background theory),
  1036 while omitting the fact binding in the user-context.  Potential
  1037 INCOMPATIBILITY for derived definitional packages: need to specify
  1038 naming policy for primitive definitions more explicitly.
  1039 
  1040 * Renamed Thm.capply to Thm.apply, and Thm.cabs to Thm.lambda in
  1041 conformance with similar operations in structure Term and Logic.
  1042 
  1043 * Antiquotation @{attributes [...]} embeds attribute source
  1044 representation into the ML text, which is particularly useful with
  1045 declarations like Local_Theory.note.
  1046 
  1047 * Structure Proof_Context follows standard naming scheme.  Old
  1048 ProofContext has been discontinued.  INCOMPATIBILITY.
  1049 
  1050 * Refined Local_Theory.declaration {syntax, pervasive}, with subtle
  1051 change of semantics: update is applied to auxiliary local theory
  1052 context as well.
  1053 
  1054 * Modernized some old-style infix operations:
  1055 
  1056   addeqcongs    ~> Simplifier.add_eqcong
  1057   deleqcongs    ~> Simplifier.del_eqcong
  1058   addcongs      ~> Simplifier.add_cong
  1059   delcongs      ~> Simplifier.del_cong
  1060   setmksimps    ~> Simplifier.set_mksimps
  1061   setmkcong     ~> Simplifier.set_mkcong
  1062   setmksym      ~> Simplifier.set_mksym
  1063   setmkeqTrue   ~> Simplifier.set_mkeqTrue
  1064   settermless   ~> Simplifier.set_termless
  1065   setsubgoaler  ~> Simplifier.set_subgoaler
  1066   addsplits     ~> Splitter.add_split
  1067   delsplits     ~> Splitter.del_split
  1068 
  1069 
  1070 *** System ***
  1071 
  1072 * USER_HOME settings variable points to cross-platform user home
  1073 directory, which coincides with HOME on POSIX systems only.  Likewise,
  1074 the Isabelle path specification "~" now expands to $USER_HOME, instead
  1075 of former $HOME.  A different default for USER_HOME may be set
  1076 explicitly in shell environment, before Isabelle settings are
  1077 evaluated.  Minor INCOMPATIBILITY: need to adapt Isabelle path where
  1078 the generic user home was intended.
  1079 
  1080 * ISABELLE_HOME_WINDOWS refers to ISABELLE_HOME in windows file name
  1081 notation, which is useful for the jEdit file browser, for example.
  1082 
  1083 * ISABELLE_JDK_HOME settings variable points to JDK with javac and jar
  1084 (not just JRE).
  1085 
  1086 
  1087 
  1088 New in Isabelle2011-1 (October 2011)
  1089 ------------------------------------
  1090 
  1091 *** General ***
  1092 
  1093 * Improved Isabelle/jEdit Prover IDE (PIDE), which can be invoked as
  1094 "isabelle jedit" or "ISABELLE_HOME/Isabelle" on the command line.
  1095 
  1096   - Management of multiple theory files directly from the editor
  1097     buffer store -- bypassing the file-system (no requirement to save
  1098     files for checking).
  1099 
  1100   - Markup of formal entities within the text buffer, with semantic
  1101     highlighting, tooltips and hyperlinks to jump to defining source
  1102     positions.
  1103 
  1104   - Improved text rendering, with sub/superscripts in the source
  1105     buffer (including support for copy/paste wrt. output panel, HTML
  1106     theory output and other non-Isabelle text boxes).
  1107 
  1108   - Refined scheduling of proof checking and printing of results,
  1109     based on interactive editor view.  (Note: jEdit folding and
  1110     narrowing allows to restrict buffer perspectives explicitly.)
  1111 
  1112   - Reduced CPU performance requirements, usable on machines with few
  1113     cores.
  1114 
  1115   - Reduced memory requirements due to pruning of unused document
  1116     versions (garbage collection).
  1117 
  1118 See also ~~/src/Tools/jEdit/README.html for further information,
  1119 including some remaining limitations.
  1120 
  1121 * Theory loader: source files are exclusively located via the master
  1122 directory of each theory node (where the .thy file itself resides).
  1123 The global load path (such as src/HOL/Library) has been discontinued.
  1124 Note that the path element ~~ may be used to reference theories in the
  1125 Isabelle home folder -- for instance, "~~/src/HOL/Library/FuncSet".
  1126 INCOMPATIBILITY.
  1127 
  1128 * Theory loader: source files are identified by content via SHA1
  1129 digests.  Discontinued former path/modtime identification and optional
  1130 ISABELLE_FILE_IDENT plugin scripts.
  1131 
  1132 * Parallelization of nested Isar proofs is subject to
  1133 Goal.parallel_proofs_threshold (default 100).  See also isabelle
  1134 usedir option -Q.
  1135 
  1136 * Name space: former unsynchronized references are now proper
  1137 configuration options, with more conventional names:
  1138 
  1139   long_names   ~> names_long
  1140   short_names  ~> names_short
  1141   unique_names ~> names_unique
  1142 
  1143 Minor INCOMPATIBILITY, need to declare options in context like this:
  1144 
  1145   declare [[names_unique = false]]
  1146 
  1147 * Literal facts `prop` may contain dummy patterns, e.g. `_ = _`.  Note
  1148 that the result needs to be unique, which means fact specifications
  1149 may have to be refined after enriching a proof context.
  1150 
  1151 * Attribute "case_names" has been refined: the assumptions in each case
  1152 can be named now by following the case name with [name1 name2 ...].
  1153 
  1154 * Isabelle/Isar reference manual has been updated and extended:
  1155   - "Synopsis" provides a catalog of main Isar language concepts.
  1156   - Formal references in syntax diagrams, via @{rail} antiquotation.
  1157   - Updated material from classic "ref" manual, notably about
  1158     "Classical Reasoner".
  1159 
  1160 
  1161 *** HOL ***
  1162 
  1163 * Class bot and top require underlying partial order rather than
  1164 preorder: uniqueness of bot and top is guaranteed.  INCOMPATIBILITY.
  1165 
  1166 * Class complete_lattice: generalized a couple of lemmas from sets;
  1167 generalized theorems INF_cong and SUP_cong.  New type classes for
  1168 complete boolean algebras and complete linear orders.  Lemmas
  1169 Inf_less_iff, less_Sup_iff, INF_less_iff, less_SUP_iff now reside in
  1170 class complete_linorder.
  1171 
  1172 Changed proposition of lemmas Inf_bool_def, Sup_bool_def, Inf_fun_def,
  1173 Sup_fun_def, Inf_apply, Sup_apply.
  1174 
  1175 Removed redundant lemmas (the right hand side gives hints how to
  1176 replace them for (metis ...), or (simp only: ...) proofs):
  1177 
  1178   Inf_singleton ~> Inf_insert [where A="{}", unfolded Inf_empty inf_top_right]
  1179   Sup_singleton ~> Sup_insert [where A="{}", unfolded Sup_empty sup_bot_right]
  1180   Inf_binary ~> Inf_insert, Inf_empty, and inf_top_right
  1181   Sup_binary ~> Sup_insert, Sup_empty, and sup_bot_right
  1182   Int_eq_Inter ~> Inf_insert, Inf_empty, and inf_top_right
  1183   Un_eq_Union ~> Sup_insert, Sup_empty, and sup_bot_right
  1184   Inter_def ~> INF_def, image_def
  1185   Union_def ~> SUP_def, image_def
  1186   INT_eq ~> INF_def, and image_def
  1187   UN_eq ~> SUP_def, and image_def
  1188   INF_subset ~> INF_superset_mono [OF _ order_refl]
  1189 
  1190 More consistent and comprehensive names:
  1191 
  1192   INTER_eq_Inter_image ~> INF_def
  1193   UNION_eq_Union_image ~> SUP_def
  1194   INFI_def ~> INF_def
  1195   SUPR_def ~> SUP_def
  1196   INF_leI ~> INF_lower
  1197   INF_leI2 ~> INF_lower2
  1198   le_INFI ~> INF_greatest
  1199   le_SUPI ~> SUP_upper
  1200   le_SUPI2 ~> SUP_upper2
  1201   SUP_leI ~> SUP_least
  1202   INFI_bool_eq ~> INF_bool_eq
  1203   SUPR_bool_eq ~> SUP_bool_eq
  1204   INFI_apply ~> INF_apply
  1205   SUPR_apply ~> SUP_apply
  1206   INTER_def ~> INTER_eq
  1207   UNION_def ~> UNION_eq
  1208 
  1209 INCOMPATIBILITY.
  1210 
  1211 * Renamed theory Complete_Lattice to Complete_Lattices.
  1212 INCOMPATIBILITY.
  1213 
  1214 * Theory Complete_Lattices: lemmas Inf_eq_top_iff, INF_eq_top_iff,
  1215 INF_image, Inf_insert, INF_top, Inf_top_conv, INF_top_conv, SUP_bot,
  1216 Sup_bot_conv, SUP_bot_conv, Sup_eq_top_iff, SUP_eq_top_iff, SUP_image,
  1217 Sup_insert are now declared as [simp].  INCOMPATIBILITY.
  1218 
  1219 * Theory Lattice: lemmas compl_inf_bot, compl_le_comp_iff,
  1220 compl_sup_top, inf_idem, inf_left_idem, inf_sup_absorb, sup_idem,
  1221 sup_inf_absob, sup_left_idem are now declared as [simp].  Minor
  1222 INCOMPATIBILITY.
  1223 
  1224 * Added syntactic classes "inf" and "sup" for the respective
  1225 constants.  INCOMPATIBILITY: Changes in the argument order of the
  1226 (mostly internal) locale predicates for some derived classes.
  1227 
  1228 * Theorem collections ball_simps and bex_simps do not contain theorems
  1229 referring to UNION any longer; these have been moved to collection
  1230 UN_ball_bex_simps.  INCOMPATIBILITY.
  1231 
  1232 * Theory Archimedean_Field: floor now is defined as parameter of a
  1233 separate type class floor_ceiling.
  1234 
  1235 * Theory Finite_Set: more coherent development of fold_set locales:
  1236 
  1237     locale fun_left_comm ~> locale comp_fun_commute
  1238     locale fun_left_comm_idem ~> locale comp_fun_idem
  1239 
  1240 Both use point-free characterization; interpretation proofs may need
  1241 adjustment.  INCOMPATIBILITY.
  1242 
  1243 * Theory Limits: Type "'a net" has been renamed to "'a filter", in
  1244 accordance with standard mathematical terminology. INCOMPATIBILITY.
  1245 
  1246 * Theory Complex_Main: The locale interpretations for the
  1247 bounded_linear and bounded_bilinear locales have been removed, in
  1248 order to reduce the number of duplicate lemmas. Users must use the
  1249 original names for distributivity theorems, potential INCOMPATIBILITY.
  1250 
  1251   divide.add ~> add_divide_distrib
  1252   divide.diff ~> diff_divide_distrib
  1253   divide.setsum ~> setsum_divide_distrib
  1254   mult.add_right ~> right_distrib
  1255   mult.diff_right ~> right_diff_distrib
  1256   mult_right.setsum ~> setsum_right_distrib
  1257   mult_left.diff ~> left_diff_distrib
  1258 
  1259 * Theory Complex_Main: Several redundant theorems have been removed or
  1260 replaced by more general versions. INCOMPATIBILITY.
  1261 
  1262   real_diff_def ~> minus_real_def
  1263   real_divide_def ~> divide_real_def
  1264   real_less_def ~> less_le
  1265   real_abs_def ~> abs_real_def
  1266   real_sgn_def ~> sgn_real_def
  1267   real_mult_commute ~> mult_commute
  1268   real_mult_assoc ~> mult_assoc
  1269   real_mult_1 ~> mult_1_left
  1270   real_add_mult_distrib ~> left_distrib
  1271   real_zero_not_eq_one ~> zero_neq_one
  1272   real_mult_inverse_left ~> left_inverse
  1273   INVERSE_ZERO ~> inverse_zero
  1274   real_le_refl ~> order_refl
  1275   real_le_antisym ~> order_antisym
  1276   real_le_trans ~> order_trans
  1277   real_le_linear ~> linear
  1278   real_le_eq_diff ~> le_iff_diff_le_0
  1279   real_add_left_mono ~> add_left_mono
  1280   real_mult_order ~> mult_pos_pos
  1281   real_mult_less_mono2 ~> mult_strict_left_mono
  1282   real_of_int_real_of_nat ~> real_of_int_of_nat_eq
  1283   real_0_le_divide_iff ~> zero_le_divide_iff
  1284   realpow_two_disj ~> power2_eq_iff
  1285   real_squared_diff_one_factored ~> square_diff_one_factored
  1286   realpow_two_diff ~> square_diff_square_factored
  1287   reals_complete2 ~> complete_real
  1288   real_sum_squared_expand ~> power2_sum
  1289   exp_ln_eq ~> ln_unique
  1290   expi_add ~> exp_add
  1291   expi_zero ~> exp_zero
  1292   lemma_DERIV_subst ~> DERIV_cong
  1293   LIMSEQ_Zfun_iff ~> tendsto_Zfun_iff
  1294   LIMSEQ_const ~> tendsto_const
  1295   LIMSEQ_norm ~> tendsto_norm
  1296   LIMSEQ_add ~> tendsto_add
  1297   LIMSEQ_minus ~> tendsto_minus
  1298   LIMSEQ_minus_cancel ~> tendsto_minus_cancel
  1299   LIMSEQ_diff ~> tendsto_diff
  1300   bounded_linear.LIMSEQ ~> bounded_linear.tendsto
  1301   bounded_bilinear.LIMSEQ ~> bounded_bilinear.tendsto
  1302   LIMSEQ_mult ~> tendsto_mult
  1303   LIMSEQ_inverse ~> tendsto_inverse
  1304   LIMSEQ_divide ~> tendsto_divide
  1305   LIMSEQ_pow ~> tendsto_power
  1306   LIMSEQ_setsum ~> tendsto_setsum
  1307   LIMSEQ_setprod ~> tendsto_setprod
  1308   LIMSEQ_norm_zero ~> tendsto_norm_zero_iff
  1309   LIMSEQ_rabs_zero ~> tendsto_rabs_zero_iff
  1310   LIMSEQ_imp_rabs ~> tendsto_rabs
  1311   LIMSEQ_add_minus ~> tendsto_add [OF _ tendsto_minus]
  1312   LIMSEQ_add_const ~> tendsto_add [OF _ tendsto_const]
  1313   LIMSEQ_diff_const ~> tendsto_diff [OF _ tendsto_const]
  1314   LIMSEQ_Complex ~> tendsto_Complex
  1315   LIM_ident ~> tendsto_ident_at
  1316   LIM_const ~> tendsto_const
  1317   LIM_add ~> tendsto_add
  1318   LIM_add_zero ~> tendsto_add_zero
  1319   LIM_minus ~> tendsto_minus
  1320   LIM_diff ~> tendsto_diff
  1321   LIM_norm ~> tendsto_norm
  1322   LIM_norm_zero ~> tendsto_norm_zero
  1323   LIM_norm_zero_cancel ~> tendsto_norm_zero_cancel
  1324   LIM_norm_zero_iff ~> tendsto_norm_zero_iff
  1325   LIM_rabs ~> tendsto_rabs
  1326   LIM_rabs_zero ~> tendsto_rabs_zero
  1327   LIM_rabs_zero_cancel ~> tendsto_rabs_zero_cancel
  1328   LIM_rabs_zero_iff ~> tendsto_rabs_zero_iff
  1329   LIM_compose ~> tendsto_compose
  1330   LIM_mult ~> tendsto_mult
  1331   LIM_scaleR ~> tendsto_scaleR
  1332   LIM_of_real ~> tendsto_of_real
  1333   LIM_power ~> tendsto_power
  1334   LIM_inverse ~> tendsto_inverse
  1335   LIM_sgn ~> tendsto_sgn
  1336   isCont_LIM_compose ~> isCont_tendsto_compose
  1337   bounded_linear.LIM ~> bounded_linear.tendsto
  1338   bounded_linear.LIM_zero ~> bounded_linear.tendsto_zero
  1339   bounded_bilinear.LIM ~> bounded_bilinear.tendsto
  1340   bounded_bilinear.LIM_prod_zero ~> bounded_bilinear.tendsto_zero
  1341   bounded_bilinear.LIM_left_zero ~> bounded_bilinear.tendsto_left_zero
  1342   bounded_bilinear.LIM_right_zero ~> bounded_bilinear.tendsto_right_zero
  1343   LIM_inverse_fun ~> tendsto_inverse [OF tendsto_ident_at]
  1344 
  1345 * Theory Complex_Main: The definition of infinite series was
  1346 generalized.  Now it is defined on the type class {topological_space,
  1347 comm_monoid_add}.  Hence it is useable also for extended real numbers.
  1348 
  1349 * Theory Complex_Main: The complex exponential function "expi" is now
  1350 a type-constrained abbreviation for "exp :: complex => complex"; thus
  1351 several polymorphic lemmas about "exp" are now applicable to "expi".
  1352 
  1353 * Code generation:
  1354 
  1355   - Theory Library/Code_Char_ord provides native ordering of
  1356     characters in the target language.
  1357 
  1358   - Commands code_module and code_library are legacy, use export_code
  1359     instead.
  1360 
  1361   - Method "evaluation" is legacy, use method "eval" instead.
  1362 
  1363   - Legacy evaluator "SML" is deactivated by default.  May be
  1364     reactivated by the following theory command:
  1365 
  1366       setup {* Value.add_evaluator ("SML", Codegen.eval_term) *}
  1367 
  1368 * Declare ext [intro] by default.  Rare INCOMPATIBILITY.
  1369 
  1370 * New proof method "induction" that gives induction hypotheses the
  1371 name "IH", thus distinguishing them from further hypotheses that come
  1372 from rule induction.  The latter are still called "hyps".  Method
  1373 "induction" is a thin wrapper around "induct" and follows the same
  1374 syntax.
  1375 
  1376 * Method "fastsimp" has been renamed to "fastforce", but "fastsimp" is
  1377 still available as a legacy feature for some time.
  1378 
  1379 * Nitpick:
  1380   - Added "need" and "total_consts" options.
  1381   - Reintroduced "show_skolems" option by popular demand.
  1382   - Renamed attribute: nitpick_def ~> nitpick_unfold.
  1383     INCOMPATIBILITY.
  1384 
  1385 * Sledgehammer:
  1386   - Use quasi-sound (and efficient) translations by default.
  1387   - Added support for the following provers: E-ToFoF, LEO-II,
  1388     Satallax, SNARK, Waldmeister, and Z3 with TPTP syntax.
  1389   - Automatically preplay and minimize proofs before showing them if
  1390     this can be done within reasonable time.
  1391   - sledgehammer available_provers ~> sledgehammer supported_provers.
  1392     INCOMPATIBILITY.
  1393   - Added "preplay_timeout", "slicing", "type_enc", "sound",
  1394     "max_mono_iters", and "max_new_mono_instances" options.
  1395   - Removed "explicit_apply" and "full_types" options as well as "Full
  1396     Types" Proof General menu item. INCOMPATIBILITY.
  1397 
  1398 * Metis:
  1399   - Removed "metisF" -- use "metis" instead. INCOMPATIBILITY.
  1400   - Obsoleted "metisFT" -- use "metis (full_types)" instead.
  1401     INCOMPATIBILITY.
  1402 
  1403 * Command 'try':
  1404   - Renamed 'try_methods' and added "simp:", "intro:", "dest:", and
  1405     "elim:" options. INCOMPATIBILITY.
  1406   - Introduced 'try' that not only runs 'try_methods' but also
  1407     'solve_direct', 'sledgehammer', 'quickcheck', and 'nitpick'.
  1408 
  1409 * Quickcheck:
  1410   - Added "eval" option to evaluate terms for the found counterexample
  1411     (currently only supported by the default (exhaustive) tester).
  1412   - Added post-processing of terms to obtain readable counterexamples
  1413     (currently only supported by the default (exhaustive) tester).
  1414   - New counterexample generator quickcheck[narrowing] enables
  1415     narrowing-based testing.  Requires the Glasgow Haskell compiler
  1416     with its installation location defined in the Isabelle settings
  1417     environment as ISABELLE_GHC.
  1418   - Removed quickcheck tester "SML" based on the SML code generator
  1419     (formly in HOL/Library).
  1420 
  1421 * Function package: discontinued option "tailrec".  INCOMPATIBILITY,
  1422 use 'partial_function' instead.
  1423 
  1424 * Theory Library/Extended_Reals replaces now the positive extended
  1425 reals found in probability theory. This file is extended by
  1426 Multivariate_Analysis/Extended_Real_Limits.
  1427 
  1428 * Theory Library/Old_Recdef: old 'recdef' package has been moved here,
  1429 from where it must be imported explicitly if it is really required.
  1430 INCOMPATIBILITY.
  1431 
  1432 * Theory Library/Wfrec: well-founded recursion combinator "wfrec" has
  1433 been moved here.  INCOMPATIBILITY.
  1434 
  1435 * Theory Library/Saturated provides type of numbers with saturated
  1436 arithmetic.
  1437 
  1438 * Theory Library/Product_Lattice defines a pointwise ordering for the
  1439 product type 'a * 'b, and provides instance proofs for various order
  1440 and lattice type classes.
  1441 
  1442 * Theory Library/Countable now provides the "countable_datatype" proof
  1443 method for proving "countable" class instances for datatypes.
  1444 
  1445 * Theory Library/Cset_Monad allows do notation for computable sets
  1446 (cset) via the generic monad ad-hoc overloading facility.
  1447 
  1448 * Library: Theories of common data structures are split into theories
  1449 for implementation, an invariant-ensuring type, and connection to an
  1450 abstract type. INCOMPATIBILITY.
  1451 
  1452   - RBT is split into RBT and RBT_Mapping.
  1453   - AssocList is split and renamed into AList and AList_Mapping.
  1454   - DList is split into DList_Impl, DList, and DList_Cset.
  1455   - Cset is split into Cset and List_Cset.
  1456 
  1457 * Theory Library/Nat_Infinity has been renamed to
  1458 Library/Extended_Nat, with name changes of the following types and
  1459 constants:
  1460 
  1461   type inat   ~> type enat
  1462   Fin         ~> enat
  1463   Infty       ~> infinity (overloaded)
  1464   iSuc        ~> eSuc
  1465   the_Fin     ~> the_enat
  1466 
  1467 Every theorem name containing "inat", "Fin", "Infty", or "iSuc" has
  1468 been renamed accordingly. INCOMPATIBILITY.
  1469 
  1470 * Session Multivariate_Analysis: The euclidean_space type class now
  1471 fixes a constant "Basis :: 'a set" consisting of the standard
  1472 orthonormal basis for the type. Users now have the option of
  1473 quantifying over this set instead of using the "basis" function, e.g.
  1474 "ALL x:Basis. P x" vs "ALL i<DIM('a). P (basis i)".
  1475 
  1476 * Session Multivariate_Analysis: Type "('a, 'b) cart" has been renamed
  1477 to "('a, 'b) vec" (the syntax "'a ^ 'b" remains unaffected). Constants
  1478 "Cart_nth" and "Cart_lambda" have been respectively renamed to
  1479 "vec_nth" and "vec_lambda"; theorems mentioning those names have
  1480 changed to match. Definition theorems for overloaded constants now use
  1481 the standard "foo_vec_def" naming scheme. A few other theorems have
  1482 been renamed as follows (INCOMPATIBILITY):
  1483 
  1484   Cart_eq          ~> vec_eq_iff
  1485   dist_nth_le_cart ~> dist_vec_nth_le
  1486   tendsto_vector   ~> vec_tendstoI
  1487   Cauchy_vector    ~> vec_CauchyI
  1488 
  1489 * Session Multivariate_Analysis: Several duplicate theorems have been
  1490 removed, and other theorems have been renamed or replaced with more
  1491 general versions. INCOMPATIBILITY.
  1492 
  1493   finite_choice ~> finite_set_choice
  1494   eventually_conjI ~> eventually_conj
  1495   eventually_and ~> eventually_conj_iff
  1496   eventually_false ~> eventually_False
  1497   setsum_norm ~> norm_setsum
  1498   Lim_sequentially ~> LIMSEQ_def
  1499   Lim_ident_at ~> LIM_ident
  1500   Lim_const ~> tendsto_const
  1501   Lim_cmul ~> tendsto_scaleR [OF tendsto_const]
  1502   Lim_neg ~> tendsto_minus
  1503   Lim_add ~> tendsto_add
  1504   Lim_sub ~> tendsto_diff
  1505   Lim_mul ~> tendsto_scaleR
  1506   Lim_vmul ~> tendsto_scaleR [OF _ tendsto_const]
  1507   Lim_null_norm ~> tendsto_norm_zero_iff [symmetric]
  1508   Lim_linear ~> bounded_linear.tendsto
  1509   Lim_component ~> tendsto_euclidean_component
  1510   Lim_component_cart ~> tendsto_vec_nth
  1511   Lim_inner ~> tendsto_inner [OF tendsto_const]
  1512   dot_lsum ~> inner_setsum_left
  1513   dot_rsum ~> inner_setsum_right
  1514   continuous_cmul ~> continuous_scaleR [OF continuous_const]
  1515   continuous_neg ~> continuous_minus
  1516   continuous_sub ~> continuous_diff
  1517   continuous_vmul ~> continuous_scaleR [OF _ continuous_const]
  1518   continuous_mul ~> continuous_scaleR
  1519   continuous_inv ~> continuous_inverse
  1520   continuous_at_within_inv ~> continuous_at_within_inverse
  1521   continuous_at_inv ~> continuous_at_inverse
  1522   continuous_at_norm ~> continuous_norm [OF continuous_at_id]
  1523   continuous_at_infnorm ~> continuous_infnorm [OF continuous_at_id]
  1524   continuous_at_component ~> continuous_component [OF continuous_at_id]
  1525   continuous_on_neg ~> continuous_on_minus
  1526   continuous_on_sub ~> continuous_on_diff
  1527   continuous_on_cmul ~> continuous_on_scaleR [OF continuous_on_const]
  1528   continuous_on_vmul ~> continuous_on_scaleR [OF _ continuous_on_const]
  1529   continuous_on_mul ~> continuous_on_scaleR
  1530   continuous_on_mul_real ~> continuous_on_mult
  1531   continuous_on_inner ~> continuous_on_inner [OF continuous_on_const]
  1532   continuous_on_norm ~> continuous_on_norm [OF continuous_on_id]
  1533   continuous_on_inverse ~> continuous_on_inv
  1534   uniformly_continuous_on_neg ~> uniformly_continuous_on_minus
  1535   uniformly_continuous_on_sub ~> uniformly_continuous_on_diff
  1536   subset_interior ~> interior_mono
  1537   subset_closure ~> closure_mono
  1538   closure_univ ~> closure_UNIV
  1539   real_arch_lt ~> reals_Archimedean2
  1540   real_arch ~> reals_Archimedean3
  1541   real_abs_norm ~> abs_norm_cancel
  1542   real_abs_sub_norm ~> norm_triangle_ineq3
  1543   norm_cauchy_schwarz_abs ~> Cauchy_Schwarz_ineq2
  1544 
  1545 * Session HOL-Probability:
  1546   - Caratheodory's extension lemma is now proved for ring_of_sets.
  1547   - Infinite products of probability measures are now available.
  1548   - Sigma closure is independent, if the generator is independent
  1549   - Use extended reals instead of positive extended
  1550     reals. INCOMPATIBILITY.
  1551 
  1552 * Session HOLCF: Discontinued legacy theorem names, INCOMPATIBILITY.
  1553 
  1554   expand_fun_below ~> fun_below_iff
  1555   below_fun_ext ~> fun_belowI
  1556   expand_cfun_eq ~> cfun_eq_iff
  1557   ext_cfun ~> cfun_eqI
  1558   expand_cfun_below ~> cfun_below_iff
  1559   below_cfun_ext ~> cfun_belowI
  1560   monofun_fun_fun ~> fun_belowD
  1561   monofun_fun_arg ~> monofunE
  1562   monofun_lub_fun ~> adm_monofun [THEN admD]
  1563   cont_lub_fun ~> adm_cont [THEN admD]
  1564   cont2cont_Rep_CFun ~> cont2cont_APP
  1565   cont_Rep_CFun_app ~> cont_APP_app
  1566   cont_Rep_CFun_app_app ~> cont_APP_app_app
  1567   cont_cfun_fun ~> cont_Rep_cfun1 [THEN contE]
  1568   cont_cfun_arg ~> cont_Rep_cfun2 [THEN contE]
  1569   contlub_cfun ~> lub_APP [symmetric]
  1570   contlub_LAM ~> lub_LAM [symmetric]
  1571   thelubI ~> lub_eqI
  1572   UU_I ~> bottomI
  1573   lift_distinct1 ~> lift.distinct(1)
  1574   lift_distinct2 ~> lift.distinct(2)
  1575   Def_not_UU ~> lift.distinct(2)
  1576   Def_inject ~> lift.inject
  1577   below_UU_iff ~> below_bottom_iff
  1578   eq_UU_iff ~> eq_bottom_iff
  1579 
  1580 
  1581 *** Document preparation ***
  1582 
  1583 * Antiquotation @{rail} layouts railroad syntax diagrams, see also
  1584 isar-ref manual, both for description and actual application of the
  1585 same.
  1586 
  1587 * Antiquotation @{value} evaluates the given term and presents its
  1588 result.
  1589 
  1590 * Antiquotations: term style "isub" provides ad-hoc conversion of
  1591 variables x1, y23 into subscripted form x\<^isub>1,
  1592 y\<^isub>2\<^isub>3.
  1593 
  1594 * Predefined LaTeX macros for Isabelle symbols \<bind> and \<then>
  1595 (e.g. see ~~/src/HOL/Library/Monad_Syntax.thy).
  1596 
  1597 * Localized \isabellestyle switch can be used within blocks or groups
  1598 like this:
  1599 
  1600   \isabellestyle{it}  %preferred default
  1601   {\isabellestylett @{text "typewriter stuff"}}
  1602 
  1603 * Discontinued special treatment of hard tabulators.  Implicit
  1604 tab-width is now defined as 1.  Potential INCOMPATIBILITY for visual
  1605 layouts.
  1606 
  1607 
  1608 *** ML ***
  1609 
  1610 * The inner syntax of sort/type/term/prop supports inlined YXML
  1611 representations within quoted string tokens.  By encoding logical
  1612 entities via Term_XML (in ML or Scala) concrete syntax can be
  1613 bypassed, which is particularly useful for producing bits of text
  1614 under external program control.
  1615 
  1616 * Antiquotations for ML and document preparation are managed as theory
  1617 data, which requires explicit setup.
  1618 
  1619 * Isabelle_Process.is_active allows tools to check if the official
  1620 process wrapper is running (Isabelle/Scala/jEdit) or the old TTY loop
  1621 (better known as Proof General).
  1622 
  1623 * Structure Proof_Context follows standard naming scheme.  Old
  1624 ProofContext is still available for some time as legacy alias.
  1625 
  1626 * Structure Timing provides various operations for timing; supersedes
  1627 former start_timing/end_timing etc.
  1628 
  1629 * Path.print is the official way to show file-system paths to users
  1630 (including quotes etc.).
  1631 
  1632 * Inner syntax: identifiers in parse trees of generic categories
  1633 "logic", "aprop", "idt" etc. carry position information (disguised as
  1634 type constraints).  Occasional INCOMPATIBILITY with non-compliant
  1635 translations that choke on unexpected type constraints.  Positions can
  1636 be stripped in ML translations via Syntax.strip_positions /
  1637 Syntax.strip_positions_ast, or via the syntax constant
  1638 "_strip_positions" within parse trees.  As last resort, positions can
  1639 be disabled via the configuration option Syntax.positions, which is
  1640 called "syntax_positions" in Isar attribute syntax.
  1641 
  1642 * Discontinued special status of various ML structures that contribute
  1643 to structure Syntax (Ast, Lexicon, Mixfix, Parser, Printer etc.): less
  1644 pervasive content, no inclusion in structure Syntax.  INCOMPATIBILITY,
  1645 refer directly to Ast.Constant, Lexicon.is_identifier,
  1646 Syntax_Trans.mk_binder_tr etc.
  1647 
  1648 * Typed print translation: discontinued show_sorts argument, which is
  1649 already available via context of "advanced" translation.
  1650 
  1651 * Refined PARALLEL_GOALS tactical: degrades gracefully for schematic
  1652 goal states; body tactic needs to address all subgoals uniformly.
  1653 
  1654 * Slightly more special eq_list/eq_set, with shortcut involving
  1655 pointer equality (assumes that eq relation is reflexive).
  1656 
  1657 * Classical tactics use proper Proof.context instead of historic types
  1658 claset/clasimpset.  Old-style declarations like addIs, addEs, addDs
  1659 operate directly on Proof.context.  Raw type claset retains its use as
  1660 snapshot of the classical context, which can be recovered via
  1661 (put_claset HOL_cs) etc.  Type clasimpset has been discontinued.
  1662 INCOMPATIBILITY, classical tactics and derived proof methods require
  1663 proper Proof.context.
  1664 
  1665 
  1666 *** System ***
  1667 
  1668 * Discontinued support for Poly/ML 5.2, which was the last version
  1669 without proper multithreading and TimeLimit implementation.
  1670 
  1671 * Discontinued old lib/scripts/polyml-platform, which has been
  1672 obsolete since Isabelle2009-2.
  1673 
  1674 * Various optional external tools are referenced more robustly and
  1675 uniformly by explicit Isabelle settings as follows:
  1676 
  1677   ISABELLE_CSDP   (formerly CSDP_EXE)
  1678   ISABELLE_GHC    (formerly EXEC_GHC or GHC_PATH)
  1679   ISABELLE_OCAML  (formerly EXEC_OCAML)
  1680   ISABELLE_SWIPL  (formerly EXEC_SWIPL)
  1681   ISABELLE_YAP    (formerly EXEC_YAP)
  1682 
  1683 Note that automated detection from the file-system or search path has
  1684 been discontinued.  INCOMPATIBILITY.
  1685 
  1686 * Scala layer provides JVM method invocation service for static
  1687 methods of type (String)String, see Invoke_Scala.method in ML.  For
  1688 example:
  1689 
  1690   Invoke_Scala.method "java.lang.System.getProperty" "java.home"
  1691 
  1692 Together with YXML.string_of_body/parse_body and XML.Encode/Decode
  1693 this allows to pass structured values between ML and Scala.
  1694 
  1695 * The IsabelleText fonts includes some further glyphs to support the
  1696 Prover IDE.  Potential INCOMPATIBILITY: users who happen to have
  1697 installed a local copy (which is normally *not* required) need to
  1698 delete or update it from ~~/lib/fonts/.
  1699 
  1700 
  1701 
  1702 New in Isabelle2011 (January 2011)
  1703 ----------------------------------
  1704 
  1705 *** General ***
  1706 
  1707 * Experimental Prover IDE based on Isabelle/Scala and jEdit (see
  1708 src/Tools/jEdit).  This also serves as IDE for Isabelle/ML, with
  1709 useful tooltips and hyperlinks produced from its static analysis.  The
  1710 bundled component provides an executable Isabelle tool that can be run
  1711 like this:
  1712 
  1713   Isabelle2011/bin/isabelle jedit
  1714 
  1715 * Significantly improved Isabelle/Isar implementation manual.
  1716 
  1717 * System settings: ISABELLE_HOME_USER now includes ISABELLE_IDENTIFIER
  1718 (and thus refers to something like $HOME/.isabelle/Isabelle2011),
  1719 while the default heap location within that directory lacks that extra
  1720 suffix.  This isolates multiple Isabelle installations from each
  1721 other, avoiding problems with old settings in new versions.
  1722 INCOMPATIBILITY, need to copy/upgrade old user settings manually.
  1723 
  1724 * Source files are always encoded as UTF-8, instead of old-fashioned
  1725 ISO-Latin-1.  INCOMPATIBILITY.  Isabelle LaTeX documents might require
  1726 the following package declarations:
  1727 
  1728   \usepackage[utf8]{inputenc}
  1729   \usepackage{textcomp}
  1730 
  1731 * Explicit treatment of UTF-8 sequences as Isabelle symbols, such that
  1732 a Unicode character is treated as a single symbol, not a sequence of
  1733 non-ASCII bytes as before.  Since Isabelle/ML string literals may
  1734 contain symbols without further backslash escapes, Unicode can now be
  1735 used here as well.  Recall that Symbol.explode in ML provides a
  1736 consistent view on symbols, while raw explode (or String.explode)
  1737 merely give a byte-oriented representation.
  1738 
  1739 * Theory loader: source files are primarily located via the master
  1740 directory of each theory node (where the .thy file itself resides).
  1741 The global load path is still partially available as legacy feature.
  1742 Minor INCOMPATIBILITY due to subtle change in file lookup: use
  1743 explicit paths, relatively to the theory.
  1744 
  1745 * Special treatment of ML file names has been discontinued.
  1746 Historically, optional extensions .ML or .sml were added on demand --
  1747 at the cost of clarity of file dependencies.  Recall that Isabelle/ML
  1748 files exclusively use the .ML extension.  Minor INCOMPATIBILTY.
  1749 
  1750 * Various options that affect pretty printing etc. are now properly
  1751 handled within the context via configuration options, instead of
  1752 unsynchronized references or print modes.  There are both ML Config.T
  1753 entities and Isar declaration attributes to access these.
  1754 
  1755   ML (Config.T)                 Isar (attribute)
  1756 
  1757   eta_contract                  eta_contract
  1758   show_brackets                 show_brackets
  1759   show_sorts                    show_sorts
  1760   show_types                    show_types
  1761   show_question_marks           show_question_marks
  1762   show_consts                   show_consts
  1763   show_abbrevs                  show_abbrevs
  1764 
  1765   Syntax.ast_trace              syntax_ast_trace
  1766   Syntax.ast_stat               syntax_ast_stat
  1767   Syntax.ambiguity_level        syntax_ambiguity_level
  1768 
  1769   Goal_Display.goals_limit      goals_limit
  1770   Goal_Display.show_main_goal   show_main_goal
  1771 
  1772   Method.rule_trace             rule_trace
  1773 
  1774   Thy_Output.display            thy_output_display
  1775   Thy_Output.quotes             thy_output_quotes
  1776   Thy_Output.indent             thy_output_indent
  1777   Thy_Output.source             thy_output_source
  1778   Thy_Output.break              thy_output_break
  1779 
  1780 Note that corresponding "..._default" references in ML may only be
  1781 changed globally at the ROOT session setup, but *not* within a theory.
  1782 The option "show_abbrevs" supersedes the former print mode
  1783 "no_abbrevs" with inverted meaning.
  1784 
  1785 * More systematic naming of some configuration options.
  1786 INCOMPATIBILITY.
  1787 
  1788   trace_simp  ~>  simp_trace
  1789   debug_simp  ~>  simp_debug
  1790 
  1791 * Support for real valued configuration options, using simplistic
  1792 floating-point notation that coincides with the inner syntax for
  1793 float_token.
  1794 
  1795 * Support for real valued preferences (with approximative PGIP type):
  1796 front-ends need to accept "pgint" values in float notation.
  1797 INCOMPATIBILITY.
  1798 
  1799 * The IsabelleText font now includes Cyrillic, Hebrew, Arabic from
  1800 DejaVu Sans.
  1801 
  1802 * Discontinued support for Poly/ML 5.0 and 5.1 versions.
  1803 
  1804 
  1805 *** Pure ***
  1806 
  1807 * Command 'type_synonym' (with single argument) replaces somewhat
  1808 outdated 'types', which is still available as legacy feature for some
  1809 time.
  1810 
  1811 * Command 'nonterminal' (with 'and' separated list of arguments)
  1812 replaces somewhat outdated 'nonterminals'.  INCOMPATIBILITY.
  1813 
  1814 * Command 'notepad' replaces former 'example_proof' for
  1815 experimentation in Isar without any result.  INCOMPATIBILITY.
  1816 
  1817 * Locale interpretation commands 'interpret' and 'sublocale' accept
  1818 lists of equations to map definitions in a locale to appropriate
  1819 entities in the context of the interpretation.  The 'interpretation'
  1820 command already provided this functionality.
  1821 
  1822 * Diagnostic command 'print_dependencies' prints the locale instances
  1823 that would be activated if the specified expression was interpreted in
  1824 the current context.  Variant "print_dependencies!" assumes a context
  1825 without interpretations.
  1826 
  1827 * Diagnostic command 'print_interps' prints interpretations in proofs
  1828 in addition to interpretations in theories.
  1829 
  1830 * Discontinued obsolete 'global' and 'local' commands to manipulate
  1831 the theory name space.  Rare INCOMPATIBILITY.  The ML functions
  1832 Sign.root_path and Sign.local_path may be applied directly where this
  1833 feature is still required for historical reasons.
  1834 
  1835 * Discontinued obsolete 'constdefs' command.  INCOMPATIBILITY, use
  1836 'definition' instead.
  1837 
  1838 * The "prems" fact, which refers to the accidental collection of
  1839 foundational premises in the context, is now explicitly marked as
  1840 legacy feature and will be discontinued soon.  Consider using "assms"
  1841 of the head statement or reference facts by explicit names.
  1842 
  1843 * Document antiquotations @{class} and @{type} print classes and type
  1844 constructors.
  1845 
  1846 * Document antiquotation @{file} checks file/directory entries within
  1847 the local file system.
  1848 
  1849 
  1850 *** HOL ***
  1851 
  1852 * Coercive subtyping: functions can be declared as coercions and type
  1853 inference will add them as necessary upon input of a term.  Theory
  1854 Complex_Main declares real :: nat => real and real :: int => real as
  1855 coercions. A coercion function f is declared like this:
  1856 
  1857   declare [[coercion f]]
  1858 
  1859 To lift coercions through type constructors (e.g. from nat => real to
  1860 nat list => real list), map functions can be declared, e.g.
  1861 
  1862   declare [[coercion_map map]]
  1863 
  1864 Currently coercion inference is activated only in theories including
  1865 real numbers, i.e. descendants of Complex_Main.  This is controlled by
  1866 the configuration option "coercion_enabled", e.g. it can be enabled in
  1867 other theories like this:
  1868 
  1869   declare [[coercion_enabled]]
  1870 
  1871 * Command 'partial_function' provides basic support for recursive
  1872 function definitions over complete partial orders.  Concrete instances
  1873 are provided for i) the option type, ii) tail recursion on arbitrary
  1874 types, and iii) the heap monad of Imperative_HOL.  See
  1875 src/HOL/ex/Fundefs.thy and src/HOL/Imperative_HOL/ex/Linked_Lists.thy
  1876 for examples.
  1877 
  1878 * Function package: f.psimps rules are no longer implicitly declared
  1879 as [simp].  INCOMPATIBILITY.
  1880 
  1881 * Datatype package: theorems generated for executable equality (class
  1882 "eq") carry proper names and are treated as default code equations.
  1883 
  1884 * Inductive package: now offers command 'inductive_simps' to
  1885 automatically derive instantiated and simplified equations for
  1886 inductive predicates, similar to 'inductive_cases'.
  1887 
  1888 * Command 'enriched_type' allows to register properties of the
  1889 functorial structure of types.
  1890 
  1891 * Improved infrastructure for term evaluation using code generator
  1892 techniques, in particular static evaluation conversions.
  1893 
  1894 * Code generator: Scala (2.8 or higher) has been added to the target
  1895 languages.
  1896 
  1897 * Code generator: globbing constant expressions "*" and "Theory.*"
  1898 have been replaced by the more idiomatic "_" and "Theory._".
  1899 INCOMPATIBILITY.
  1900 
  1901 * Code generator: export_code without explicit file declaration prints
  1902 to standard output.  INCOMPATIBILITY.
  1903 
  1904 * Code generator: do not print function definitions for case
  1905 combinators any longer.
  1906 
  1907 * Code generator: simplification with rules determined with
  1908 src/Tools/Code/code_simp.ML and method "code_simp".
  1909 
  1910 * Code generator for records: more idiomatic representation of record
  1911 types.  Warning: records are not covered by ancient SML code
  1912 generation any longer.  INCOMPATIBILITY.  In cases of need, a suitable
  1913 rep_datatype declaration helps to succeed then:
  1914 
  1915   record 'a foo = ...
  1916   ...
  1917   rep_datatype foo_ext ...
  1918 
  1919 * Records: logical foundation type for records does not carry a
  1920 '_type' suffix any longer (obsolete due to authentic syntax).
  1921 INCOMPATIBILITY.
  1922 
  1923 * Quickcheck now by default uses exhaustive testing instead of random
  1924 testing.  Random testing can be invoked by "quickcheck [random]",
  1925 exhaustive testing by "quickcheck [exhaustive]".
  1926 
  1927 * Quickcheck instantiates polymorphic types with small finite
  1928 datatypes by default. This enables a simple execution mechanism to
  1929 handle quantifiers and function equality over the finite datatypes.
  1930 
  1931 * Quickcheck random generator has been renamed from "code" to
  1932 "random".  INCOMPATIBILITY.
  1933 
  1934 * Quickcheck now has a configurable time limit which is set to 30
  1935 seconds by default. This can be changed by adding [timeout = n] to the
  1936 quickcheck command. The time limit for Auto Quickcheck is still set
  1937 independently.
  1938 
  1939 * Quickcheck in locales considers interpretations of that locale for
  1940 counter example search.
  1941 
  1942 * Sledgehammer:
  1943   - Added "smt" and "remote_smt" provers based on the "smt" proof
  1944     method. See the Sledgehammer manual for details ("isabelle doc
  1945     sledgehammer").
  1946   - Renamed commands:
  1947     sledgehammer atp_info ~> sledgehammer running_provers
  1948     sledgehammer atp_kill ~> sledgehammer kill_provers
  1949     sledgehammer available_atps ~> sledgehammer available_provers
  1950     INCOMPATIBILITY.
  1951   - Renamed options:
  1952     sledgehammer [atps = ...] ~> sledgehammer [provers = ...]
  1953     sledgehammer [atp = ...] ~> sledgehammer [prover = ...]
  1954     sledgehammer [timeout = 77 s] ~> sledgehammer [timeout = 77]
  1955     (and "ms" and "min" are no longer supported)
  1956     INCOMPATIBILITY.
  1957 
  1958 * Nitpick:
  1959   - Renamed options:
  1960     nitpick [timeout = 77 s] ~> nitpick [timeout = 77]
  1961     nitpick [tac_timeout = 777 ms] ~> nitpick [tac_timeout = 0.777]
  1962     INCOMPATIBILITY.
  1963   - Added support for partial quotient types.
  1964   - Added local versions of the "Nitpick.register_xxx" functions.
  1965   - Added "whack" option.
  1966   - Allow registration of quotient types as codatatypes.
  1967   - Improved "merge_type_vars" option to merge more types.
  1968   - Removed unsound "fast_descrs" option.
  1969   - Added custom symmetry breaking for datatypes, making it possible to reach
  1970     higher cardinalities.
  1971   - Prevent the expansion of too large definitions.
  1972 
  1973 * Proof methods "metis" and "meson" now have configuration options
  1974 "meson_trace", "metis_trace", and "metis_verbose" that can be enabled
  1975 to diagnose these tools. E.g.
  1976 
  1977     using [[metis_trace = true]]
  1978 
  1979 * Auto Solve: Renamed "Auto Solve Direct".  The tool is now available
  1980 manually as command 'solve_direct'.
  1981 
  1982 * The default SMT solver Z3 must be enabled explicitly (due to
  1983 licensing issues) by setting the environment variable
  1984 Z3_NON_COMMERCIAL in etc/settings of the component, for example.  For
  1985 commercial applications, the SMT solver CVC3 is provided as fall-back;
  1986 changing the SMT solver is done via the configuration option
  1987 "smt_solver".
  1988 
  1989 * Remote SMT solvers need to be referred to by the "remote_" prefix,
  1990 i.e. "remote_cvc3" and "remote_z3".
  1991 
  1992 * Added basic SMT support for datatypes, records, and typedefs using
  1993 the oracle mode (no proofs).  Direct support of pairs has been dropped
  1994 in exchange (pass theorems fst_conv snd_conv pair_collapse to the SMT
  1995 support for a similar behavior).  Minor INCOMPATIBILITY.
  1996 
  1997 * Changed SMT configuration options:
  1998   - Renamed:
  1999     z3_proofs ~> smt_oracle (with inverted meaning)
  2000     z3_trace_assms ~> smt_trace_used_facts
  2001     INCOMPATIBILITY.
  2002   - Added:
  2003     smt_verbose
  2004     smt_random_seed
  2005     smt_datatypes
  2006     smt_infer_triggers
  2007     smt_monomorph_limit
  2008     cvc3_options
  2009     remote_cvc3_options
  2010     remote_z3_options
  2011     yices_options
  2012 
  2013 * Boogie output files (.b2i files) need to be declared in the theory
  2014 header.
  2015 
  2016 * Simplification procedure "list_to_set_comprehension" rewrites list
  2017 comprehensions applied to List.set to set comprehensions.  Occasional
  2018 INCOMPATIBILITY, may be deactivated like this:
  2019 
  2020   declare [[simproc del: list_to_set_comprehension]]
  2021 
  2022 * Removed old version of primrec package.  INCOMPATIBILITY.
  2023 
  2024 * Removed simplifier congruence rule of "prod_case", as has for long
  2025 been the case with "split".  INCOMPATIBILITY.
  2026 
  2027 * String.literal is a type, but not a datatype.  INCOMPATIBILITY.
  2028 
  2029 * Removed [split_format ... and ... and ...] version of
  2030 [split_format].  Potential INCOMPATIBILITY.
  2031 
  2032 * Predicate "sorted" now defined inductively, with nice induction
  2033 rules.  INCOMPATIBILITY: former sorted.simps now named sorted_simps.
  2034 
  2035 * Constant "contents" renamed to "the_elem", to free the generic name
  2036 contents for other uses.  INCOMPATIBILITY.
  2037 
  2038 * Renamed class eq and constant eq (for code generation) to class
  2039 equal and constant equal, plus renaming of related facts and various
  2040 tuning.  INCOMPATIBILITY.
  2041 
  2042 * Dropped type classes mult_mono and mult_mono1.  INCOMPATIBILITY.
  2043 
  2044 * Removed output syntax "'a ~=> 'b" for "'a => 'b option".
  2045 INCOMPATIBILITY.
  2046 
  2047 * Renamed theory Fset to Cset, type Fset.fset to Cset.set, in order to
  2048 avoid confusion with finite sets.  INCOMPATIBILITY.
  2049 
  2050 * Abandoned locales equiv, congruent and congruent2 for equivalence
  2051 relations.  INCOMPATIBILITY: use equivI rather than equiv_intro (same
  2052 for congruent(2)).
  2053 
  2054 * Some previously unqualified names have been qualified:
  2055 
  2056   types
  2057     bool ~> HOL.bool
  2058     nat ~> Nat.nat
  2059 
  2060   constants
  2061     Trueprop ~> HOL.Trueprop
  2062     True ~> HOL.True
  2063     False ~> HOL.False
  2064     op & ~> HOL.conj
  2065     op | ~> HOL.disj
  2066     op --> ~> HOL.implies
  2067     op = ~> HOL.eq
  2068     Not ~> HOL.Not
  2069     The ~> HOL.The
  2070     All ~> HOL.All
  2071     Ex ~> HOL.Ex
  2072     Ex1 ~> HOL.Ex1
  2073     Let ~> HOL.Let
  2074     If ~> HOL.If
  2075     Ball ~> Set.Ball
  2076     Bex ~> Set.Bex
  2077     Suc ~> Nat.Suc
  2078     Pair ~> Product_Type.Pair
  2079     fst ~> Product_Type.fst
  2080     snd ~> Product_Type.snd
  2081     curry ~> Product_Type.curry
  2082     op : ~> Set.member
  2083     Collect ~> Set.Collect
  2084 
  2085 INCOMPATIBILITY.
  2086 
  2087 * More canonical naming convention for some fundamental definitions:
  2088 
  2089     bot_bool_eq ~> bot_bool_def
  2090     top_bool_eq ~> top_bool_def
  2091     inf_bool_eq ~> inf_bool_def
  2092     sup_bool_eq ~> sup_bool_def
  2093     bot_fun_eq  ~> bot_fun_def
  2094     top_fun_eq  ~> top_fun_def
  2095     inf_fun_eq  ~> inf_fun_def
  2096     sup_fun_eq  ~> sup_fun_def
  2097 
  2098 INCOMPATIBILITY.
  2099 
  2100 * More stylized fact names:
  2101 
  2102   expand_fun_eq ~> fun_eq_iff
  2103   expand_set_eq ~> set_eq_iff
  2104   set_ext       ~> set_eqI
  2105   nat_number    ~> eval_nat_numeral
  2106 
  2107 INCOMPATIBILITY.
  2108 
  2109 * Refactoring of code-generation specific operations in theory List:
  2110 
  2111   constants
  2112     null ~> List.null
  2113 
  2114   facts
  2115     mem_iff ~> member_def
  2116     null_empty ~> null_def
  2117 
  2118 INCOMPATIBILITY.  Note that these were not supposed to be used
  2119 regularly unless for striking reasons; their main purpose was code
  2120 generation.
  2121 
  2122 Various operations from the Haskell prelude are used for generating
  2123 Haskell code.
  2124 
  2125 * Term "bij f" is now an abbreviation of "bij_betw f UNIV UNIV".  Term
  2126 "surj f" is now an abbreviation of "range f = UNIV".  The theorems
  2127 bij_def and surj_def are unchanged.  INCOMPATIBILITY.
  2128 
  2129 * Abolished some non-alphabetic type names: "prod" and "sum" replace
  2130 "*" and "+" respectively.  INCOMPATIBILITY.
  2131 
  2132 * Name "Plus" of disjoint sum operator "<+>" is now hidden.  Write
  2133 "Sum_Type.Plus" instead.
  2134 
  2135 * Constant "split" has been merged with constant "prod_case"; names of
  2136 ML functions, facts etc. involving split have been retained so far,
  2137 though.  INCOMPATIBILITY.
  2138 
  2139 * Dropped old infix syntax "_ mem _" for List.member; use "_ : set _"
  2140 instead.  INCOMPATIBILITY.
  2141 
  2142 * Removed lemma "Option.is_none_none" which duplicates "is_none_def".
  2143 INCOMPATIBILITY.
  2144 
  2145 * Former theory Library/Enum is now part of the HOL-Main image.
  2146 INCOMPATIBILITY: all constants of the Enum theory now have to be
  2147 referred to by its qualified name.
  2148 
  2149   enum    ~>  Enum.enum
  2150   nlists  ~>  Enum.nlists
  2151   product ~>  Enum.product
  2152 
  2153 * Theory Library/Monad_Syntax provides do-syntax for monad types.
  2154 Syntax in Library/State_Monad has been changed to avoid ambiguities.
  2155 INCOMPATIBILITY.
  2156 
  2157 * Theory Library/SetsAndFunctions has been split into
  2158 Library/Function_Algebras and Library/Set_Algebras; canonical names
  2159 for instance definitions for functions; various improvements.
  2160 INCOMPATIBILITY.
  2161 
  2162 * Theory Library/Multiset provides stable quicksort implementation of
  2163 sort_key.
  2164 
  2165 * Theory Library/Multiset: renamed empty_idemp ~> empty_neutral.
  2166 INCOMPATIBILITY.
  2167 
  2168 * Session Multivariate_Analysis: introduced a type class for euclidean
  2169 space.  Most theorems are now stated in terms of euclidean spaces
  2170 instead of finite cartesian products.
  2171 
  2172   types
  2173     real ^ 'n ~>  'a::real_vector
  2174               ~>  'a::euclidean_space
  2175               ~>  'a::ordered_euclidean_space
  2176         (depends on your needs)
  2177 
  2178   constants
  2179      _ $ _        ~> _ $$ _
  2180      \<chi> x. _  ~> \<chi>\<chi> x. _
  2181      CARD('n)     ~> DIM('a)
  2182 
  2183 Also note that the indices are now natural numbers and not from some
  2184 finite type. Finite cartesian products of euclidean spaces, products
  2185 of euclidean spaces the real and complex numbers are instantiated to
  2186 be euclidean_spaces.  INCOMPATIBILITY.
  2187 
  2188 * Session Probability: introduced pextreal as positive extended real
  2189 numbers.  Use pextreal as value for measures.  Introduce the
  2190 Radon-Nikodym derivative, product spaces and Fubini's theorem for
  2191 arbitrary sigma finite measures.  Introduces Lebesgue measure based on
  2192 the integral in Multivariate Analysis.  INCOMPATIBILITY.
  2193 
  2194 * Session Imperative_HOL: revamped, corrected dozens of inadequacies.
  2195 INCOMPATIBILITY.
  2196 
  2197 * Session SPARK (with image HOL-SPARK) provides commands to load and
  2198 prove verification conditions generated by the SPARK Ada program
  2199 verifier.  See also src/HOL/SPARK and src/HOL/SPARK/Examples.
  2200 
  2201 
  2202 *** HOL-Algebra ***
  2203 
  2204 * Theorems for additive ring operations (locale abelian_monoid and
  2205 descendants) are generated by interpretation from their multiplicative
  2206 counterparts.  Names (in particular theorem names) have the mandatory
  2207 qualifier 'add'.  Previous theorem names are redeclared for
  2208 compatibility.
  2209 
  2210 * Structure "int_ring" is now an abbreviation (previously a
  2211 definition).  This fits more natural with advanced interpretations.
  2212 
  2213 
  2214 *** HOLCF ***
  2215 
  2216 * The domain package now runs in definitional mode by default: The
  2217 former command 'new_domain' is now called 'domain'.  To use the domain
  2218 package in its original axiomatic mode, use 'domain (unsafe)'.
  2219 INCOMPATIBILITY.
  2220 
  2221 * The new class "domain" is now the default sort.  Class "predomain"
  2222 is an unpointed version of "domain". Theories can be updated by
  2223 replacing sort annotations as shown below.  INCOMPATIBILITY.
  2224 
  2225   'a::type ~> 'a::countable
  2226   'a::cpo  ~> 'a::predomain
  2227   'a::pcpo ~> 'a::domain
  2228 
  2229 * The old type class "rep" has been superseded by class "domain".
  2230 Accordingly, users of the definitional package must remove any
  2231 "default_sort rep" declarations.  INCOMPATIBILITY.
  2232 
  2233 * The domain package (definitional mode) now supports unpointed
  2234 predomain argument types, as long as they are marked 'lazy'. (Strict
  2235 arguments must be in class "domain".) For example, the following
  2236 domain definition now works:
  2237 
  2238   domain natlist = nil | cons (lazy "nat discr") (lazy "natlist")
  2239 
  2240 * Theory HOLCF/Library/HOL_Cpo provides cpo and predomain class
  2241 instances for types from main HOL: bool, nat, int, char, 'a + 'b,
  2242 'a option, and 'a list.  Additionally, it configures fixrec and the
  2243 domain package to work with these types.  For example:
  2244 
  2245   fixrec isInl :: "('a + 'b) u -> tr"
  2246     where "isInl$(up$(Inl x)) = TT" | "isInl$(up$(Inr y)) = FF"
  2247 
  2248   domain V = VFun (lazy "V -> V") | VCon (lazy "nat") (lazy "V list")
  2249 
  2250 * The "(permissive)" option of fixrec has been replaced with a
  2251 per-equation "(unchecked)" option. See
  2252 src/HOL/HOLCF/Tutorial/Fixrec_ex.thy for examples. INCOMPATIBILITY.
  2253 
  2254 * The "bifinite" class no longer fixes a constant "approx"; the class
  2255 now just asserts that such a function exists.  INCOMPATIBILITY.
  2256 
  2257 * Former type "alg_defl" has been renamed to "defl".  HOLCF no longer
  2258 defines an embedding of type 'a defl into udom by default; instances
  2259 of "bifinite" and "domain" classes are available in
  2260 src/HOL/HOLCF/Library/Defl_Bifinite.thy.
  2261 
  2262 * The syntax "REP('a)" has been replaced with "DEFL('a)".
  2263 
  2264 * The predicate "directed" has been removed.  INCOMPATIBILITY.
  2265 
  2266 * The type class "finite_po" has been removed.  INCOMPATIBILITY.
  2267 
  2268 * The function "cprod_map" has been renamed to "prod_map".
  2269 INCOMPATIBILITY.
  2270 
  2271 * The monadic bind operator on each powerdomain has new binder syntax
  2272 similar to sets, e.g. "\<Union>\<sharp>x\<in>xs. t" represents
  2273 "upper_bind\<cdot>xs\<cdot>(\<Lambda> x. t)".
  2274 
  2275 * The infix syntax for binary union on each powerdomain has changed
  2276 from e.g. "+\<sharp>" to "\<union>\<sharp>", for consistency with set
  2277 syntax.  INCOMPATIBILITY.
  2278 
  2279 * The constant "UU" has been renamed to "bottom".  The syntax "UU" is
  2280 still supported as an input translation.
  2281 
  2282 * Renamed some theorems (the original names are also still available).
  2283 
  2284   expand_fun_below   ~> fun_below_iff
  2285   below_fun_ext      ~> fun_belowI
  2286   expand_cfun_eq     ~> cfun_eq_iff
  2287   ext_cfun           ~> cfun_eqI
  2288   expand_cfun_below  ~> cfun_below_iff
  2289   below_cfun_ext     ~> cfun_belowI
  2290   cont2cont_Rep_CFun ~> cont2cont_APP
  2291 
  2292 * The Abs and Rep functions for various types have changed names.
  2293 Related theorem names have also changed to match. INCOMPATIBILITY.
  2294 
  2295   Rep_CFun  ~> Rep_cfun
  2296   Abs_CFun  ~> Abs_cfun
  2297   Rep_Sprod ~> Rep_sprod
  2298   Abs_Sprod ~> Abs_sprod
  2299   Rep_Ssum  ~> Rep_ssum
  2300   Abs_Ssum  ~> Abs_ssum
  2301 
  2302 * Lemmas with names of the form *_defined_iff or *_strict_iff have
  2303 been renamed to *_bottom_iff.  INCOMPATIBILITY.
  2304 
  2305 * Various changes to bisimulation/coinduction with domain package:
  2306 
  2307   - Definitions of "bisim" constants no longer mention definedness.
  2308   - With mutual recursion, "bisim" predicate is now curried.
  2309   - With mutual recursion, each type gets a separate coind theorem.
  2310   - Variable names in bisim_def and coinduct rules have changed.
  2311 
  2312 INCOMPATIBILITY.
  2313 
  2314 * Case combinators generated by the domain package for type "foo" are
  2315 now named "foo_case" instead of "foo_when".  INCOMPATIBILITY.
  2316 
  2317 * Several theorems have been renamed to more accurately reflect the
  2318 names of constants and types involved.  INCOMPATIBILITY.
  2319 
  2320   thelub_const    ~> lub_const
  2321   lub_const       ~> is_lub_const
  2322   thelubI         ~> lub_eqI
  2323   is_lub_lub      ~> is_lubD2
  2324   lubI            ~> is_lub_lub
  2325   unique_lub      ~> is_lub_unique
  2326   is_ub_lub       ~> is_lub_rangeD1
  2327   lub_bin_chain   ~> is_lub_bin_chain
  2328   lub_fun         ~> is_lub_fun
  2329   thelub_fun      ~> lub_fun
  2330   thelub_cfun     ~> lub_cfun
  2331   thelub_Pair     ~> lub_Pair
  2332   lub_cprod       ~> is_lub_prod
  2333   thelub_cprod    ~> lub_prod
  2334   minimal_cprod   ~> minimal_prod
  2335   inst_cprod_pcpo ~> inst_prod_pcpo
  2336   UU_I            ~> bottomI
  2337   compact_UU      ~> compact_bottom
  2338   deflation_UU    ~> deflation_bottom
  2339   finite_deflation_UU ~> finite_deflation_bottom
  2340 
  2341 * Many legacy theorem names have been discontinued.  INCOMPATIBILITY.
  2342 
  2343   sq_ord_less_eq_trans ~> below_eq_trans
  2344   sq_ord_eq_less_trans ~> eq_below_trans
  2345   refl_less            ~> below_refl
  2346   trans_less           ~> below_trans
  2347   antisym_less         ~> below_antisym
  2348   antisym_less_inverse ~> po_eq_conv [THEN iffD1]
  2349   box_less             ~> box_below
  2350   rev_trans_less       ~> rev_below_trans
  2351   not_less2not_eq      ~> not_below2not_eq
  2352   less_UU_iff          ~> below_UU_iff
  2353   flat_less_iff        ~> flat_below_iff
  2354   adm_less             ~> adm_below
  2355   adm_not_less         ~> adm_not_below
  2356   adm_compact_not_less ~> adm_compact_not_below
  2357   less_fun_def         ~> below_fun_def
  2358   expand_fun_less      ~> fun_below_iff
  2359   less_fun_ext         ~> fun_belowI
  2360   less_discr_def       ~> below_discr_def
  2361   discr_less_eq        ~> discr_below_eq
  2362   less_unit_def        ~> below_unit_def
  2363   less_cprod_def       ~> below_prod_def
  2364   prod_lessI           ~> prod_belowI
  2365   Pair_less_iff        ~> Pair_below_iff
  2366   fst_less_iff         ~> fst_below_iff
  2367   snd_less_iff         ~> snd_below_iff
  2368   expand_cfun_less     ~> cfun_below_iff
  2369   less_cfun_ext        ~> cfun_belowI
  2370   injection_less       ~> injection_below
  2371   less_up_def          ~> below_up_def
  2372   not_Iup_less         ~> not_Iup_below
  2373   Iup_less             ~> Iup_below
  2374   up_less              ~> up_below
  2375   Def_inject_less_eq   ~> Def_below_Def
  2376   Def_less_is_eq       ~> Def_below_iff
  2377   spair_less_iff       ~> spair_below_iff
  2378   less_sprod           ~> below_sprod
  2379   spair_less           ~> spair_below
  2380   sfst_less_iff        ~> sfst_below_iff
  2381   ssnd_less_iff        ~> ssnd_below_iff
  2382   fix_least_less       ~> fix_least_below
  2383   dist_less_one        ~> dist_below_one
  2384   less_ONE             ~> below_ONE
  2385   ONE_less_iff         ~> ONE_below_iff
  2386   less_sinlD           ~> below_sinlD
  2387   less_sinrD           ~> below_sinrD
  2388 
  2389 
  2390 *** FOL and ZF ***
  2391 
  2392 * All constant names are now qualified internally and use proper
  2393 identifiers, e.g. "IFOL.eq" instead of "op =".  INCOMPATIBILITY.
  2394 
  2395 
  2396 *** ML ***
  2397 
  2398 * Antiquotation @{assert} inlines a function bool -> unit that raises
  2399 Fail if the argument is false.  Due to inlining the source position of
  2400 failed assertions is included in the error output.
  2401 
  2402 * Discontinued antiquotation @{theory_ref}, which is obsolete since ML
  2403 text is in practice always evaluated with a stable theory checkpoint.
  2404 Minor INCOMPATIBILITY, use (Theory.check_thy @{theory}) instead.
  2405 
  2406 * Antiquotation @{theory A} refers to theory A from the ancestry of
  2407 the current context, not any accidental theory loader state as before.
  2408 Potential INCOMPATIBILITY, subtle change in semantics.
  2409 
  2410 * Syntax.pretty_priority (default 0) configures the required priority
  2411 of pretty-printed output and thus affects insertion of parentheses.
  2412 
  2413 * Syntax.default_root (default "any") configures the inner syntax
  2414 category (nonterminal symbol) for parsing of terms.
  2415 
  2416 * Former exception Library.UnequalLengths now coincides with
  2417 ListPair.UnequalLengths.
  2418 
  2419 * Renamed structure MetaSimplifier to Raw_Simplifier.  Note that the
  2420 main functionality is provided by structure Simplifier.
  2421 
  2422 * Renamed raw "explode" function to "raw_explode" to emphasize its
  2423 meaning.  Note that internally to Isabelle, Symbol.explode is used in
  2424 almost all situations.
  2425 
  2426 * Discontinued obsolete function sys_error and exception SYS_ERROR.
  2427 See implementation manual for further details on exceptions in
  2428 Isabelle/ML.
  2429 
  2430 * Renamed setmp_noncritical to Unsynchronized.setmp to emphasize its
  2431 meaning.
  2432 
  2433 * Renamed structure PureThy to Pure_Thy and moved most of its
  2434 operations to structure Global_Theory, to emphasize that this is
  2435 rarely-used global-only stuff.
  2436 
  2437 * Discontinued Output.debug.  Minor INCOMPATIBILITY, use plain writeln
  2438 instead (or tracing for high-volume output).
  2439 
  2440 * Configuration option show_question_marks only affects regular pretty
  2441 printing of types and terms, not raw Term.string_of_vname.
  2442 
  2443 * ML_Context.thm and ML_Context.thms are no longer pervasive.  Rare
  2444 INCOMPATIBILITY, superseded by static antiquotations @{thm} and
  2445 @{thms} for most purposes.
  2446 
  2447 * ML structure Unsynchronized is never opened, not even in Isar
  2448 interaction mode as before.  Old Unsynchronized.set etc. have been
  2449 discontinued -- use plain := instead.  This should be *rare* anyway,
  2450 since modern tools always work via official context data, notably
  2451 configuration options.
  2452 
  2453 * Parallel and asynchronous execution requires special care concerning
  2454 interrupts.  Structure Exn provides some convenience functions that
  2455 avoid working directly with raw Interrupt.  User code must not absorb
  2456 interrupts -- intermediate handling (for cleanup etc.) needs to be
  2457 followed by re-raising of the original exception.  Another common
  2458 source of mistakes are "handle _" patterns, which make the meaning of
  2459 the program subject to physical effects of the environment.
  2460 
  2461 
  2462 
  2463 New in Isabelle2009-2 (June 2010)
  2464 ---------------------------------
  2465 
  2466 *** General ***
  2467 
  2468 * Authentic syntax for *all* logical entities (type classes, type
  2469 constructors, term constants): provides simple and robust
  2470 correspondence between formal entities and concrete syntax.  Within
  2471 the parse tree / AST representations, "constants" are decorated by
  2472 their category (class, type, const) and spelled out explicitly with
  2473 their full internal name.
  2474 
  2475 Substantial INCOMPATIBILITY concerning low-level syntax declarations
  2476 and translations (translation rules and translation functions in ML).
  2477 Some hints on upgrading:
  2478 
  2479   - Many existing uses of 'syntax' and 'translations' can be replaced
  2480     by more modern 'type_notation', 'notation' and 'abbreviation',
  2481     which are independent of this issue.
  2482 
  2483   - 'translations' require markup within the AST; the term syntax
  2484     provides the following special forms:
  2485 
  2486       CONST c   -- produces syntax version of constant c from context
  2487       XCONST c  -- literally c, checked as constant from context
  2488       c         -- literally c, if declared by 'syntax'
  2489 
  2490     Plain identifiers are treated as AST variables -- occasionally the
  2491     system indicates accidental variables via the error "rhs contains
  2492     extra variables".
  2493 
  2494     Type classes and type constructors are marked according to their
  2495     concrete syntax.  Some old translations rules need to be written
  2496     for the "type" category, using type constructor application
  2497     instead of pseudo-term application of the default category
  2498     "logic".
  2499 
  2500   - 'parse_translation' etc. in ML may use the following
  2501     antiquotations:
  2502 
  2503       @{class_syntax c}   -- type class c within parse tree / AST
  2504       @{term_syntax c}    -- type constructor c within parse tree / AST
  2505       @{const_syntax c}   -- ML version of "CONST c" above
  2506       @{syntax_const c}   -- literally c (checked wrt. 'syntax' declarations)
  2507 
  2508   - Literal types within 'typed_print_translations', i.e. those *not*
  2509     represented as pseudo-terms are represented verbatim.  Use @{class
  2510     c} or @{type_name c} here instead of the above syntax
  2511     antiquotations.
  2512 
  2513 Note that old non-authentic syntax was based on unqualified base
  2514 names, so all of the above "constant" names would coincide.  Recall
  2515 that 'print_syntax' and ML_command "set Syntax.trace_ast" help to
  2516 diagnose syntax problems.
  2517 
  2518 * Type constructors admit general mixfix syntax, not just infix.
  2519 
  2520 * Concrete syntax may be attached to local entities without a proof
  2521 body, too.  This works via regular mixfix annotations for 'fix',
  2522 'def', 'obtain' etc. or via the explicit 'write' command, which is
  2523 similar to the 'notation' command in theory specifications.
  2524 
  2525 * Discontinued unnamed infix syntax (legacy feature for many years) --
  2526 need to specify constant name and syntax separately.  Internal ML
  2527 datatype constructors have been renamed from InfixName to Infix etc.
  2528 Minor INCOMPATIBILITY.
  2529 
  2530 * Schematic theorem statements need to be explicitly markup as such,
  2531 via commands 'schematic_lemma', 'schematic_theorem',
  2532 'schematic_corollary'.  Thus the relevance of the proof is made
  2533 syntactically clear, which impacts performance in a parallel or
  2534 asynchronous interactive environment.  Minor INCOMPATIBILITY.
  2535 
  2536 * Use of cumulative prems via "!" in some proof methods has been
  2537 discontinued (old legacy feature).
  2538 
  2539 * References 'trace_simp' and 'debug_simp' have been replaced by
  2540 configuration options stored in the context. Enabling tracing (the
  2541 case of debugging is similar) in proofs works via
  2542 
  2543   using [[trace_simp = true]]
  2544 
  2545 Tracing is then active for all invocations of the simplifier in
  2546 subsequent goal refinement steps. Tracing may also still be enabled or
  2547 disabled via the ProofGeneral settings menu.
  2548 
  2549 * Separate commands 'hide_class', 'hide_type', 'hide_const',
  2550 'hide_fact' replace the former 'hide' KIND command.  Minor
  2551 INCOMPATIBILITY.
  2552 
  2553 * Improved parallelism of proof term normalization: usedir -p2 -q0 is
  2554 more efficient than combinations with -q1 or -q2.
  2555 
  2556 
  2557 *** Pure ***
  2558 
  2559 * Proofterms record type-class reasoning explicitly, using the
  2560 "unconstrain" operation internally.  This eliminates all sort
  2561 constraints from a theorem and proof, introducing explicit
  2562 OFCLASS-premises.  On the proof term level, this operation is
  2563 automatically applied at theorem boundaries, such that closed proofs
  2564 are always free of sort constraints.  INCOMPATIBILITY for tools that
  2565 inspect proof terms.
  2566 
  2567 * Local theory specifications may depend on extra type variables that
  2568 are not present in the result type -- arguments TYPE('a) :: 'a itself
  2569 are added internally.  For example:
  2570 
  2571   definition unitary :: bool where "unitary = (ALL (x::'a) y. x = y)"
  2572 
  2573 * Predicates of locales introduced by classes carry a mandatory
  2574 "class" prefix.  INCOMPATIBILITY.
  2575 
  2576 * Vacuous class specifications observe default sort.  INCOMPATIBILITY.
  2577 
  2578 * Old 'axclass' command has been discontinued.  INCOMPATIBILITY, use
  2579 'class' instead.
  2580 
  2581 * Command 'code_reflect' allows to incorporate generated ML code into
  2582 runtime environment; replaces immature code_datatype antiquotation.
  2583 INCOMPATIBILITY.
  2584 
  2585 * Code generator: simple concept for abstract datatypes obeying
  2586 invariants.
  2587 
  2588 * Code generator: details of internal data cache have no impact on the
  2589 user space functionality any longer.
  2590 
  2591 * Methods "unfold_locales" and "intro_locales" ignore non-locale
  2592 subgoals.  This is more appropriate for interpretations with 'where'.
  2593 INCOMPATIBILITY.
  2594 
  2595 * Command 'example_proof' opens an empty proof body.  This allows to
  2596 experiment with Isar, without producing any persistent result.
  2597 
  2598 * Commands 'type_notation' and 'no_type_notation' declare type syntax
  2599 within a local theory context, with explicit checking of the
  2600 constructors involved (in contrast to the raw 'syntax' versions).
  2601 
  2602 * Commands 'types' and 'typedecl' now work within a local theory
  2603 context -- without introducing dependencies on parameters or
  2604 assumptions, which is not possible in Isabelle/Pure.
  2605 
  2606 * Command 'defaultsort' has been renamed to 'default_sort', it works
  2607 within a local theory context.  Minor INCOMPATIBILITY.
  2608 
  2609 
  2610 *** HOL ***
  2611 
  2612 * Command 'typedef' now works within a local theory context -- without
  2613 introducing dependencies on parameters or assumptions, which is not
  2614 possible in Isabelle/Pure/HOL.  Note that the logical environment may
  2615 contain multiple interpretations of local typedefs (with different
  2616 non-emptiness proofs), even in a global theory context.
  2617 
  2618 * New package for quotient types.  Commands 'quotient_type' and
  2619 'quotient_definition' may be used for defining types and constants by
  2620 quotient constructions.  An example is the type of integers created by
  2621 quotienting pairs of natural numbers:
  2622 
  2623   fun
  2624     intrel :: "(nat * nat) => (nat * nat) => bool"
  2625   where
  2626     "intrel (x, y) (u, v) = (x + v = u + y)"
  2627 
  2628   quotient_type int = "nat * nat" / intrel
  2629     by (auto simp add: equivp_def expand_fun_eq)
  2630 
  2631   quotient_definition
  2632     "0::int" is "(0::nat, 0::nat)"
  2633 
  2634 The method "lifting" can be used to lift of theorems from the
  2635 underlying "raw" type to the quotient type.  The example
  2636 src/HOL/Quotient_Examples/FSet.thy includes such a quotient
  2637 construction and provides a reasoning infrastructure for finite sets.
  2638 
  2639 * Renamed Library/Quotient.thy to Library/Quotient_Type.thy to avoid
  2640 clash with new theory Quotient in Main HOL.
  2641 
  2642 * Moved the SMT binding into the main HOL session, eliminating
  2643 separate HOL-SMT session.
  2644 
  2645 * List membership infix mem operation is only an input abbreviation.
  2646 INCOMPATIBILITY.
  2647 
  2648 * Theory Library/Word.thy has been removed.  Use library Word/Word.thy
  2649 for future developements; former Library/Word.thy is still present in
  2650 the AFP entry RSAPPS.
  2651 
  2652 * Theorem Int.int_induct renamed to Int.int_of_nat_induct and is no
  2653 longer shadowed.  INCOMPATIBILITY.
  2654 
  2655 * Dropped theorem duplicate comp_arith; use semiring_norm instead.
  2656 INCOMPATIBILITY.
  2657 
  2658 * Dropped theorem RealPow.real_sq_order; use power2_le_imp_le instead.
  2659 INCOMPATIBILITY.
  2660 
  2661 * Dropped normalizing_semiring etc; use the facts in semiring classes
  2662 instead.  INCOMPATIBILITY.
  2663 
  2664 * Dropped several real-specific versions of lemmas about floor and
  2665 ceiling; use the generic lemmas from theory "Archimedean_Field"
  2666 instead.  INCOMPATIBILITY.
  2667 
  2668   floor_number_of_eq         ~> floor_number_of
  2669   le_floor_eq_number_of      ~> number_of_le_floor
  2670   le_floor_eq_zero           ~> zero_le_floor
  2671   le_floor_eq_one            ~> one_le_floor
  2672   floor_less_eq_number_of    ~> floor_less_number_of
  2673   floor_less_eq_zero         ~> floor_less_zero
  2674   floor_less_eq_one          ~> floor_less_one
  2675   less_floor_eq_number_of    ~> number_of_less_floor
  2676   less_floor_eq_zero         ~> zero_less_floor
  2677   less_floor_eq_one          ~> one_less_floor
  2678   floor_le_eq_number_of      ~> floor_le_number_of
  2679   floor_le_eq_zero           ~> floor_le_zero
  2680   floor_le_eq_one            ~> floor_le_one
  2681   floor_subtract_number_of   ~> floor_diff_number_of
  2682   floor_subtract_one         ~> floor_diff_one
  2683   ceiling_number_of_eq       ~> ceiling_number_of
  2684   ceiling_le_eq_number_of    ~> ceiling_le_number_of
  2685   ceiling_le_zero_eq         ~> ceiling_le_zero
  2686   ceiling_le_eq_one          ~> ceiling_le_one
  2687   less_ceiling_eq_number_of  ~> number_of_less_ceiling
  2688   less_ceiling_eq_zero       ~> zero_less_ceiling
  2689   less_ceiling_eq_one        ~> one_less_ceiling
  2690   ceiling_less_eq_number_of  ~> ceiling_less_number_of
  2691   ceiling_less_eq_zero       ~> ceiling_less_zero
  2692   ceiling_less_eq_one        ~> ceiling_less_one
  2693   le_ceiling_eq_number_of    ~> number_of_le_ceiling
  2694   le_ceiling_eq_zero         ~> zero_le_ceiling
  2695   le_ceiling_eq_one          ~> one_le_ceiling
  2696   ceiling_subtract_number_of ~> ceiling_diff_number_of
  2697   ceiling_subtract_one       ~> ceiling_diff_one
  2698 
  2699 * Theory "Finite_Set": various folding_XXX locales facilitate the
  2700 application of the various fold combinators on finite sets.
  2701 
  2702 * Library theory "RBT" renamed to "RBT_Impl"; new library theory "RBT"
  2703 provides abstract red-black tree type which is backed by "RBT_Impl" as
  2704 implementation.  INCOMPATIBILTY.
  2705 
  2706 * Theory Library/Coinductive_List has been removed -- superseded by
  2707 AFP/thys/Coinductive.
  2708 
  2709 * Theory PReal, including the type "preal" and related operations, has
  2710 been removed.  INCOMPATIBILITY.
  2711 
  2712 * Real: new development using Cauchy Sequences.
  2713 
  2714 * Split off theory "Big_Operators" containing setsum, setprod,
  2715 Inf_fin, Sup_fin, Min, Max from theory Finite_Set.  INCOMPATIBILITY.
  2716 
  2717 * Theory "Rational" renamed to "Rat", for consistency with "Nat",
  2718 "Int" etc.  INCOMPATIBILITY.
  2719 
  2720 * Constant Rat.normalize needs to be qualified.  INCOMPATIBILITY.
  2721 
  2722 * New set of rules "ac_simps" provides combined assoc / commute
  2723 rewrites for all interpretations of the appropriate generic locales.
  2724 
  2725 * Renamed theory "OrderedGroup" to "Groups" and split theory
  2726 "Ring_and_Field" into theories "Rings" and "Fields"; for more
  2727 appropriate and more consistent names suitable for name prefixes
  2728 within the HOL theories.  INCOMPATIBILITY.
  2729 
  2730 * Some generic constants have been put to appropriate theories:
  2731   - less_eq, less: Orderings
  2732   - zero, one, plus, minus, uminus, times, abs, sgn: Groups
  2733   - inverse, divide: Rings
  2734 INCOMPATIBILITY.
  2735 
  2736 * More consistent naming of type classes involving orderings (and
  2737 lattices):
  2738 
  2739     lower_semilattice                   ~> semilattice_inf
  2740     upper_semilattice                   ~> semilattice_sup
  2741 
  2742     dense_linear_order                  ~> dense_linorder
  2743 
  2744     pordered_ab_group_add               ~> ordered_ab_group_add
  2745     pordered_ab_group_add_abs           ~> ordered_ab_group_add_abs
  2746     pordered_ab_semigroup_add           ~> ordered_ab_semigroup_add
  2747     pordered_ab_semigroup_add_imp_le    ~> ordered_ab_semigroup_add_imp_le
  2748     pordered_cancel_ab_semigroup_add    ~> ordered_cancel_ab_semigroup_add
  2749     pordered_cancel_comm_semiring       ~> ordered_cancel_comm_semiring
  2750     pordered_cancel_semiring            ~> ordered_cancel_semiring
  2751     pordered_comm_monoid_add            ~> ordered_comm_monoid_add
  2752     pordered_comm_ring                  ~> ordered_comm_ring
  2753     pordered_comm_semiring              ~> ordered_comm_semiring
  2754     pordered_ring                       ~> ordered_ring
  2755     pordered_ring_abs                   ~> ordered_ring_abs
  2756     pordered_semiring                   ~> ordered_semiring
  2757 
  2758     ordered_ab_group_add                ~> linordered_ab_group_add
  2759     ordered_ab_semigroup_add            ~> linordered_ab_semigroup_add
  2760     ordered_cancel_ab_semigroup_add     ~> linordered_cancel_ab_semigroup_add
  2761     ordered_comm_semiring_strict        ~> linordered_comm_semiring_strict
  2762     ordered_field                       ~> linordered_field
  2763     ordered_field_no_lb                 ~> linordered_field_no_lb
  2764     ordered_field_no_ub                 ~> linordered_field_no_ub
  2765     ordered_field_dense_linear_order    ~> dense_linordered_field
  2766     ordered_idom                        ~> linordered_idom
  2767     ordered_ring                        ~> linordered_ring
  2768     ordered_ring_le_cancel_factor       ~> linordered_ring_le_cancel_factor
  2769     ordered_ring_less_cancel_factor     ~> linordered_ring_less_cancel_factor
  2770     ordered_ring_strict                 ~> linordered_ring_strict
  2771     ordered_semidom                     ~> linordered_semidom
  2772     ordered_semiring                    ~> linordered_semiring
  2773     ordered_semiring_1                  ~> linordered_semiring_1
  2774     ordered_semiring_1_strict           ~> linordered_semiring_1_strict
  2775     ordered_semiring_strict             ~> linordered_semiring_strict
  2776 
  2777   The following slightly odd type classes have been moved to a
  2778   separate theory Library/Lattice_Algebras:
  2779 
  2780     lordered_ab_group_add               ~> lattice_ab_group_add
  2781     lordered_ab_group_add_abs           ~> lattice_ab_group_add_abs
  2782     lordered_ab_group_add_meet          ~> semilattice_inf_ab_group_add
  2783     lordered_ab_group_add_join          ~> semilattice_sup_ab_group_add
  2784     lordered_ring                       ~> lattice_ring
  2785 
  2786 INCOMPATIBILITY.
  2787 
  2788 * Refined field classes:
  2789   - classes division_ring_inverse_zero, field_inverse_zero,
  2790     linordered_field_inverse_zero include rule inverse 0 = 0 --
  2791     subsumes former division_by_zero class;
  2792   - numerous lemmas have been ported from field to division_ring.
  2793 INCOMPATIBILITY.
  2794 
  2795 * Refined algebra theorem collections:
  2796   - dropped theorem group group_simps, use algebra_simps instead;
  2797   - dropped theorem group ring_simps, use field_simps instead;
  2798   - proper theorem collection field_simps subsumes former theorem
  2799     groups field_eq_simps and field_simps;
  2800   - dropped lemma eq_minus_self_iff which is a duplicate for
  2801     equal_neg_zero.
  2802 INCOMPATIBILITY.
  2803 
  2804 * Theory Finite_Set and List: some lemmas have been generalized from
  2805 sets to lattices:
  2806 
  2807   fun_left_comm_idem_inter      ~> fun_left_comm_idem_inf
  2808   fun_left_comm_idem_union      ~> fun_left_comm_idem_sup
  2809   inter_Inter_fold_inter        ~> inf_Inf_fold_inf
  2810   union_Union_fold_union        ~> sup_Sup_fold_sup
  2811   Inter_fold_inter              ~> Inf_fold_inf
  2812   Union_fold_union              ~> Sup_fold_sup
  2813   inter_INTER_fold_inter        ~> inf_INFI_fold_inf
  2814   union_UNION_fold_union        ~> sup_SUPR_fold_sup
  2815   INTER_fold_inter              ~> INFI_fold_inf
  2816   UNION_fold_union              ~> SUPR_fold_sup
  2817 
  2818 * Theory "Complete_Lattice": lemmas top_def and bot_def have been
  2819 replaced by the more convenient lemmas Inf_empty and Sup_empty.
  2820 Dropped lemmas Inf_insert_simp and Sup_insert_simp, which are subsumed
  2821 by Inf_insert and Sup_insert.  Lemmas Inf_UNIV and Sup_UNIV replace
  2822 former Inf_Univ and Sup_Univ.  Lemmas inf_top_right and sup_bot_right
  2823 subsume inf_top and sup_bot respectively.  INCOMPATIBILITY.
  2824 
  2825 * Reorganized theory Multiset: swapped notation of pointwise and
  2826 multiset order:
  2827 
  2828   - pointwise ordering is instance of class order with standard syntax
  2829     <= and <;
  2830   - multiset ordering has syntax <=# and <#; partial order properties
  2831     are provided by means of interpretation with prefix
  2832     multiset_order;
  2833   - less duplication, less historical organization of sections,
  2834     conversion from associations lists to multisets, rudimentary code
  2835     generation;
  2836   - use insert_DiffM2 [symmetric] instead of elem_imp_eq_diff_union,
  2837     if needed.
  2838 
  2839 Renamed:
  2840 
  2841   multiset_eq_conv_count_eq  ~>  multiset_ext_iff
  2842   multi_count_ext  ~>  multiset_ext
  2843   diff_union_inverse2  ~>  diff_union_cancelR
  2844 
  2845 INCOMPATIBILITY.
  2846 
  2847 * Theory Permutation: replaced local "remove" by List.remove1.
  2848 
  2849 * Code generation: ML and OCaml code is decorated with signatures.
  2850 
  2851 * Theory List: added transpose.
  2852 
  2853 * Library/Nat_Bijection.thy is a collection of bijective functions
  2854 between nat and other types, which supersedes the older libraries
  2855 Library/Nat_Int_Bij.thy and HOLCF/NatIso.thy.  INCOMPATIBILITY.
  2856 
  2857   Constants:
  2858   Nat_Int_Bij.nat2_to_nat         ~> prod_encode
  2859   Nat_Int_Bij.nat_to_nat2         ~> prod_decode
  2860   Nat_Int_Bij.int_to_nat_bij      ~> int_encode
  2861   Nat_Int_Bij.nat_to_int_bij      ~> int_decode
  2862   Countable.pair_encode           ~> prod_encode
  2863   NatIso.prod2nat                 ~> prod_encode
  2864   NatIso.nat2prod                 ~> prod_decode
  2865   NatIso.sum2nat                  ~> sum_encode
  2866   NatIso.nat2sum                  ~> sum_decode
  2867   NatIso.list2nat                 ~> list_encode
  2868   NatIso.nat2list                 ~> list_decode
  2869   NatIso.set2nat                  ~> set_encode
  2870   NatIso.nat2set                  ~> set_decode
  2871 
  2872   Lemmas:
  2873   Nat_Int_Bij.bij_nat_to_int_bij  ~> bij_int_decode
  2874   Nat_Int_Bij.nat2_to_nat_inj     ~> inj_prod_encode
  2875   Nat_Int_Bij.nat2_to_nat_surj    ~> surj_prod_encode
  2876   Nat_Int_Bij.nat_to_nat2_inj     ~> inj_prod_decode
  2877   Nat_Int_Bij.nat_to_nat2_surj    ~> surj_prod_decode
  2878   Nat_Int_Bij.i2n_n2i_id          ~> int_encode_inverse
  2879   Nat_Int_Bij.n2i_i2n_id          ~> int_decode_inverse
  2880   Nat_Int_Bij.surj_nat_to_int_bij ~> surj_int_encode
  2881   Nat_Int_Bij.surj_int_to_nat_bij ~> surj_int_decode
  2882   Nat_Int_Bij.inj_nat_to_int_bij  ~> inj_int_encode
  2883   Nat_Int_Bij.inj_int_to_nat_bij  ~> inj_int_decode
  2884   Nat_Int_Bij.bij_nat_to_int_bij  ~> bij_int_encode
  2885   Nat_Int_Bij.bij_int_to_nat_bij  ~> bij_int_decode
  2886 
  2887 * Sledgehammer:
  2888   - Renamed ATP commands:
  2889     atp_info     ~> sledgehammer running_atps
  2890     atp_kill     ~> sledgehammer kill_atps
  2891     atp_messages ~> sledgehammer messages
  2892     atp_minimize ~> sledgehammer minimize
  2893     print_atps   ~> sledgehammer available_atps
  2894     INCOMPATIBILITY.
  2895   - Added user's manual ("isabelle doc sledgehammer").
  2896   - Added option syntax and "sledgehammer_params" to customize
  2897     Sledgehammer's behavior.  See the manual for details.
  2898   - Modified the Isar proof reconstruction code so that it produces
  2899     direct proofs rather than proofs by contradiction.  (This feature
  2900     is still experimental.)
  2901   - Made Isar proof reconstruction work for SPASS, remote ATPs, and in
  2902     full-typed mode.
  2903   - Added support for TPTP syntax for SPASS via the "spass_tptp" ATP.
  2904 
  2905 * Nitpick:
  2906   - Added and implemented "binary_ints" and "bits" options.
  2907   - Added "std" option and implemented support for nonstandard models.
  2908   - Added and implemented "finitize" option to improve the precision
  2909     of infinite datatypes based on a monotonicity analysis.
  2910   - Added support for quotient types.
  2911   - Added support for "specification" and "ax_specification"
  2912     constructs.
  2913   - Added support for local definitions (for "function" and
  2914     "termination" proofs).
  2915   - Added support for term postprocessors.
  2916   - Optimized "Multiset.multiset" and "FinFun.finfun".
  2917   - Improved efficiency of "destroy_constrs" optimization.
  2918   - Fixed soundness bugs related to "destroy_constrs" optimization and
  2919     record getters.
  2920   - Fixed soundness bug related to higher-order constructors.
  2921   - Fixed soundness bug when "full_descrs" is enabled.
  2922   - Improved precision of set constructs.
  2923   - Added "atoms" option.
  2924   - Added cache to speed up repeated Kodkod invocations on the same
  2925     problems.
  2926   - Renamed "MiniSatJNI", "zChaffJNI", "BerkMinAlloy", and
  2927     "SAT4JLight" to "MiniSat_JNI", "zChaff_JNI", "BerkMin_Alloy", and
  2928     "SAT4J_Light".  INCOMPATIBILITY.
  2929   - Removed "skolemize", "uncurry", "sym_break", "flatten_prop",
  2930     "sharing_depth", and "show_skolems" options.  INCOMPATIBILITY.
  2931   - Removed "nitpick_intro" attribute.  INCOMPATIBILITY.
  2932 
  2933 * Method "induct" now takes instantiations of the form t, where t is not
  2934   a variable, as a shorthand for "x == t", where x is a fresh variable.
  2935   If this is not intended, t has to be enclosed in parentheses.
  2936   By default, the equalities generated by definitional instantiations
  2937   are pre-simplified, which may cause parameters of inductive cases
  2938   to disappear, or may even delete some of the inductive cases.
  2939   Use "induct (no_simp)" instead of "induct" to restore the old
  2940   behaviour. The (no_simp) option is also understood by the "cases"
  2941   and "nominal_induct" methods, which now perform pre-simplification, too.
  2942   INCOMPATIBILITY.
  2943 
  2944 
  2945 *** HOLCF ***
  2946 
  2947 * Variable names in lemmas generated by the domain package have
  2948 changed; the naming scheme is now consistent with the HOL datatype
  2949 package.  Some proof scripts may be affected, INCOMPATIBILITY.
  2950 
  2951 * The domain package no longer defines the function "foo_copy" for
  2952 recursive domain "foo".  The reach lemma is now stated directly in
  2953 terms of "foo_take".  Lemmas and proofs that mention "foo_copy" must
  2954 be reformulated in terms of "foo_take", INCOMPATIBILITY.
  2955 
  2956 * Most definedness lemmas generated by the domain package (previously
  2957 of the form "x ~= UU ==> foo$x ~= UU") now have an if-and-only-if form
  2958 like "foo$x = UU <-> x = UU", which works better as a simp rule.
  2959 Proofs that used definedness lemmas as intro rules may break,
  2960 potential INCOMPATIBILITY.
  2961 
  2962 * Induction and casedist rules generated by the domain package now
  2963 declare proper case_names (one called "bottom", and one named for each
  2964 constructor).  INCOMPATIBILITY.
  2965 
  2966 * For mutually-recursive domains, separate "reach" and "take_lemma"
  2967 rules are generated for each domain, INCOMPATIBILITY.
  2968 
  2969   foo_bar.reach       ~> foo.reach  bar.reach
  2970   foo_bar.take_lemmas ~> foo.take_lemma  bar.take_lemma
  2971 
  2972 * Some lemmas generated by the domain package have been renamed for
  2973 consistency with the datatype package, INCOMPATIBILITY.
  2974 
  2975   foo.ind        ~> foo.induct
  2976   foo.finite_ind ~> foo.finite_induct
  2977   foo.coind      ~> foo.coinduct
  2978   foo.casedist   ~> foo.exhaust
  2979   foo.exhaust    ~> foo.nchotomy
  2980 
  2981 * For consistency with other definition packages, the fixrec package
  2982 now generates qualified theorem names, INCOMPATIBILITY.
  2983 
  2984   foo_simps  ~> foo.simps
  2985   foo_unfold ~> foo.unfold
  2986   foo_induct ~> foo.induct
  2987 
  2988 * The "fixrec_simp" attribute has been removed.  The "fixrec_simp"
  2989 method and internal fixrec proofs now use the default simpset instead.
  2990 INCOMPATIBILITY.
  2991 
  2992 * The "contlub" predicate has been removed.  Proof scripts should use
  2993 lemma contI2 in place of monocontlub2cont, INCOMPATIBILITY.
  2994 
  2995 * The "admw" predicate has been removed, INCOMPATIBILITY.
  2996 
  2997 * The constants cpair, cfst, and csnd have been removed in favor of
  2998 Pair, fst, and snd from Isabelle/HOL, INCOMPATIBILITY.
  2999 
  3000 
  3001 *** ML ***
  3002 
  3003 * Antiquotations for basic formal entities:
  3004 
  3005     @{class NAME}         -- type class
  3006     @{class_syntax NAME}  -- syntax representation of the above
  3007 
  3008     @{type_name NAME}     -- logical type
  3009     @{type_abbrev NAME}   -- type abbreviation
  3010     @{nonterminal NAME}   -- type of concrete syntactic category
  3011     @{type_syntax NAME}   -- syntax representation of any of the above
  3012 
  3013     @{const_name NAME}    -- logical constant (INCOMPATIBILITY)
  3014     @{const_abbrev NAME}  -- abbreviated constant
  3015     @{const_syntax NAME}  -- syntax representation of any of the above
  3016 
  3017 * Antiquotation @{syntax_const NAME} ensures that NAME refers to a raw
  3018 syntax constant (cf. 'syntax' command).
  3019 
  3020 * Antiquotation @{make_string} inlines a function to print arbitrary
  3021 values similar to the ML toplevel.  The result is compiler dependent
  3022 and may fall back on "?" in certain situations.
  3023 
  3024 * Diagnostic commands 'ML_val' and 'ML_command' may refer to
  3025 antiquotations @{Isar.state} and @{Isar.goal}.  This replaces impure
  3026 Isar.state() and Isar.goal(), which belong to the old TTY loop and do
  3027 not work with the asynchronous Isar document model.
  3028 
  3029 * Configuration options now admit dynamic default values, depending on
  3030 the context or even global references.
  3031 
  3032 * SHA1.digest digests strings according to SHA-1 (see RFC 3174).  It
  3033 uses an efficient external library if available (for Poly/ML).
  3034 
  3035 * Renamed some important ML structures, while keeping the old names
  3036 for some time as aliases within the structure Legacy:
  3037 
  3038   OuterKeyword  ~>  Keyword
  3039   OuterLex      ~>  Token
  3040   OuterParse    ~>  Parse
  3041   OuterSyntax   ~>  Outer_Syntax
  3042   PrintMode     ~>  Print_Mode
  3043   SpecParse     ~>  Parse_Spec
  3044   ThyInfo       ~>  Thy_Info
  3045   ThyLoad       ~>  Thy_Load
  3046   ThyOutput     ~>  Thy_Output
  3047   TypeInfer     ~>  Type_Infer
  3048 
  3049 Note that "open Legacy" simplifies porting of sources, but forgetting
  3050 to remove it again will complicate porting again in the future.
  3051 
  3052 * Most operations that refer to a global context are named
  3053 accordingly, e.g. Simplifier.global_context or
  3054 ProofContext.init_global.  There are some situations where a global
  3055 context actually works, but under normal circumstances one needs to
  3056 pass the proper local context through the code!
  3057 
  3058 * Discontinued old TheoryDataFun with its copy/init operation -- data
  3059 needs to be pure.  Functor Theory_Data_PP retains the traditional
  3060 Pretty.pp argument to merge, which is absent in the standard
  3061 Theory_Data version.
  3062 
  3063 * Sorts.certify_sort and derived "cert" operations for types and terms
  3064 no longer minimize sorts.  Thus certification at the boundary of the
  3065 inference kernel becomes invariant under addition of class relations,
  3066 which is an important monotonicity principle.  Sorts are now minimized
  3067 in the syntax layer only, at the boundary between the end-user and the
  3068 system.  Subtle INCOMPATIBILITY, may have to use Sign.minimize_sort
  3069 explicitly in rare situations.
  3070 
  3071 * Renamed old-style Drule.standard to Drule.export_without_context, to
  3072 emphasize that this is in no way a standard operation.
  3073 INCOMPATIBILITY.
  3074 
  3075 * Subgoal.FOCUS (and variants): resulting goal state is normalized as
  3076 usual for resolution.  Rare INCOMPATIBILITY.
  3077 
  3078 * Renamed varify/unvarify operations to varify_global/unvarify_global
  3079 to emphasize that these only work in a global situation (which is
  3080 quite rare).
  3081 
  3082 * Curried take and drop in library.ML; negative length is interpreted
  3083 as infinity (as in chop).  Subtle INCOMPATIBILITY.
  3084 
  3085 * Proof terms: type substitutions on proof constants now use canonical
  3086 order of type variables.  INCOMPATIBILITY for tools working with proof
  3087 terms.
  3088 
  3089 * Raw axioms/defs may no longer carry sort constraints, and raw defs
  3090 may no longer carry premises.  User-level specifications are
  3091 transformed accordingly by Thm.add_axiom/add_def.
  3092 
  3093 
  3094 *** System ***
  3095 
  3096 * Discontinued special HOL_USEDIR_OPTIONS for the main HOL image;
  3097 ISABELLE_USEDIR_OPTIONS applies uniformly to all sessions.  Note that
  3098 proof terms are enabled unconditionally in the new HOL-Proofs image.
  3099 
  3100 * Discontinued old ISABELLE and ISATOOL environment settings (legacy
  3101 feature since Isabelle2009).  Use ISABELLE_PROCESS and ISABELLE_TOOL,
  3102 respectively.
  3103 
  3104 * Old lib/scripts/polyml-platform is superseded by the
  3105 ISABELLE_PLATFORM setting variable, which defaults to the 32 bit
  3106 variant, even on a 64 bit machine.  The following example setting
  3107 prefers 64 bit if available:
  3108 
  3109   ML_PLATFORM="${ISABELLE_PLATFORM64:-$ISABELLE_PLATFORM}"
  3110 
  3111 * The preliminary Isabelle/jEdit application demonstrates the emerging
  3112 Isabelle/Scala layer for advanced prover interaction and integration.
  3113 See src/Tools/jEdit or "isabelle jedit" provided by the properly built
  3114 component.
  3115 
  3116 * "IsabelleText" is a Unicode font derived from Bitstream Vera Mono
  3117 and Bluesky TeX fonts.  It provides the usual Isabelle symbols,
  3118 similar to the default assignment of the document preparation system
  3119 (cf. isabellesym.sty).  The Isabelle/Scala class Isabelle_System
  3120 provides some operations for direct access to the font without asking
  3121 the user for manual installation.
  3122 
  3123 
  3124 
  3125 New in Isabelle2009-1 (December 2009)
  3126 -------------------------------------
  3127 
  3128 *** General ***
  3129 
  3130 * Discontinued old form of "escaped symbols" such as \\<forall>.  Only
  3131 one backslash should be used, even in ML sources.
  3132 
  3133 
  3134 *** Pure ***
  3135 
  3136 * Locale interpretation propagates mixins along the locale hierarchy.
  3137 The currently only available mixins are the equations used to map
  3138 local definitions to terms of the target domain of an interpretation.
  3139 
  3140 * Reactivated diagnostic command 'print_interps'.  Use "print_interps
  3141 loc" to print all interpretations of locale "loc" in the theory.
  3142 Interpretations in proofs are not shown.
  3143 
  3144 * Thoroughly revised locales tutorial.  New section on conditional
  3145 interpretation.
  3146 
  3147 * On instantiation of classes, remaining undefined class parameters
  3148 are formally declared.  INCOMPATIBILITY.
  3149 
  3150 
  3151 *** Document preparation ***
  3152 
  3153 * New generalized style concept for printing terms: @{foo (style) ...}
  3154 instead of @{foo_style style ...}  (old form is still retained for
  3155 backward compatibility).  Styles can be also applied for
  3156 antiquotations prop, term_type and typeof.
  3157 
  3158 
  3159 *** HOL ***
  3160 
  3161 * New proof method "smt" for a combination of first-order logic with
  3162 equality, linear and nonlinear (natural/integer/real) arithmetic, and
  3163 fixed-size bitvectors; there is also basic support for higher-order
  3164 features (esp. lambda abstractions).  It is an incomplete decision
  3165 procedure based on external SMT solvers using the oracle mechanism;
  3166 for the SMT solver Z3, this method is proof-producing.  Certificates
  3167 are provided to avoid calling the external solvers solely for
  3168 re-checking proofs.  Due to a remote SMT service there is no need for
  3169 installing SMT solvers locally.  See src/HOL/SMT.
  3170 
  3171 * New commands to load and prove verification conditions generated by
  3172 the Boogie program verifier or derived systems (e.g. the Verifying C
  3173 Compiler (VCC) or Spec#).  See src/HOL/Boogie.
  3174 
  3175 * New counterexample generator tool 'nitpick' based on the Kodkod
  3176 relational model finder.  See src/HOL/Tools/Nitpick and
  3177 src/HOL/Nitpick_Examples.
  3178 
  3179 * New commands 'code_pred' and 'values' to invoke the predicate
  3180 compiler and to enumerate values of inductive predicates.
  3181 
  3182 * A tabled implementation of the reflexive transitive closure.
  3183 
  3184 * New implementation of quickcheck uses generic code generator;
  3185 default generators are provided for all suitable HOL types, records
  3186 and datatypes.  Old quickcheck can be re-activated importing theory
  3187 Library/SML_Quickcheck.
  3188 
  3189 * New testing tool Mirabelle for automated proof tools.  Applies
  3190 several tools and tactics like sledgehammer, metis, or quickcheck, to
  3191 every proof step in a theory.  To be used in batch mode via the
  3192 "mirabelle" utility.
  3193 
  3194 * New proof method "sos" (sum of squares) for nonlinear real
  3195 arithmetic (originally due to John Harison). It requires theory
  3196 Library/Sum_Of_Squares.  It is not a complete decision procedure but
  3197 works well in practice on quantifier-free real arithmetic with +, -,
  3198 *, ^, =, <= and <, i.e. boolean combinations of equalities and
  3199 inequalities between polynomials.  It makes use of external
  3200 semidefinite programming solvers.  Method "sos" generates a
  3201 certificate that can be pasted into the proof thus avoiding the need
  3202 to call an external tool every time the proof is checked.  See
  3203 src/HOL/Library/Sum_Of_Squares.
  3204 
  3205 * New method "linarith" invokes existing linear arithmetic decision
  3206 procedure only.
  3207 
  3208 * New command 'atp_minimal' reduces result produced by Sledgehammer.
  3209 
  3210 * New Sledgehammer option "Full Types" in Proof General settings menu.
  3211 Causes full type information to be output to the ATPs.  This slows
  3212 ATPs down considerably but eliminates a source of unsound "proofs"
  3213 that fail later.
  3214 
  3215 * New method "metisFT": A version of metis that uses full type
  3216 information in order to avoid failures of proof reconstruction.
  3217 
  3218 * New evaluator "approximate" approximates an real valued term using
  3219 the same method as the approximation method.
  3220 
  3221 * Method "approximate" now supports arithmetic expressions as
  3222 boundaries of intervals and implements interval splitting and Taylor
  3223 series expansion.
  3224 
  3225 * ML antiquotation @{code_datatype} inserts definition of a datatype
  3226 generated by the code generator; e.g. see src/HOL/Predicate.thy.
  3227 
  3228 * New theory SupInf of the supremum and infimum operators for sets of
  3229 reals.
  3230 
  3231 * New theory Probability, which contains a development of measure
  3232 theory, eventually leading to Lebesgue integration and probability.
  3233 
  3234 * Extended Multivariate Analysis to include derivation and Brouwer's
  3235 fixpoint theorem.
  3236 
  3237 * Reorganization of number theory, INCOMPATIBILITY:
  3238   - new number theory development for nat and int, in theories Divides
  3239     and GCD as well as in new session Number_Theory
  3240   - some constants and facts now suffixed with _nat and _int
  3241     accordingly
  3242   - former session NumberTheory now named Old_Number_Theory, including
  3243     theories Legacy_GCD and Primes (prefer Number_Theory if possible)
  3244   - moved theory Pocklington from src/HOL/Library to
  3245     src/HOL/Old_Number_Theory
  3246 
  3247 * Theory GCD includes functions Gcd/GCD and Lcm/LCM for the gcd and
  3248 lcm of finite and infinite sets. It is shown that they form a complete
  3249 lattice.
  3250 
  3251 * Class semiring_div requires superclass no_zero_divisors and proof of
  3252 div_mult_mult1; theorems div_mult_mult1, div_mult_mult2,
  3253 div_mult_mult1_if, div_mult_mult1 and div_mult_mult2 have been
  3254 generalized to class semiring_div, subsuming former theorems
  3255 zdiv_zmult_zmult1, zdiv_zmult_zmult1_if, zdiv_zmult_zmult1 and
  3256 zdiv_zmult_zmult2.  div_mult_mult1 is now [simp] by default.
  3257 INCOMPATIBILITY.
  3258 
  3259 * Refinements to lattice classes and sets:
  3260   - less default intro/elim rules in locale variant, more default
  3261     intro/elim rules in class variant: more uniformity
  3262   - lemma ge_sup_conv renamed to le_sup_iff, in accordance with
  3263     le_inf_iff
  3264   - dropped lemma alias inf_ACI for inf_aci (same for sup_ACI and
  3265     sup_aci)
  3266   - renamed ACI to inf_sup_aci
  3267   - new class "boolean_algebra"
  3268   - class "complete_lattice" moved to separate theory
  3269     "Complete_Lattice"; corresponding constants (and abbreviations)
  3270     renamed and with authentic syntax:
  3271     Set.Inf ~>    Complete_Lattice.Inf
  3272     Set.Sup ~>    Complete_Lattice.Sup
  3273     Set.INFI ~>   Complete_Lattice.INFI
  3274     Set.SUPR ~>   Complete_Lattice.SUPR
  3275     Set.Inter ~>  Complete_Lattice.Inter
  3276     Set.Union ~>  Complete_Lattice.Union
  3277     Set.INTER ~>  Complete_Lattice.INTER
  3278     Set.UNION ~>  Complete_Lattice.UNION
  3279   - authentic syntax for
  3280     Set.Pow
  3281     Set.image
  3282   - mere abbreviations:
  3283     Set.empty               (for bot)
  3284     Set.UNIV                (for top)
  3285     Set.inter               (for inf, formerly Set.Int)
  3286     Set.union               (for sup, formerly Set.Un)
  3287     Complete_Lattice.Inter  (for Inf)
  3288     Complete_Lattice.Union  (for Sup)
  3289     Complete_Lattice.INTER  (for INFI)
  3290     Complete_Lattice.UNION  (for SUPR)
  3291   - object-logic definitions as far as appropriate
  3292 
  3293 INCOMPATIBILITY.  Care is required when theorems Int_subset_iff or
  3294 Un_subset_iff are explicitly deleted as default simp rules; then also
  3295 their lattice counterparts le_inf_iff and le_sup_iff have to be
  3296 deleted to achieve the desired effect.
  3297 
  3298 * Rules inf_absorb1, inf_absorb2, sup_absorb1, sup_absorb2 are no simp
  3299 rules by default any longer; the same applies to min_max.inf_absorb1
  3300 etc.  INCOMPATIBILITY.
  3301 
  3302 * Rules sup_Int_eq and sup_Un_eq are no longer declared as
  3303 pred_set_conv by default.  INCOMPATIBILITY.
  3304 
  3305 * Power operations on relations and functions are now one dedicated
  3306 constant "compow" with infix syntax "^^".  Power operation on
  3307 multiplicative monoids retains syntax "^" and is now defined generic
  3308 in class power.  INCOMPATIBILITY.
  3309 
  3310 * Relation composition "R O S" now has a more standard argument order:
  3311 "R O S = {(x, z). EX y. (x, y) : R & (y, z) : S}".  INCOMPATIBILITY,
  3312 rewrite propositions with "S O R" --> "R O S". Proofs may occasionally
  3313 break, since the O_assoc rule was not rewritten like this.  Fix using
  3314 O_assoc[symmetric].  The same applies to the curried version "R OO S".
  3315 
  3316 * Function "Inv" is renamed to "inv_into" and function "inv" is now an
  3317 abbreviation for "inv_into UNIV".  Lemmas are renamed accordingly.
  3318 INCOMPATIBILITY.
  3319 
  3320 * Most rules produced by inductive and datatype package have mandatory
  3321 prefixes.  INCOMPATIBILITY.
  3322 
  3323 * Changed "DERIV_intros" to a dynamic fact, which can be augmented by
  3324 the attribute of the same name.  Each of the theorems in the list
  3325 DERIV_intros assumes composition with an additional function and
  3326 matches a variable to the derivative, which has to be solved by the
  3327 Simplifier.  Hence (auto intro!: DERIV_intros) computes the derivative
  3328 of most elementary terms.  Former Maclauren.DERIV_tac and
  3329 Maclauren.deriv_tac should be replaced by (auto intro!: DERIV_intros).
  3330 INCOMPATIBILITY.
  3331 
  3332 * Code generator attributes follow the usual underscore convention:
  3333     code_unfold     replaces    code unfold
  3334     code_post       replaces    code post
  3335     etc.
  3336   INCOMPATIBILITY.
  3337 
  3338 * Renamed methods:
  3339     sizechange -> size_change
  3340     induct_scheme -> induction_schema
  3341   INCOMPATIBILITY.
  3342 
  3343 * Discontinued abbreviation "arbitrary" of constant "undefined".
  3344 INCOMPATIBILITY, use "undefined" directly.
  3345 
  3346 * Renamed theorems:
  3347     Suc_eq_add_numeral_1 -> Suc_eq_plus1
  3348     Suc_eq_add_numeral_1_left -> Suc_eq_plus1_left
  3349     Suc_plus1 -> Suc_eq_plus1
  3350     *anti_sym -> *antisym*
  3351     vector_less_eq_def -> vector_le_def
  3352   INCOMPATIBILITY.
  3353 
  3354 * Added theorem List.map_map as [simp].  Removed List.map_compose.
  3355 INCOMPATIBILITY.
  3356 
  3357 * Removed predicate "M hassize n" (<--> card M = n & finite M).
  3358 INCOMPATIBILITY.
  3359 
  3360 
  3361 *** HOLCF ***
  3362 
  3363 * Theory Representable defines a class "rep" of domains that are
  3364 representable (via an ep-pair) in the universal domain type "udom".
  3365 Instances are provided for all type constructors defined in HOLCF.
  3366 
  3367 * The 'new_domain' command is a purely definitional version of the
  3368 domain package, for representable domains.  Syntax is identical to the
  3369 old domain package.  The 'new_domain' package also supports indirect
  3370 recursion using previously-defined type constructors.  See
  3371 src/HOLCF/ex/New_Domain.thy for examples.
  3372 
  3373 * Method "fixrec_simp" unfolds one step of a fixrec-defined constant
  3374 on the left-hand side of an equation, and then performs
  3375 simplification.  Rewriting is done using rules declared with the
  3376 "fixrec_simp" attribute.  The "fixrec_simp" method is intended as a
  3377 replacement for "fixpat"; see src/HOLCF/ex/Fixrec_ex.thy for examples.
  3378 
  3379 * The pattern-match compiler in 'fixrec' can now handle constructors
  3380 with HOL function types.  Pattern-match combinators for the Pair
  3381 constructor are pre-configured.
  3382 
  3383 * The 'fixrec' package now produces better fixed-point induction rules
  3384 for mutually-recursive definitions:  Induction rules have conclusions
  3385 of the form "P foo bar" instead of "P <foo, bar>".
  3386 
  3387 * The constant "sq_le" (with infix syntax "<<" or "\<sqsubseteq>") has
  3388 been renamed to "below".  The name "below" now replaces "less" in many
  3389 theorem names.  (Legacy theorem names using "less" are still supported
  3390 as well.)
  3391 
  3392 * The 'fixrec' package now supports "bottom patterns".  Bottom
  3393 patterns can be used to generate strictness rules, or to make
  3394 functions more strict (much like the bang-patterns supported by the
  3395 Glasgow Haskell Compiler).  See src/HOLCF/ex/Fixrec_ex.thy for
  3396 examples.
  3397 
  3398 
  3399 *** ML ***
  3400 
  3401 * Support for Poly/ML 5.3.0, with improved reporting of compiler
  3402 errors and run-time exceptions, including detailed source positions.
  3403 
  3404 * Structure Name_Space (formerly NameSpace) now manages uniquely
  3405 identified entries, with some additional information such as source
  3406 position, logical grouping etc.
  3407 
  3408 * Theory and context data is now introduced by the simplified and
  3409 modernized functors Theory_Data, Proof_Data, Generic_Data.  Data needs
  3410 to be pure, but the old TheoryDataFun for mutable data (with explicit
  3411 copy operation) is still available for some time.
  3412 
  3413 * Structure Synchronized (cf. src/Pure/Concurrent/synchronized.ML)
  3414 provides a high-level programming interface to synchronized state
  3415 variables with atomic update.  This works via pure function
  3416 application within a critical section -- its runtime should be as
  3417 short as possible; beware of deadlocks if critical code is nested,
  3418 either directly or indirectly via other synchronized variables!
  3419 
  3420 * Structure Unsynchronized (cf. src/Pure/ML-Systems/unsynchronized.ML)
  3421 wraps raw ML references, explicitly indicating their non-thread-safe
  3422 behaviour.  The Isar toplevel keeps this structure open, to
  3423 accommodate Proof General as well as quick and dirty interactive
  3424 experiments with references.
  3425 
  3426 * PARALLEL_CHOICE and PARALLEL_GOALS provide basic support for
  3427 parallel tactical reasoning.
  3428 
  3429 * Tacticals Subgoal.FOCUS, Subgoal.FOCUS_PREMS, Subgoal.FOCUS_PARAMS
  3430 are similar to SUBPROOF, but are slightly more flexible: only the
  3431 specified parts of the subgoal are imported into the context, and the
  3432 body tactic may introduce new subgoals and schematic variables.
  3433 
  3434 * Old tactical METAHYPS, which does not observe the proof context, has
  3435 been renamed to Old_Goals.METAHYPS and awaits deletion.  Use SUBPROOF
  3436 or Subgoal.FOCUS etc.
  3437 
  3438 * Renamed functor TableFun to Table, and GraphFun to Graph.  (Since
  3439 functors have their own ML name space there is no point to mark them
  3440 separately.)  Minor INCOMPATIBILITY.
  3441 
  3442 * Renamed NamedThmsFun to Named_Thms.  INCOMPATIBILITY.
  3443 
  3444 * Renamed several structures FooBar to Foo_Bar.  Occasional,
  3445 INCOMPATIBILITY.
  3446 
  3447 * Operations of structure Skip_Proof no longer require quick_and_dirty
  3448 mode, which avoids critical setmp.
  3449 
  3450 * Eliminated old Attrib.add_attributes, Method.add_methods and related
  3451 combinators for "args".  INCOMPATIBILITY, need to use simplified
  3452 Attrib/Method.setup introduced in Isabelle2009.
  3453 
  3454 * Proper context for simpset_of, claset_of, clasimpset_of.  May fall
  3455 back on global_simpset_of, global_claset_of, global_clasimpset_of as
  3456 last resort.  INCOMPATIBILITY.
  3457 
  3458 * Display.pretty_thm now requires a proper context (cf. former
  3459 ProofContext.pretty_thm).  May fall back on Display.pretty_thm_global
  3460 or even Display.pretty_thm_without_context as last resort.
  3461 INCOMPATIBILITY.
  3462 
  3463 * Discontinued Display.pretty_ctyp/cterm etc.  INCOMPATIBILITY, use
  3464 Syntax.pretty_typ/term directly, preferably with proper context
  3465 instead of global theory.
  3466 
  3467 
  3468 *** System ***
  3469 
  3470 * Further fine tuning of parallel proof checking, scales up to 8 cores
  3471 (max. speedup factor 5.0).  See also Goal.parallel_proofs in ML and
  3472 usedir option -q.
  3473 
  3474 * Support for additional "Isabelle components" via etc/components, see
  3475 also the system manual.
  3476 
  3477 * The isabelle makeall tool now operates on all components with
  3478 IsaMakefile, not just hardwired "logics".
  3479 
  3480 * Removed "compress" option from isabelle-process and isabelle usedir;
  3481 this is always enabled.
  3482 
  3483 * Discontinued support for Poly/ML 4.x versions.
  3484 
  3485 * Isabelle tool "wwwfind" provides web interface for 'find_theorems'
  3486 on a given logic image.  This requires the lighttpd webserver and is
  3487 currently supported on Linux only.
  3488 
  3489 
  3490 
  3491 New in Isabelle2009 (April 2009)
  3492 --------------------------------
  3493 
  3494 *** General ***
  3495 
  3496 * Simplified main Isabelle executables, with less surprises on
  3497 case-insensitive file-systems (such as Mac OS).
  3498 
  3499   - The main Isabelle tool wrapper is now called "isabelle" instead of
  3500     "isatool."
  3501 
  3502   - The former "isabelle" alias for "isabelle-process" has been
  3503     removed (should rarely occur to regular users).
  3504 
  3505   - The former "isabelle-interface" and its alias "Isabelle" have been
  3506     removed (interfaces are now regular Isabelle tools).
  3507 
  3508 Within scripts and make files, the Isabelle environment variables
  3509 ISABELLE_TOOL and ISABELLE_PROCESS replace old ISATOOL and ISABELLE,
  3510 respectively.  (The latter are still available as legacy feature.)
  3511 
  3512 The old isabelle-interface wrapper could react in confusing ways if
  3513 the interface was uninstalled or changed otherwise.  Individual
  3514 interface tool configuration is now more explicit, see also the
  3515 Isabelle system manual.  In particular, Proof General is now available
  3516 via "isabelle emacs".
  3517 
  3518 INCOMPATIBILITY, need to adapt derivative scripts.  Users may need to
  3519 purge installed copies of Isabelle executables and re-run "isabelle
  3520 install -p ...", or use symlinks.
  3521 
  3522 * The default for ISABELLE_HOME_USER is now ~/.isabelle instead of the
  3523 old ~/isabelle, which was slightly non-standard and apt to cause
  3524 surprises on case-insensitive file-systems (such as Mac OS).
  3525 
  3526 INCOMPATIBILITY, need to move existing ~/isabelle/etc,
  3527 ~/isabelle/heaps, ~/isabelle/browser_info to the new place.  Special
  3528 care is required when using older releases of Isabelle.  Note that
  3529 ISABELLE_HOME_USER can be changed in Isabelle/etc/settings of any
  3530 Isabelle distribution, in order to use the new ~/.isabelle uniformly.
  3531 
  3532 * Proofs of fully specified statements are run in parallel on
  3533 multi-core systems.  A speedup factor of 2.5 to 3.2 can be expected on
  3534 a regular 4-core machine, if the initial heap space is made reasonably
  3535 large (cf. Poly/ML option -H).  (Requires Poly/ML 5.2.1 or later.)
  3536 
  3537 * The main reference manuals ("isar-ref", "implementation", and
  3538 "system") have been updated and extended.  Formally checked references
  3539 as hyperlinks are now available uniformly.
  3540 
  3541 
  3542 *** Pure ***
  3543 
  3544 * Complete re-implementation of locales.  INCOMPATIBILITY in several
  3545 respects.  The most important changes are listed below.  See the
  3546 Tutorial on Locales ("locales" manual) for details.
  3547 
  3548 - In locale expressions, instantiation replaces renaming.  Parameters
  3549 must be declared in a for clause.  To aid compatibility with previous
  3550 parameter inheritance, in locale declarations, parameters that are not
  3551 'touched' (instantiation position "_" or omitted) are implicitly added
  3552 with their syntax at the beginning of the for clause.
  3553 
  3554 - Syntax from abbreviations and definitions in locales is available in
  3555 locale expressions and context elements.  The latter is particularly
  3556 useful in locale declarations.
  3557 
  3558 - More flexible mechanisms to qualify names generated by locale
  3559 expressions.  Qualifiers (prefixes) may be specified in locale
  3560 expressions, and can be marked as mandatory (syntax: "name!:") or
  3561 optional (syntax "name?:").  The default depends for plain "name:"
  3562 depends on the situation where a locale expression is used: in
  3563 commands 'locale' and 'sublocale' prefixes are optional, in
  3564 'interpretation' and 'interpret' prefixes are mandatory.  The old
  3565 implicit qualifiers derived from the parameter names of a locale are
  3566 no longer generated.
  3567 
  3568 - Command "sublocale l < e" replaces "interpretation l < e".  The
  3569 instantiation clause in "interpretation" and "interpret" (square
  3570 brackets) is no longer available.  Use locale expressions.
  3571 
  3572 - When converting proof scripts, mandatory qualifiers in
  3573 'interpretation' and 'interpret' should be retained by default, even
  3574 if this is an INCOMPATIBILITY compared to former behavior.  In the
  3575 worst case, use the "name?:" form for non-mandatory ones.  Qualifiers
  3576 in locale expressions range over a single locale instance only.
  3577 
  3578 - Dropped locale element "includes".  This is a major INCOMPATIBILITY.
  3579 In existing theorem specifications replace the includes element by the
  3580 respective context elements of the included locale, omitting those
  3581 that are already present in the theorem specification.  Multiple
  3582 assume elements of a locale should be replaced by a single one
  3583 involving the locale predicate.  In the proof body, declarations (most
  3584 notably theorems) may be regained by interpreting the respective
  3585 locales in the proof context as required (command "interpret").
  3586 
  3587 If using "includes" in replacement of a target solely because the
  3588 parameter types in the theorem are not as general as in the target,
  3589 consider declaring a new locale with additional type constraints on
  3590 the parameters (context element "constrains").
  3591 
  3592 - Discontinued "locale (open)".  INCOMPATIBILITY.
  3593 
  3594 - Locale interpretation commands no longer attempt to simplify goal.
  3595 INCOMPATIBILITY: in rare situations the generated goal differs.  Use
  3596 methods intro_locales and unfold_locales to clarify.
  3597 
  3598 - Locale interpretation commands no longer accept interpretation
  3599 attributes.  INCOMPATIBILITY.
  3600 
  3601 * Class declaration: so-called "base sort" must not be given in import
  3602 list any longer, but is inferred from the specification.  Particularly
  3603 in HOL, write
  3604 
  3605     class foo = ...
  3606 
  3607 instead of
  3608 
  3609     class foo = type + ...
  3610 
  3611 * Class target: global versions of theorems stemming do not carry a
  3612 parameter prefix any longer.  INCOMPATIBILITY.
  3613 
  3614 * Class 'instance' command no longer accepts attached definitions.
  3615 INCOMPATIBILITY, use proper 'instantiation' target instead.
  3616 
  3617 * Recovered hiding of consts, which was accidentally broken in
  3618 Isabelle2007.  Potential INCOMPATIBILITY, ``hide const c'' really
  3619 makes c inaccessible; consider using ``hide (open) const c'' instead.
  3620 
  3621 * Slightly more coherent Pure syntax, with updated documentation in
  3622 isar-ref manual.  Removed locales meta_term_syntax and
  3623 meta_conjunction_syntax: TERM and &&& (formerly &&) are now permanent,
  3624 INCOMPATIBILITY in rare situations.  Note that &&& should not be used
  3625 directly in regular applications.
  3626 
  3627 * There is a new syntactic category "float_const" for signed decimal
  3628 fractions (e.g. 123.45 or -123.45).
  3629 
  3630 * Removed exotic 'token_translation' command.  INCOMPATIBILITY, use ML
  3631 interface with 'setup' command instead.
  3632 
  3633 * Command 'local_setup' is similar to 'setup', but operates on a local
  3634 theory context.
  3635 
  3636 * The 'axiomatization' command now only works within a global theory
  3637 context.  INCOMPATIBILITY.
  3638 
  3639 * Goal-directed proof now enforces strict proof irrelevance wrt. sort
  3640 hypotheses.  Sorts required in the course of reasoning need to be
  3641 covered by the constraints in the initial statement, completed by the
  3642 type instance information of the background theory.  Non-trivial sort
  3643 hypotheses, which rarely occur in practice, may be specified via
  3644 vacuous propositions of the form SORT_CONSTRAINT('a::c).  For example:
  3645 
  3646   lemma assumes "SORT_CONSTRAINT('a::empty)" shows False ...
  3647 
  3648 The result contains an implicit sort hypotheses as before --
  3649 SORT_CONSTRAINT premises are eliminated as part of the canonical rule
  3650 normalization.
  3651 
  3652 * Generalized Isar history, with support for linear undo, direct state
  3653 addressing etc.
  3654 
  3655 * Changed defaults for unify configuration options:
  3656 
  3657   unify_trace_bound = 50 (formerly 25)
  3658   unify_search_bound = 60 (formerly 30)
  3659 
  3660 * Different bookkeeping for code equations (INCOMPATIBILITY):
  3661 
  3662   a) On theory merge, the last set of code equations for a particular
  3663      constant is taken (in accordance with the policy applied by other
  3664      parts of the code generator framework).
  3665 
  3666   b) Code equations stemming from explicit declarations (e.g. code
  3667      attribute) gain priority over default code equations stemming
  3668      from definition, primrec, fun etc.
  3669 
  3670 * Keyword 'code_exception' now named 'code_abort'.  INCOMPATIBILITY.
  3671 
  3672 * Unified theorem tables for both code generators.  Thus [code
  3673 func] has disappeared and only [code] remains.  INCOMPATIBILITY.
  3674 
  3675 * Command 'find_consts' searches for constants based on type and name
  3676 patterns, e.g.
  3677 
  3678     find_consts "_ => bool"
  3679 
  3680 By default, matching is against subtypes, but it may be restricted to
  3681 the whole type.  Searching by name is possible.  Multiple queries are
  3682 conjunctive and queries may be negated by prefixing them with a
  3683 hyphen:
  3684 
  3685     find_consts strict: "_ => bool" name: "Int" -"int => int"
  3686 
  3687 * New 'find_theorems' criterion "solves" matches theorems that
  3688 directly solve the current goal (modulo higher-order unification).
  3689 
  3690 * Auto solve feature for main theorem statements: whenever a new goal
  3691 is stated, "find_theorems solves" is called; any theorems that could
  3692 solve the lemma directly are listed as part of the goal state.
  3693 Cf. associated options in Proof General Isabelle settings menu,
  3694 enabled by default, with reasonable timeout for pathological cases of
  3695 higher-order unification.
  3696 
  3697 
  3698 *** Document preparation ***
  3699 
  3700 * Antiquotation @{lemma} now imitates a regular terminal proof,
  3701 demanding keyword 'by' and supporting the full method expression
  3702 syntax just like the Isar command 'by'.
  3703 
  3704 
  3705 *** HOL ***
  3706 
  3707 * Integrated main parts of former image HOL-Complex with HOL.  Entry
  3708 points Main and Complex_Main remain as before.
  3709 
  3710 * Logic image HOL-Plain provides a minimal HOL with the most important
  3711 tools available (inductive, datatype, primrec, ...).  This facilitates
  3712 experimentation and tool development.  Note that user applications
  3713 (and library theories) should never refer to anything below theory
  3714 Main, as before.
  3715 
  3716 * Logic image HOL-Main stops at theory Main, and thus facilitates
  3717 experimentation due to shorter build times.
  3718 
  3719 * Logic image HOL-NSA contains theories of nonstandard analysis which
  3720 were previously part of former HOL-Complex.  Entry point Hyperreal
  3721 remains valid, but theories formerly using Complex_Main should now use
  3722 new entry point Hypercomplex.
  3723 
  3724 * Generic ATP manager for Sledgehammer, based on ML threads instead of
  3725 Posix processes.  Avoids potentially expensive forking of the ML
  3726 process.  New thread-based implementation also works on non-Unix
  3727 platforms (Cygwin).  Provers are no longer hardwired, but defined
  3728 within the theory via plain ML wrapper functions.  Basic Sledgehammer
  3729 commands are covered in the isar-ref manual.
  3730 
  3731 * Wrapper scripts for remote SystemOnTPTP service allows to use
  3732 sledgehammer without local ATP installation (Vampire etc.). Other
  3733 provers may be included via suitable ML wrappers, see also
  3734 src/HOL/ATP_Linkup.thy.
  3735 
  3736 * ATP selection (E/Vampire/Spass) is now via Proof General's settings
  3737 menu.
  3738 
  3739 * The metis method no longer fails because the theorem is too trivial
  3740 (contains the empty clause).
  3741 
  3742 * The metis method now fails in the usual manner, rather than raising
  3743 an exception, if it determines that it cannot prove the theorem.
  3744 
  3745 * Method "coherent" implements a prover for coherent logic (see also
  3746 src/Tools/coherent.ML).
  3747 
  3748 * Constants "undefined" and "default" replace "arbitrary".  Usually
  3749 "undefined" is the right choice to replace "arbitrary", though
  3750 logically there is no difference.  INCOMPATIBILITY.
  3751 
  3752 * Command "value" now integrates different evaluation mechanisms.  The
  3753 result of the first successful evaluation mechanism is printed.  In
  3754 square brackets a particular named evaluation mechanisms may be
  3755 specified (currently, [SML], [code] or [nbe]).  See further
  3756 src/HOL/ex/Eval_Examples.thy.
  3757 
  3758 * Normalization by evaluation now allows non-leftlinear equations.
  3759 Declare with attribute [code nbe].
  3760 
  3761 * Methods "case_tac" and "induct_tac" now refer to the very same rules
  3762 as the structured Isar versions "cases" and "induct", cf. the
  3763 corresponding "cases" and "induct" attributes.  Mutual induction rules
  3764 are now presented as a list of individual projections
  3765 (e.g. foo_bar.inducts for types foo and bar); the old format with
  3766 explicit HOL conjunction is no longer supported.  INCOMPATIBILITY, in
  3767 rare situations a different rule is selected --- notably nested tuple
  3768 elimination instead of former prod.exhaust: use explicit (case_tac t
  3769 rule: prod.exhaust) here.
  3770 
  3771 * Attributes "cases", "induct", "coinduct" support "del" option.
  3772 
  3773 * Removed fact "case_split_thm", which duplicates "case_split".
  3774 
  3775 * The option datatype has been moved to a new theory Option.  Renamed
  3776 option_map to Option.map, and o2s to Option.set, INCOMPATIBILITY.
  3777 
  3778 * New predicate "strict_mono" classifies strict functions on partial
  3779 orders.  With strict functions on linear orders, reasoning about
  3780 (in)equalities is facilitated by theorems "strict_mono_eq",
  3781 "strict_mono_less_eq" and "strict_mono_less".
  3782 
  3783 * Some set operations are now proper qualified constants with
  3784 authentic syntax.  INCOMPATIBILITY:
  3785 
  3786     op Int ~>   Set.Int
  3787     op Un ~>    Set.Un
  3788     INTER ~>    Set.INTER
  3789     UNION ~>    Set.UNION
  3790     Inter ~>    Set.Inter
  3791     Union ~>    Set.Union
  3792     {} ~>       Set.empty
  3793     UNIV ~>     Set.UNIV
  3794 
  3795 * Class complete_lattice with operations Inf, Sup, INFI, SUPR now in
  3796 theory Set.
  3797 
  3798 * Auxiliary class "itself" has disappeared -- classes without any
  3799 parameter are treated as expected by the 'class' command.
  3800 
  3801 * Leibnitz's Series for Pi and the arcus tangens and logarithm series.
  3802 
  3803 * Common decision procedures (Cooper, MIR, Ferrack, Approximation,
  3804 Dense_Linear_Order) are now in directory HOL/Decision_Procs.
  3805 
  3806 * Theory src/HOL/Decision_Procs/Approximation provides the new proof
  3807 method "approximation".  It proves formulas on real values by using
  3808 interval arithmetic.  In the formulas are also the transcendental
  3809 functions sin, cos, tan, atan, ln, exp and the constant pi are
  3810 allowed. For examples see
  3811 src/HOL/Descision_Procs/ex/Approximation_Ex.thy.
  3812 
  3813 * Theory "Reflection" now resides in HOL/Library.
  3814 
  3815 * Entry point to Word library now simply named "Word".
  3816 INCOMPATIBILITY.
  3817 
  3818 * Made source layout more coherent with logical distribution
  3819 structure:
  3820 
  3821     src/HOL/Library/RType.thy ~> src/HOL/Typerep.thy
  3822     src/HOL/Library/Code_Message.thy ~> src/HOL/
  3823     src/HOL/Library/GCD.thy ~> src/HOL/
  3824     src/HOL/Library/Order_Relation.thy ~> src/HOL/
  3825     src/HOL/Library/Parity.thy ~> src/HOL/
  3826     src/HOL/Library/Univ_Poly.thy ~> src/HOL/
  3827     src/HOL/Real/ContNotDenum.thy ~> src/HOL/Library/
  3828     src/HOL/Real/Lubs.thy ~> src/HOL/
  3829     src/HOL/Real/PReal.thy ~> src/HOL/
  3830     src/HOL/Real/Rational.thy ~> src/HOL/
  3831     src/HOL/Real/RComplete.thy ~> src/HOL/
  3832     src/HOL/Real/RealDef.thy ~> src/HOL/
  3833     src/HOL/Real/RealPow.thy ~> src/HOL/
  3834     src/HOL/Real/Real.thy ~> src/HOL/
  3835     src/HOL/Complex/Complex_Main.thy ~> src/HOL/
  3836     src/HOL/Complex/Complex.thy ~> src/HOL/
  3837     src/HOL/Complex/FrechetDeriv.thy ~> src/HOL/Library/
  3838     src/HOL/Complex/Fundamental_Theorem_Algebra.thy ~> src/HOL/Library/
  3839     src/HOL/Hyperreal/Deriv.thy ~> src/HOL/
  3840     src/HOL/Hyperreal/Fact.thy ~> src/HOL/
  3841     src/HOL/Hyperreal/Integration.thy ~> src/HOL/
  3842     src/HOL/Hyperreal/Lim.thy ~> src/HOL/
  3843     src/HOL/Hyperreal/Ln.thy ~> src/HOL/
  3844     src/HOL/Hyperreal/Log.thy ~> src/HOL/
  3845     src/HOL/Hyperreal/MacLaurin.thy ~> src/HOL/
  3846     src/HOL/Hyperreal/NthRoot.thy ~> src/HOL/
  3847     src/HOL/Hyperreal/Series.thy ~> src/HOL/
  3848     src/HOL/Hyperreal/SEQ.thy ~> src/HOL/
  3849     src/HOL/Hyperreal/Taylor.thy ~> src/HOL/
  3850     src/HOL/Hyperreal/Transcendental.thy ~> src/HOL/
  3851     src/HOL/Real/Float ~> src/HOL/Library/
  3852     src/HOL/Real/HahnBanach ~> src/HOL/HahnBanach
  3853     src/HOL/Real/RealVector.thy ~> src/HOL/
  3854 
  3855     src/HOL/arith_data.ML ~> src/HOL/Tools
  3856     src/HOL/hologic.ML ~> src/HOL/Tools
  3857     src/HOL/simpdata.ML ~> src/HOL/Tools
  3858     src/HOL/int_arith1.ML ~> src/HOL/Tools/int_arith.ML
  3859     src/HOL/int_factor_simprocs.ML ~> src/HOL/Tools
  3860     src/HOL/nat_simprocs.ML ~> src/HOL/Tools
  3861     src/HOL/Real/float_arith.ML ~> src/HOL/Tools
  3862     src/HOL/Real/float_syntax.ML ~> src/HOL/Tools
  3863     src/HOL/Real/rat_arith.ML ~> src/HOL/Tools
  3864     src/HOL/Real/real_arith.ML ~> src/HOL/Tools
  3865 
  3866     src/HOL/Library/Array.thy ~> src/HOL/Imperative_HOL
  3867     src/HOL/Library/Heap_Monad.thy ~> src/HOL/Imperative_HOL
  3868     src/HOL/Library/Heap.thy ~> src/HOL/Imperative_HOL
  3869     src/HOL/Library/Imperative_HOL.thy ~> src/HOL/Imperative_HOL
  3870     src/HOL/Library/Ref.thy ~> src/HOL/Imperative_HOL
  3871     src/HOL/Library/Relational.thy ~> src/HOL/Imperative_HOL
  3872 
  3873 * If methods "eval" and "evaluation" encounter a structured proof
  3874 state with !!/==>, only the conclusion is evaluated to True (if
  3875 possible), avoiding strange error messages.
  3876 
  3877 * Method "sizechange" automates termination proofs using (a
  3878 modification of) the size-change principle.  Requires SAT solver.  See
  3879 src/HOL/ex/Termination.thy for examples.
  3880 
  3881 * Simplifier: simproc for let expressions now unfolds if bound
  3882 variable occurs at most once in let expression body.  INCOMPATIBILITY.
  3883 
  3884 * Method "arith": Linear arithmetic now ignores all inequalities when
  3885 fast_arith_neq_limit is exceeded, instead of giving up entirely.
  3886 
  3887 * New attribute "arith" for facts that should always be used
  3888 automatically by arithmetic. It is intended to be used locally in
  3889 proofs, e.g.
  3890 
  3891   assumes [arith]: "x > 0"
  3892 
  3893 Global usage is discouraged because of possible performance impact.
  3894 
  3895 * New classes "top" and "bot" with corresponding operations "top" and
  3896 "bot" in theory Orderings; instantiation of class "complete_lattice"
  3897 requires instantiation of classes "top" and "bot".  INCOMPATIBILITY.
  3898 
  3899 * Changed definition lemma "less_fun_def" in order to provide an
  3900 instance for preorders on functions; use lemma "less_le" instead.
  3901 INCOMPATIBILITY.
  3902 
  3903 * Theory Orderings: class "wellorder" moved here, with explicit
  3904 induction rule "less_induct" as assumption.  For instantiation of
  3905 "wellorder" by means of predicate "wf", use rule wf_wellorderI.
  3906 INCOMPATIBILITY.
  3907 
  3908 * Theory Orderings: added class "preorder" as superclass of "order".
  3909 INCOMPATIBILITY: Instantiation proofs for order, linorder
  3910 etc. slightly changed.  Some theorems named order_class.* now named
  3911 preorder_class.*.
  3912 
  3913 * Theory Relation: renamed "refl" to "refl_on", "reflexive" to "refl,
  3914 "diag" to "Id_on".
  3915 
  3916 * Theory Finite_Set: added a new fold combinator of type
  3917 
  3918   ('a => 'b => 'b) => 'b => 'a set => 'b
  3919 
  3920 Occasionally this is more convenient than the old fold combinator
  3921 which is now defined in terms of the new one and renamed to
  3922 fold_image.
  3923 
  3924 * Theories Ring_and_Field and OrderedGroup: The lemmas "group_simps"
  3925 and "ring_simps" have been replaced by "algebra_simps" (which can be
  3926 extended with further lemmas!).  At the moment both still exist but
  3927 the former will disappear at some point.
  3928 
  3929 * Theory Power: Lemma power_Suc is now declared as a simp rule in
  3930 class recpower.  Type-specific simp rules for various recpower types
  3931 have been removed.  INCOMPATIBILITY, rename old lemmas as follows:
  3932 
  3933 rat_power_0    -> power_0
  3934 rat_power_Suc  -> power_Suc
  3935 realpow_0      -> power_0
  3936 realpow_Suc    -> power_Suc
  3937 complexpow_0   -> power_0
  3938 complexpow_Suc -> power_Suc
  3939 power_poly_0   -> power_0
  3940 power_poly_Suc -> power_Suc
  3941 
  3942 * Theories Ring_and_Field and Divides: Definition of "op dvd" has been
  3943 moved to separate class dvd in Ring_and_Field; a couple of lemmas on
  3944 dvd has been generalized to class comm_semiring_1.  Likewise a bunch
  3945 of lemmas from Divides has been generalized from nat to class
  3946 semiring_div.  INCOMPATIBILITY.  This involves the following theorem
  3947 renames resulting from duplicate elimination:
  3948 
  3949     dvd_def_mod ~>          dvd_eq_mod_eq_0
  3950     zero_dvd_iff ~>         dvd_0_left_iff
  3951     dvd_0 ~>                dvd_0_right
  3952     DIVISION_BY_ZERO_DIV ~> div_by_0
  3953     DIVISION_BY_ZERO_MOD ~> mod_by_0
  3954     mult_div ~>             div_mult_self2_is_id
  3955     mult_mod ~>             mod_mult_self2_is_0
  3956 
  3957 * Theory IntDiv: removed many lemmas that are instances of class-based
  3958 generalizations (from Divides and Ring_and_Field).  INCOMPATIBILITY,
  3959 rename old lemmas as follows:
  3960 
  3961 dvd_diff               -> nat_dvd_diff
  3962 dvd_zminus_iff         -> dvd_minus_iff
  3963 mod_add1_eq            -> mod_add_eq
  3964 mod_mult1_eq           -> mod_mult_right_eq
  3965 mod_mult1_eq'          -> mod_mult_left_eq
  3966 mod_mult_distrib_mod   -> mod_mult_eq
  3967 nat_mod_add_left_eq    -> mod_add_left_eq
  3968 nat_mod_add_right_eq   -> mod_add_right_eq
  3969 nat_mod_div_trivial    -> mod_div_trivial
  3970 nat_mod_mod_trivial    -> mod_mod_trivial
  3971 zdiv_zadd_self1        -> div_add_self1
  3972 zdiv_zadd_self2        -> div_add_self2
  3973 zdiv_zmult_self1       -> div_mult_self2_is_id
  3974 zdiv_zmult_self2       -> div_mult_self1_is_id
  3975 zdvd_triv_left         -> dvd_triv_left
  3976 zdvd_triv_right        -> dvd_triv_right
  3977 zdvd_zmult_cancel_disj -> dvd_mult_cancel_left
  3978 zmod_eq0_zdvd_iff      -> dvd_eq_mod_eq_0[symmetric]
  3979 zmod_zadd_left_eq      -> mod_add_left_eq
  3980 zmod_zadd_right_eq     -> mod_add_right_eq
  3981 zmod_zadd_self1        -> mod_add_self1
  3982 zmod_zadd_self2        -> mod_add_self2
  3983 zmod_zadd1_eq          -> mod_add_eq
  3984 zmod_zdiff1_eq         -> mod_diff_eq
  3985 zmod_zdvd_zmod         -> mod_mod_cancel
  3986 zmod_zmod_cancel       -> mod_mod_cancel
  3987 zmod_zmult_self1       -> mod_mult_self2_is_0
  3988 zmod_zmult_self2       -> mod_mult_self1_is_0
  3989 zmod_1                 -> mod_by_1
  3990 zdiv_1                 -> div_by_1
  3991 zdvd_abs1              -> abs_dvd_iff
  3992 zdvd_abs2              -> dvd_abs_iff
  3993 zdvd_refl              -> dvd_refl
  3994 zdvd_trans             -> dvd_trans
  3995 zdvd_zadd              -> dvd_add
  3996 zdvd_zdiff             -> dvd_diff
  3997 zdvd_zminus_iff        -> dvd_minus_iff
  3998 zdvd_zminus2_iff       -> minus_dvd_iff
  3999 zdvd_zmultD            -> dvd_mult_right
  4000 zdvd_zmultD2           -> dvd_mult_left
  4001 zdvd_zmult_mono        -> mult_dvd_mono
  4002 zdvd_0_right           -> dvd_0_right
  4003 zdvd_0_left            -> dvd_0_left_iff
  4004 zdvd_1_left            -> one_dvd
  4005 zminus_dvd_iff         -> minus_dvd_iff
  4006 
  4007 * Theory Rational: 'Fract k 0' now equals '0'.  INCOMPATIBILITY.
  4008 
  4009 * The real numbers offer decimal input syntax: 12.34 is translated
  4010 into 1234/10^2. This translation is not reversed upon output.
  4011 
  4012 * Theory Library/Polynomial defines an abstract type 'a poly of
  4013 univariate polynomials with coefficients of type 'a.  In addition to
  4014 the standard ring operations, it also supports div and mod.  Code
  4015 generation is also supported, using list-style constructors.
  4016 
  4017 * Theory Library/Inner_Product defines a class of real_inner for real
  4018 inner product spaces, with an overloaded operation inner :: 'a => 'a
  4019 => real.  Class real_inner is a subclass of real_normed_vector from
  4020 theory RealVector.
  4021 
  4022 * Theory Library/Product_Vector provides instances for the product
  4023 type 'a * 'b of several classes from RealVector and Inner_Product.
  4024 Definitions of addition, subtraction, scalar multiplication, norms,
  4025 and inner products are included.
  4026 
  4027 * Theory Library/Bit defines the field "bit" of integers modulo 2.  In
  4028 addition to the field operations, numerals and case syntax are also
  4029 supported.
  4030 
  4031 * Theory Library/Diagonalize provides constructive version of Cantor's
  4032 first diagonalization argument.
  4033 
  4034 * Theory Library/GCD: Curried operations gcd, lcm (for nat) and zgcd,
  4035 zlcm (for int); carried together from various gcd/lcm developements in
  4036 the HOL Distribution.  Constants zgcd and zlcm replace former igcd and
  4037 ilcm; corresponding theorems renamed accordingly.  INCOMPATIBILITY,
  4038 may recover tupled syntax as follows:
  4039 
  4040     hide (open) const gcd
  4041     abbreviation gcd where
  4042       "gcd == (%(a, b). GCD.gcd a b)"
  4043     notation (output)
  4044       GCD.gcd ("gcd '(_, _')")
  4045 
  4046 The same works for lcm, zgcd, zlcm.
  4047 
  4048 * Theory Library/Nat_Infinity: added addition, numeral syntax and more
  4049 instantiations for algebraic structures.  Removed some duplicate
  4050 theorems.  Changes in simp rules.  INCOMPATIBILITY.
  4051 
  4052 * ML antiquotation @{code} takes a constant as argument and generates
  4053 corresponding code in background and inserts name of the corresponding
  4054 resulting ML value/function/datatype constructor binding in place.
  4055 All occurrences of @{code} with a single ML block are generated
  4056 simultaneously.  Provides a generic and safe interface for
  4057 instrumentalizing code generation.  See
  4058 src/HOL/Decision_Procs/Ferrack.thy for a more ambitious application.
  4059 In future you ought to refrain from ad-hoc compiling generated SML
  4060 code on the ML toplevel.  Note that (for technical reasons) @{code}
  4061 cannot refer to constants for which user-defined serializations are
  4062 set.  Refer to the corresponding ML counterpart directly in that
  4063 cases.
  4064 
  4065 * Command 'rep_datatype': instead of theorem names the command now
  4066 takes a list of terms denoting the constructors of the type to be
  4067 represented as datatype.  The characteristic theorems have to be
  4068 proven.  INCOMPATIBILITY.  Also observe that the following theorems
  4069 have disappeared in favour of existing ones:
  4070 
  4071     unit_induct                 ~> unit.induct
  4072     prod_induct                 ~> prod.induct
  4073     sum_induct                  ~> sum.induct
  4074     Suc_Suc_eq                  ~> nat.inject
  4075     Suc_not_Zero Zero_not_Suc   ~> nat.distinct
  4076 
  4077 
  4078 *** HOL-Algebra ***
  4079 
  4080 * New locales for orders and lattices where the equivalence relation
  4081 is not restricted to equality.  INCOMPATIBILITY: all order and lattice
  4082 locales use a record structure with field eq for the equivalence.
  4083 
  4084 * New theory of factorial domains.
  4085 
  4086 * Units_l_inv and Units_r_inv are now simp rules by default.
  4087 INCOMPATIBILITY.  Simplifier proof that require deletion of l_inv
  4088 and/or r_inv will now also require deletion of these lemmas.
  4089 
  4090 * Renamed the following theorems, INCOMPATIBILITY:
  4091 
  4092 UpperD ~> Upper_memD
  4093 LowerD ~> Lower_memD
  4094 least_carrier ~> least_closed
  4095 greatest_carrier ~> greatest_closed
  4096 greatest_Lower_above ~> greatest_Lower_below
  4097 one_zero ~> carrier_one_zero
  4098 one_not_zero ~> carrier_one_not_zero  (collision with assumption)
  4099 
  4100 
  4101 *** HOL-Nominal ***
  4102 
  4103 * Nominal datatypes can now contain type-variables.
  4104 
  4105 * Commands 'nominal_inductive' and 'equivariance' work with local
  4106 theory targets.
  4107 
  4108 * Nominal primrec can now works with local theory targets and its
  4109 specification syntax now conforms to the general format as seen in
  4110 'inductive' etc.
  4111 
  4112 * Method "perm_simp" honours the standard simplifier attributes
  4113 (no_asm), (no_asm_use) etc.
  4114 
  4115 * The new predicate #* is defined like freshness, except that on the
  4116 left hand side can be a set or list of atoms.
  4117 
  4118 * Experimental command 'nominal_inductive2' derives strong induction
  4119 principles for inductive definitions.  In contrast to
  4120 'nominal_inductive', which can only deal with a fixed number of
  4121 binders, it can deal with arbitrary expressions standing for sets of
  4122 atoms to be avoided.  The only inductive definition we have at the
  4123 moment that needs this generalisation is the typing rule for Lets in
  4124 the algorithm W:
  4125 
  4126  Gamma |- t1 : T1   (x,close Gamma T1)::Gamma |- t2 : T2   x#Gamma
  4127  -----------------------------------------------------------------
  4128          Gamma |- Let x be t1 in t2 : T2
  4129 
  4130 In this rule one wants to avoid all the binders that are introduced by
  4131 "close Gamma T1".  We are looking for other examples where this
  4132 feature might be useful.  Please let us know.
  4133 
  4134 
  4135 *** HOLCF ***
  4136 
  4137 * Reimplemented the simplification procedure for proving continuity
  4138 subgoals.  The new simproc is extensible; users can declare additional
  4139 continuity introduction rules with the attribute [cont2cont].
  4140 
  4141 * The continuity simproc now uses a different introduction rule for
  4142 solving continuity subgoals on terms with lambda abstractions.  In
  4143 some rare cases the new simproc may fail to solve subgoals that the
  4144 old one could solve, and "simp add: cont2cont_LAM" may be necessary.
  4145 Potential INCOMPATIBILITY.
  4146 
  4147 * Command 'fixrec': specification syntax now conforms to the general
  4148 format as seen in 'inductive' etc.  See src/HOLCF/ex/Fixrec_ex.thy for
  4149 examples.  INCOMPATIBILITY.
  4150 
  4151 
  4152 *** ZF ***
  4153 
  4154 * Proof of Zorn's Lemma for partial orders.
  4155 
  4156 
  4157 *** ML ***
  4158 
  4159 * Multithreading for Poly/ML 5.1/5.2 is no longer supported, only for
  4160 Poly/ML 5.2.1 or later.  Important note: the TimeLimit facility
  4161 depends on multithreading, so timouts will not work before Poly/ML
  4162 5.2.1!
  4163 
  4164 * High-level support for concurrent ML programming, see
  4165 src/Pure/Cuncurrent.  The data-oriented model of "future values" is
  4166 particularly convenient to organize independent functional
  4167 computations.  The concept of "synchronized variables" provides a
  4168 higher-order interface for components with shared state, avoiding the
  4169 delicate details of mutexes and condition variables.  (Requires
  4170 Poly/ML 5.2.1 or later.)
  4171 
  4172 * ML bindings produced via Isar commands are stored within the Isar
  4173 context (theory or proof).  Consequently, commands like 'use' and 'ML'
  4174 become thread-safe and work with undo as expected (concerning
  4175 top-level bindings, not side-effects on global references).
  4176 INCOMPATIBILITY, need to provide proper Isar context when invoking the
  4177 compiler at runtime; really global bindings need to be given outside a
  4178 theory.  (Requires Poly/ML 5.2 or later.)
  4179 
  4180 * Command 'ML_prf' is analogous to 'ML' but works within a proof
  4181 context.  Top-level ML bindings are stored within the proof context in
  4182 a purely sequential fashion, disregarding the nested proof structure.
  4183 ML bindings introduced by 'ML_prf' are discarded at the end of the
  4184 proof.  (Requires Poly/ML 5.2 or later.)
  4185 
  4186 * Simplified ML attribute and method setup, cf. functions Attrib.setup
  4187 and Method.setup, as well as Isar commands 'attribute_setup' and
  4188 'method_setup'.  INCOMPATIBILITY for 'method_setup', need to simplify
  4189 existing code accordingly, or use plain 'setup' together with old
  4190 Method.add_method.
  4191 
  4192 * Simplified ML oracle interface Thm.add_oracle promotes 'a -> cterm
  4193 to 'a -> thm, while results are always tagged with an authentic oracle
  4194 name.  The Isar command 'oracle' is now polymorphic, no argument type
  4195 is specified.  INCOMPATIBILITY, need to simplify existing oracle code
  4196 accordingly.  Note that extra performance may be gained by producing
  4197 the cterm carefully, avoiding slow Thm.cterm_of.
  4198 
  4199 * Simplified interface for defining document antiquotations via
  4200 ThyOutput.antiquotation, ThyOutput.output, and optionally
  4201 ThyOutput.maybe_pretty_source.  INCOMPATIBILITY, need to simplify user
  4202 antiquotations accordingly, see src/Pure/Thy/thy_output.ML for common
  4203 examples.
  4204 
  4205 * More systematic treatment of long names, abstract name bindings, and
  4206 name space operations.  Basic operations on qualified names have been
  4207 move from structure NameSpace to Long_Name, e.g. Long_Name.base_name,
  4208 Long_Name.append.  Old type bstring has been mostly replaced by
  4209 abstract type binding (see structure Binding), which supports precise
  4210 qualification by packages and local theory targets, as well as proper
  4211 tracking of source positions.  INCOMPATIBILITY, need to wrap old
  4212 bstring values into Binding.name, or better pass through abstract
  4213 bindings everywhere.  See further src/Pure/General/long_name.ML,
  4214 src/Pure/General/binding.ML and src/Pure/General/name_space.ML
  4215 
  4216 * Result facts (from PureThy.note_thms, ProofContext.note_thms,
  4217 LocalTheory.note etc.) now refer to the *full* internal name, not the
  4218 bstring as before.  INCOMPATIBILITY, not detected by ML type-checking!
  4219 
  4220 * Disposed old type and term read functions (Sign.read_def_typ,
  4221 Sign.read_typ, Sign.read_def_terms, Sign.read_term,
  4222 Thm.read_def_cterms, Thm.read_cterm etc.).  INCOMPATIBILITY, should
  4223 use regular Syntax.read_typ, Syntax.read_term, Syntax.read_typ_global,
  4224 Syntax.read_term_global etc.; see also OldGoals.read_term as last
  4225 resort for legacy applications.
  4226 
  4227 * Disposed old declarations, tactics, tactic combinators that refer to
  4228 the simpset or claset of an implicit theory (such as Addsimps,
  4229 Simp_tac, SIMPSET).  INCOMPATIBILITY, should use @{simpset} etc. in
  4230 embedded ML text, or local_simpset_of with a proper context passed as
  4231 explicit runtime argument.
  4232 
  4233 * Rules and tactics that read instantiations (read_instantiate,
  4234 res_inst_tac, thin_tac, subgoal_tac etc.) now demand a proper proof
  4235 context, which is required for parsing and type-checking.  Moreover,
  4236 the variables are specified as plain indexnames, not string encodings
  4237 thereof.  INCOMPATIBILITY.
  4238 
  4239 * Generic Toplevel.add_hook interface allows to analyze the result of
  4240 transactions.  E.g. see src/Pure/ProofGeneral/proof_general_pgip.ML
  4241 for theorem dependency output of transactions resulting in a new
  4242 theory state.
  4243 
  4244 * ML antiquotations: block-structured compilation context indicated by
  4245 \<lbrace> ... \<rbrace>; additional antiquotation forms:
  4246 
  4247   @{binding name}                         - basic name binding
  4248   @{let ?pat = term}                      - term abbreviation (HO matching)
  4249   @{note name = fact}                     - fact abbreviation
  4250   @{thm fact}                             - singleton fact (with attributes)
  4251   @{thms fact}                            - general fact (with attributes)
  4252   @{lemma prop by method}                 - singleton goal
  4253   @{lemma prop by meth1 meth2}            - singleton goal
  4254   @{lemma prop1 ... propN by method}      - general goal
  4255   @{lemma prop1 ... propN by meth1 meth2} - general goal
  4256   @{lemma (open) ...}                     - open derivation
  4257 
  4258 
  4259 *** System ***
  4260 
  4261 * The Isabelle "emacs" tool provides a specific interface to invoke
  4262 Proof General / Emacs, with more explicit failure if that is not
  4263 installed (the old isabelle-interface script silently falls back on
  4264 isabelle-process).  The PROOFGENERAL_HOME setting determines the
  4265 installation location of the Proof General distribution.
  4266 
  4267 * Isabelle/lib/classes/Pure.jar provides basic support to integrate
  4268 the Isabelle process into a JVM/Scala application.  See
  4269 Isabelle/lib/jedit/plugin for a minimal example.  (The obsolete Java
  4270 process wrapper has been discontinued.)
  4271 
  4272 * Added homegrown Isabelle font with unicode layout, see lib/fonts.
  4273 
  4274 * Various status messages (with exact source position information) are
  4275 emitted, if proper markup print mode is enabled.  This allows
  4276 user-interface components to provide detailed feedback on internal
  4277 prover operations.
  4278 
  4279 
  4280 
  4281 New in Isabelle2008 (June 2008)
  4282 -------------------------------
  4283 
  4284 *** General ***
  4285 
  4286 * The Isabelle/Isar Reference Manual (isar-ref) has been reorganized
  4287 and updated, with formally checked references as hyperlinks.
  4288 
  4289 * Theory loader: use_thy (and similar operations) no longer set the
  4290 implicit ML context, which was occasionally hard to predict and in
  4291 conflict with concurrency.  INCOMPATIBILITY, use ML within Isar which
  4292 provides a proper context already.
  4293 
  4294 * Theory loader: old-style ML proof scripts being *attached* to a thy
  4295 file are no longer supported.  INCOMPATIBILITY, regular 'uses' and
  4296 'use' within a theory file will do the job.
  4297 
  4298 * Name space merge now observes canonical order, i.e. the second space
  4299 is inserted into the first one, while existing entries in the first
  4300 space take precedence.  INCOMPATIBILITY in rare situations, may try to
  4301 swap theory imports.
  4302 
  4303 * Syntax: symbol \<chi> is now considered a letter.  Potential
  4304 INCOMPATIBILITY in identifier syntax etc.
  4305 
  4306 * Outer syntax: string tokens no longer admit escaped white space,
  4307 which was an accidental (undocumented) feature.  INCOMPATIBILITY, use
  4308 white space without escapes.
  4309 
  4310 * Outer syntax: string tokens may contain arbitrary character codes
  4311 specified via 3 decimal digits (as in SML).  E.g. "foo\095bar" for
  4312 "foo_bar".
  4313 
  4314 
  4315 *** Pure ***
  4316 
  4317 * Context-dependent token translations.  Default setup reverts locally
  4318 fixed variables, and adds hilite markup for undeclared frees.
  4319 
  4320 * Unused theorems can be found using the new command 'unused_thms'.
  4321 There are three ways of invoking it:
  4322 
  4323 (1) unused_thms
  4324      Only finds unused theorems in the current theory.
  4325 
  4326 (2) unused_thms thy_1 ... thy_n -
  4327      Finds unused theorems in the current theory and all of its ancestors,
  4328      excluding the theories thy_1 ... thy_n and all of their ancestors.
  4329 
  4330 (3) unused_thms thy_1 ... thy_n - thy'_1 ... thy'_m
  4331      Finds unused theorems in the theories thy'_1 ... thy'_m and all of
  4332      their ancestors, excluding the theories thy_1 ... thy_n and all of
  4333      their ancestors.
  4334 
  4335 In order to increase the readability of the list produced by
  4336 unused_thms, theorems that have been created by a particular instance
  4337 of a theory command such as 'inductive' or 'function' are considered
  4338 to belong to the same "group", meaning that if at least one theorem in
  4339 this group is used, the other theorems in the same group are no longer
  4340 reported as unused.  Moreover, if all theorems in the group are
  4341 unused, only one theorem in the group is displayed.
  4342 
  4343 Note that proof objects have to be switched on in order for
  4344 unused_thms to work properly (i.e. !proofs must be >= 1, which is
  4345 usually the case when using Proof General with the default settings).
  4346 
  4347 * Authentic naming of facts disallows ad-hoc overwriting of previous
  4348 theorems within the same name space.  INCOMPATIBILITY, need to remove
  4349 duplicate fact bindings, or even accidental fact duplications.  Note
  4350 that tools may maintain dynamically scoped facts systematically, using
  4351 PureThy.add_thms_dynamic.
  4352 
  4353 * Command 'hide' now allows to hide from "fact" name space as well.
  4354 
  4355 * Eliminated destructive theorem database, simpset, claset, and
  4356 clasimpset.  Potential INCOMPATIBILITY, really need to observe linear
  4357 update of theories within ML code.
  4358 
  4359 * Eliminated theory ProtoPure and CPure, leaving just one Pure theory.
  4360 INCOMPATIBILITY, object-logics depending on former Pure require
  4361 additional setup PureThy.old_appl_syntax_setup; object-logics
  4362 depending on former CPure need to refer to Pure.
  4363 
  4364 * Commands 'use' and 'ML' are now purely functional, operating on
  4365 theory/local_theory.  Removed former 'ML_setup' (on theory), use 'ML'
  4366 instead.  Added 'ML_val' as mere diagnostic replacement for 'ML'.
  4367 INCOMPATIBILITY.
  4368 
  4369 * Command 'setup': discontinued implicit version with ML reference.
  4370 
  4371 * Instantiation target allows for simultaneous specification of class
  4372 instance operations together with an instantiation proof.
  4373 Type-checking phase allows to refer to class operations uniformly.
  4374 See src/HOL/Complex/Complex.thy for an Isar example and
  4375 src/HOL/Library/Eval.thy for an ML example.
  4376 
  4377 * Indexing of literal facts: be more serious about including only
  4378 facts from the visible specification/proof context, but not the
  4379 background context (locale etc.).  Affects `prop` notation and method
  4380 "fact".  INCOMPATIBILITY: need to name facts explicitly in rare
  4381 situations.
  4382 
  4383 * Method "cases", "induct", "coinduct": removed obsolete/undocumented
  4384 "(open)" option, which used to expose internal bound variables to the
  4385 proof text.
  4386 
  4387 * Isar statements: removed obsolete case "rule_context".
  4388 INCOMPATIBILITY, better use explicit fixes/assumes.
  4389 
  4390 * Locale proofs: default proof step now includes 'unfold_locales';
  4391 hence 'proof' without argument may be used to unfold locale
  4392 predicates.
  4393 
  4394 
  4395 *** Document preparation ***
  4396 
  4397 * Simplified pdfsetup.sty: color/hyperref is used unconditionally for
  4398 both pdf and dvi (hyperlinks usually work in xdvi as well); removed
  4399 obsolete thumbpdf setup (contemporary PDF viewers do this on the
  4400 spot); renamed link color from "darkblue" to "linkcolor" (default
  4401 value unchanged, can be redefined via \definecolor); no longer sets
  4402 "a4paper" option (unnecessary or even intrusive).
  4403 
  4404 * Antiquotation @{lemma A method} proves proposition A by the given
  4405 method (either a method name or a method name plus (optional) method
  4406 arguments in parentheses) and prints A just like @{prop A}.
  4407 
  4408 
  4409 *** HOL ***
  4410 
  4411 * New primrec package.  Specification syntax conforms in style to
  4412 definition/function/....  No separate induction rule is provided.  The
  4413 "primrec" command distinguishes old-style and new-style specifications
  4414 by syntax.  The former primrec package is now named OldPrimrecPackage.
  4415 When adjusting theories, beware: constants stemming from new-style
  4416 primrec specifications have authentic syntax.
  4417 
  4418 * Metis prover is now an order of magnitude faster, and also works
  4419 with multithreading.
  4420 
  4421 * Metis: the maximum number of clauses that can be produced from a
  4422 theorem is now given by the attribute max_clauses.  Theorems that
  4423 exceed this number are ignored, with a warning printed.
  4424 
  4425 * Sledgehammer no longer produces structured proofs by default. To
  4426 enable, declare [[sledgehammer_full = true]].  Attributes
  4427 reconstruction_modulus, reconstruction_sorts renamed
  4428 sledgehammer_modulus, sledgehammer_sorts.  INCOMPATIBILITY.
  4429 
  4430 * Method "induct_scheme" derives user-specified induction rules
  4431 from well-founded induction and completeness of patterns. This factors
  4432 out some operations that are done internally by the function package
  4433 and makes them available separately.  See
  4434 src/HOL/ex/Induction_Scheme.thy for examples.
  4435 
  4436 * More flexible generation of measure functions for termination
  4437 proofs: Measure functions can be declared by proving a rule of the
  4438 form "is_measure f" and giving it the [measure_function] attribute.
  4439 The "is_measure" predicate is logically meaningless (always true), and
  4440 just guides the heuristic.  To find suitable measure functions, the
  4441 termination prover sets up the goal "is_measure ?f" of the appropriate
  4442 type and generates all solutions by prolog-style backwards proof using
  4443 the declared rules.
  4444 
  4445 This setup also deals with rules like 
  4446 
  4447   "is_measure f ==> is_measure (list_size f)"
  4448 
  4449 which accommodates nested datatypes that recurse through lists.
  4450 Similar rules are predeclared for products and option types.
  4451 
  4452 * Turned the type of sets "'a set" into an abbreviation for "'a => bool"
  4453 
  4454   INCOMPATIBILITIES:
  4455 
  4456   - Definitions of overloaded constants on sets have to be replaced by
  4457     definitions on => and bool.
  4458 
  4459   - Some definitions of overloaded operators on sets can now be proved
  4460     using the definitions of the operators on => and bool.  Therefore,
  4461     the following theorems have been renamed:
  4462 
  4463       subset_def   -> subset_eq
  4464       psubset_def  -> psubset_eq
  4465       set_diff_def -> set_diff_eq
  4466       Compl_def    -> Compl_eq
  4467       Sup_set_def  -> Sup_set_eq
  4468       Inf_set_def  -> Inf_set_eq
  4469       sup_set_def  -> sup_set_eq
  4470       inf_set_def  -> inf_set_eq
  4471 
  4472   - Due to the incompleteness of the HO unification algorithm, some
  4473     rules such as subst may require manual instantiation, if some of
  4474     the unknowns in the rule is a set.
  4475 
  4476   - Higher order unification and forward proofs:
  4477     The proof pattern
  4478 
  4479       have "P (S::'a set)" <...>
  4480       then have "EX S. P S" ..
  4481 
  4482     no longer works (due to the incompleteness of the HO unification
  4483     algorithm) and must be replaced by the pattern
  4484 
  4485       have "EX S. P S"
  4486       proof
  4487         show "P S" <...>
  4488       qed
  4489 
  4490   - Calculational reasoning with subst (or similar rules):
  4491     The proof pattern
  4492 
  4493       have "P (S::'a set)" <...>
  4494       also have "S = T" <...>
  4495       finally have "P T" .
  4496 
  4497     no longer works (for similar reasons as the previous example) and
  4498     must be replaced by something like
  4499 
  4500       have "P (S::'a set)" <...>
  4501       moreover have "S = T" <...>
  4502       ultimately have "P T" by simp
  4503 
  4504   - Tactics or packages written in ML code:
  4505     Code performing pattern matching on types via
  4506 
  4507       Type ("set", [T]) => ...
  4508 
  4509     must be rewritten. Moreover, functions like strip_type or
  4510     binder_types no longer return the right value when applied to a
  4511     type of the form
  4512 
  4513       T1 => ... => Tn => U => bool
  4514 
  4515     rather than
  4516 
  4517       T1 => ... => Tn => U set
  4518 
  4519 * Merged theories Wellfounded_Recursion, Accessible_Part and
  4520 Wellfounded_Relations to theory Wellfounded.
  4521 
  4522 * Explicit class "eq" for executable equality.  INCOMPATIBILITY.
  4523 
  4524 * Class finite no longer treats UNIV as class parameter.  Use class
  4525 enum from theory Library/Enum instead to achieve a similar effect.
  4526 INCOMPATIBILITY.
  4527 
  4528 * Theory List: rule list_induct2 now has explicitly named cases "Nil"
  4529 and "Cons".  INCOMPATIBILITY.
  4530 
  4531 * HOL (and FOL): renamed variables in rules imp_elim and swap.
  4532 Potential INCOMPATIBILITY.
  4533 
  4534 * Theory Product_Type: duplicated lemmas split_Pair_apply and
  4535 injective_fst_snd removed, use split_eta and prod_eqI instead.
  4536 Renamed upd_fst to apfst and upd_snd to apsnd.  INCOMPATIBILITY.
  4537 
  4538 * Theory Nat: removed redundant lemmas that merely duplicate lemmas of
  4539 the same name in theory Orderings:
  4540 
  4541   less_trans
  4542   less_linear
  4543   le_imp_less_or_eq
  4544   le_less_trans
  4545   less_le_trans
  4546   less_not_sym
  4547   less_asym
  4548 
  4549 Renamed less_imp_le to less_imp_le_nat, and less_irrefl to
  4550 less_irrefl_nat.  Potential INCOMPATIBILITY due to more general types
  4551 and different variable names.
  4552 
  4553 * Library/Option_ord.thy: Canonical order on option type.
  4554 
  4555 * Library/RBT.thy: Red-black trees, an efficient implementation of
  4556 finite maps.
  4557 
  4558 * Library/Countable.thy: Type class for countable types.
  4559 
  4560 * Theory Int: The representation of numerals has changed.  The infix
  4561 operator BIT and the bit datatype with constructors B0 and B1 have
  4562 disappeared.  INCOMPATIBILITY, use "Int.Bit0 x" and "Int.Bit1 y" in
  4563 place of "x BIT bit.B0" and "y BIT bit.B1", respectively.  Theorems
  4564 involving BIT, B0, or B1 have been renamed with "Bit0" or "Bit1"
  4565 accordingly.
  4566 
  4567 * Theory Nat: definition of <= and < on natural numbers no longer
  4568 depend on well-founded relations.  INCOMPATIBILITY.  Definitions
  4569 le_def and less_def have disappeared.  Consider lemmas not_less
  4570 [symmetric, where ?'a = nat] and less_eq [symmetric] instead.
  4571 
  4572 * Theory Finite_Set: locales ACf, ACe, ACIf, ACIfSL and ACIfSLlin
  4573 (whose purpose mainly is for various fold_set functionals) have been
  4574 abandoned in favor of the existing algebraic classes
  4575 ab_semigroup_mult, comm_monoid_mult, ab_semigroup_idem_mult,
  4576 lower_semilattice (resp. upper_semilattice) and linorder.
  4577 INCOMPATIBILITY.
  4578 
  4579 * Theory Transitive_Closure: induct and cases rules now declare proper
  4580 case_names ("base" and "step").  INCOMPATIBILITY.
  4581 
  4582 * Theorem Inductive.lfp_ordinal_induct generalized to complete
  4583 lattices.  The form set-specific version is available as
  4584 Inductive.lfp_ordinal_induct_set.
  4585 
  4586 * Renamed theorems "power.simps" to "power_int.simps".
  4587 INCOMPATIBILITY.
  4588 
  4589 * Class semiring_div provides basic abstract properties of semirings
  4590 with division and modulo operations.  Subsumes former class dvd_mod.
  4591 
  4592 * Merged theories IntDef, Numeral and IntArith into unified theory
  4593 Int.  INCOMPATIBILITY.
  4594 
  4595 * Theory Library/Code_Index: type "index" now represents natural
  4596 numbers rather than integers.  INCOMPATIBILITY.
  4597 
  4598 * New class "uminus" with operation "uminus" (split of from class
  4599 "minus" which now only has operation "minus", binary).
  4600 INCOMPATIBILITY.
  4601 
  4602 * Constants "card", "internal_split", "option_map" now with authentic
  4603 syntax.  INCOMPATIBILITY.
  4604 
  4605 * Definitions subset_def, psubset_def, set_diff_def, Compl_def,
  4606 le_bool_def, less_bool_def, le_fun_def, less_fun_def, inf_bool_def,
  4607 sup_bool_def, Inf_bool_def, Sup_bool_def, inf_fun_def, sup_fun_def,
  4608 Inf_fun_def, Sup_fun_def, inf_set_def, sup_set_def, Inf_set_def,
  4609 Sup_set_def, le_def, less_def, option_map_def now with object
  4610 equality.  INCOMPATIBILITY.
  4611 
  4612 * Records. Removed K_record, and replaced it by pure lambda term
  4613 %x. c. The simplifier setup is now more robust against eta expansion.
  4614 INCOMPATIBILITY: in cases explicitly referring to K_record.
  4615 
  4616 * Library/Multiset: {#a, b, c#} abbreviates {#a#} + {#b#} + {#c#}.
  4617 
  4618 * Library/ListVector: new theory of arithmetic vector operations.
  4619 
  4620 * Library/Order_Relation: new theory of various orderings as sets of
  4621 pairs.  Defines preorders, partial orders, linear orders and
  4622 well-orders on sets and on types.
  4623 
  4624 
  4625 *** ZF ***
  4626 
  4627 * Renamed some theories to allow to loading both ZF and HOL in the
  4628 same session:
  4629 
  4630   Datatype  -> Datatype_ZF
  4631   Inductive -> Inductive_ZF
  4632   Int       -> Int_ZF
  4633   IntDiv    -> IntDiv_ZF
  4634   Nat       -> Nat_ZF
  4635   List      -> List_ZF
  4636   Main      -> Main_ZF
  4637 
  4638 INCOMPATIBILITY: ZF theories that import individual theories below
  4639 Main might need to be adapted.  Regular theory Main is still
  4640 available, as trivial extension of Main_ZF.
  4641 
  4642 
  4643 *** ML ***
  4644 
  4645 * ML within Isar: antiquotation @{const name} or @{const
  4646 name(typargs)} produces statically-checked Const term.
  4647 
  4648 * Functor NamedThmsFun: data is available to the user as dynamic fact
  4649 (of the same name).  Removed obsolete print command.
  4650 
  4651 * Removed obsolete "use_legacy_bindings" function.
  4652 
  4653 * The ``print mode'' is now a thread-local value derived from a global
  4654 template (the former print_mode reference), thus access becomes
  4655 non-critical.  The global print_mode reference is for session
  4656 management only; user-code should use print_mode_value,
  4657 print_mode_active, PrintMode.setmp etc.  INCOMPATIBILITY.
  4658 
  4659 * Functions system/system_out provide a robust way to invoke external
  4660 shell commands, with propagation of interrupts (requires Poly/ML
  4661 5.2.1).  Do not use OS.Process.system etc. from the basis library!
  4662 
  4663 
  4664 *** System ***
  4665 
  4666 * Default settings: PROOFGENERAL_OPTIONS no longer impose xemacs ---
  4667 in accordance with Proof General 3.7, which prefers GNU emacs.
  4668 
  4669 * isatool tty runs Isabelle process with plain tty interaction;
  4670 optional line editor may be specified via ISABELLE_LINE_EDITOR
  4671 setting, the default settings attempt to locate "ledit" and "rlwrap".
  4672 
  4673 * isatool browser now works with Cygwin as well, using general
  4674 "javapath" function defined in Isabelle process environment.
  4675 
  4676 * YXML notation provides a simple and efficient alternative to
  4677 standard XML transfer syntax.  See src/Pure/General/yxml.ML and
  4678 isatool yxml as described in the Isabelle system manual.
  4679 
  4680 * JVM class isabelle.IsabelleProcess (located in Isabelle/lib/classes)
  4681 provides general wrapper for managing an Isabelle process in a robust
  4682 fashion, with ``cooked'' output from stdin/stderr.
  4683 
  4684 * Rudimentary Isabelle plugin for jEdit (see Isabelle/lib/jedit),
  4685 based on Isabelle/JVM process wrapper (see Isabelle/lib/classes).
  4686 
  4687 * Removed obsolete THIS_IS_ISABELLE_BUILD feature.  NB: the documented
  4688 way of changing the user's settings is via
  4689 ISABELLE_HOME_USER/etc/settings, which is a fully featured bash
  4690 script.
  4691 
  4692 * Multithreading.max_threads := 0 refers to the number of actual CPU
  4693 cores of the underlying machine, which is a good starting point for
  4694 optimal performance tuning.  The corresponding usedir option -M allows
  4695 "max" as an alias for "0".  WARNING: does not work on certain versions
  4696 of Mac OS (with Poly/ML 5.1).
  4697 
  4698 * isabelle-process: non-ML sessions are run with "nice", to reduce the
  4699 adverse effect of Isabelle flooding interactive front-ends (notably
  4700 ProofGeneral / XEmacs).
  4701 
  4702 
  4703 
  4704 New in Isabelle2007 (November 2007)
  4705 -----------------------------------
  4706 
  4707 *** General ***
  4708 
  4709 * More uniform information about legacy features, notably a
  4710 warning/error of "Legacy feature: ...", depending on the state of the
  4711 tolerate_legacy_features flag (default true). FUTURE INCOMPATIBILITY:
  4712 legacy features will disappear eventually.
  4713 
  4714 * Theory syntax: the header format ``theory A = B + C:'' has been
  4715 discontinued in favour of ``theory A imports B C begin''.  Use isatool
  4716 fixheaders to convert existing theory files.  INCOMPATIBILITY.
  4717 
  4718 * Theory syntax: the old non-Isar theory file format has been
  4719 discontinued altogether.  Note that ML proof scripts may still be used
  4720 with Isar theories; migration is usually quite simple with the ML
  4721 function use_legacy_bindings.  INCOMPATIBILITY.
  4722 
  4723 * Theory syntax: some popular names (e.g. 'class', 'declaration',
  4724 'fun', 'help', 'if') are now keywords.  INCOMPATIBILITY, use double
  4725 quotes.
  4726 
  4727 * Theory loader: be more serious about observing the static theory
  4728 header specifications (including optional directories), but not the
  4729 accidental file locations of previously successful loads.  The strict
  4730 update policy of former update_thy is now already performed by
  4731 use_thy, so the former has been removed; use_thys updates several
  4732 theories simultaneously, just as 'imports' within a theory header
  4733 specification, but without merging the results.  Potential
  4734 INCOMPATIBILITY: may need to refine theory headers and commands
  4735 ROOT.ML which depend on load order.
  4736 
  4737 * Theory loader: optional support for content-based file
  4738 identification, instead of the traditional scheme of full physical
  4739 path plus date stamp; configured by the ISABELLE_FILE_IDENT setting
  4740 (cf. the system manual).  The new scheme allows to work with
  4741 non-finished theories in persistent session images, such that source
  4742 files may be moved later on without requiring reloads.
  4743 
  4744 * Theory loader: old-style ML proof scripts being *attached* to a thy
  4745 file (with the same base name as the theory) are considered a legacy
  4746 feature, which will disappear eventually. Even now, the theory loader
  4747 no longer maintains dependencies on such files.
  4748 
  4749 * Syntax: the scope for resolving ambiguities via type-inference is
  4750 now limited to individual terms, instead of whole simultaneous
  4751 specifications as before. This greatly reduces the complexity of the
  4752 syntax module and improves flexibility by separating parsing and
  4753 type-checking. INCOMPATIBILITY: additional type-constraints (explicit
  4754 'fixes' etc.) are required in rare situations.
  4755 
  4756 * Syntax: constants introduced by new-style packages ('definition',
  4757 'abbreviation' etc.) are passed through the syntax module in
  4758 ``authentic mode''. This means that associated mixfix annotations
  4759 really stick to such constants, independently of potential name space
  4760 ambiguities introduced later on. INCOMPATIBILITY: constants in parse
  4761 trees are represented slightly differently, may need to adapt syntax
  4762 translations accordingly. Use CONST marker in 'translations' and
  4763 @{const_syntax} antiquotation in 'parse_translation' etc.
  4764 
  4765 * Legacy goal package: reduced interface to the bare minimum required
  4766 to keep existing proof scripts running.  Most other user-level
  4767 functions are now part of the OldGoals structure, which is *not* open
  4768 by default (consider isatool expandshort before open OldGoals).
  4769 Removed top_sg, prin, printyp, pprint_term/typ altogether, because
  4770 these tend to cause confusion about the actual goal (!) context being
  4771 used here, which is not necessarily the same as the_context().
  4772 
  4773 * Command 'find_theorems': supports "*" wild-card in "name:"
  4774 criterion; "with_dups" option.  Certain ProofGeneral versions might
  4775 support a specific search form (see ProofGeneral/CHANGES).
  4776 
  4777 * The ``prems limit'' option (cf. ProofContext.prems_limit) is now -1
  4778 by default, which means that "prems" (and also "fixed variables") are
  4779 suppressed from proof state output.  Note that the ProofGeneral
  4780 settings mechanism allows to change and save options persistently, but
  4781 older versions of Isabelle will fail to start up if a negative prems
  4782 limit is imposed.
  4783 
  4784 * Local theory targets may be specified by non-nested blocks of
  4785 ``context/locale/class ... begin'' followed by ``end''.  The body may
  4786 contain definitions, theorems etc., including any derived mechanism
  4787 that has been implemented on top of these primitives.  This concept
  4788 generalizes the existing ``theorem (in ...)'' towards more versatility
  4789 and scalability.
  4790 
  4791 * Proof General interface: proper undo of final 'end' command;
  4792 discontinued Isabelle/classic mode (ML proof scripts).
  4793 
  4794 
  4795 *** Document preparation ***
  4796 
  4797 * Added antiquotation @{theory name} which prints the given name,
  4798 after checking that it refers to a valid ancestor theory in the
  4799 current context.
  4800 
  4801 * Added antiquotations @{ML_type text} and @{ML_struct text} which
  4802 check the given source text as ML type/structure, printing verbatim.
  4803 
  4804 * Added antiquotation @{abbrev "c args"} which prints the abbreviation
  4805 "c args == rhs" given in the current context.  (Any number of
  4806 arguments may be given on the LHS.)
  4807 
  4808 
  4809 *** Pure ***
  4810 
  4811 * The 'class' package offers a combination of axclass and locale to
  4812 achieve Haskell-like type classes in Isabelle.  Definitions and
  4813 theorems within a class context produce both relative results (with
  4814 implicit parameters according to the locale context), and polymorphic
  4815 constants with qualified polymorphism (according to the class
  4816 context).  Within the body context of a 'class' target, a separate
  4817 syntax layer ("user space type system") takes care of converting
  4818 between global polymorphic consts and internal locale representation.
  4819 See src/HOL/ex/Classpackage.thy for examples (as well as main HOL).
  4820 "isatool doc classes" provides a tutorial.
  4821 
  4822 * Generic code generator framework allows to generate executable
  4823 code for ML and Haskell (including Isabelle classes).  A short usage
  4824 sketch:
  4825 
  4826     internal compilation:
  4827         export_code <list of constants (term syntax)> in SML
  4828     writing SML code to a file:
  4829         export_code <list of constants (term syntax)> in SML <filename>
  4830     writing OCaml code to a file:
  4831         export_code <list of constants (term syntax)> in OCaml <filename>
  4832     writing Haskell code to a bunch of files:
  4833         export_code <list of constants (term syntax)> in Haskell <filename>
  4834 
  4835     evaluating closed propositions to True/False using code generation:
  4836         method ``eval''
  4837 
  4838 Reasonable default setup of framework in HOL.
  4839 
  4840 Theorem attributs for selecting and transforming function equations theorems:
  4841 
  4842     [code fun]:        select a theorem as function equation for a specific constant
  4843     [code fun del]:    deselect a theorem as function equation for a specific constant
  4844     [code inline]:     select an equation theorem for unfolding (inlining) in place
  4845     [code inline del]: deselect an equation theorem for unfolding (inlining) in place
  4846 
  4847 User-defined serializations (target in {SML, OCaml, Haskell}):
  4848 
  4849     code_const <and-list of constants (term syntax)>
  4850       {(target) <and-list of const target syntax>}+
  4851 
  4852     code_type <and-list of type constructors>
  4853       {(target) <and-list of type target syntax>}+
  4854 
  4855     code_instance <and-list of instances>
  4856       {(target)}+
  4857         where instance ::= <type constructor> :: <class>
  4858 
  4859     code_class <and_list of classes>
  4860       {(target) <and-list of class target syntax>}+
  4861         where class target syntax ::= <class name> {where {<classop> == <target syntax>}+}?
  4862 
  4863 code_instance and code_class only are effective to target Haskell.
  4864 
  4865 For example usage see src/HOL/ex/Codegenerator.thy and
  4866 src/HOL/ex/Codegenerator_Pretty.thy.  A separate tutorial on code
  4867 generation from Isabelle/HOL theories is available via "isatool doc
  4868 codegen".
  4869 
  4870 * Code generator: consts in 'consts_code' Isar commands are now
  4871 referred to by usual term syntax (including optional type
  4872 annotations).
  4873 
  4874 * Command 'no_translations' removes translation rules from theory
  4875 syntax.
  4876 
  4877 * Overloaded definitions are now actually checked for acyclic
  4878 dependencies.  The overloading scheme is slightly more general than
  4879 that of Haskell98, although Isabelle does not demand an exact
  4880 correspondence to type class and instance declarations.
  4881 INCOMPATIBILITY, use ``defs (unchecked overloaded)'' to admit more
  4882 exotic versions of overloading -- at the discretion of the user!
  4883 
  4884 Polymorphic constants are represented via type arguments, i.e. the
  4885 instantiation that matches an instance against the most general
  4886 declaration given in the signature.  For example, with the declaration
  4887 c :: 'a => 'a => 'a, an instance c :: nat => nat => nat is represented
  4888 as c(nat).  Overloading is essentially simultaneous structural
  4889 recursion over such type arguments.  Incomplete specification patterns
  4890 impose global constraints on all occurrences, e.g. c('a * 'a) on the
  4891 LHS means that more general c('a * 'b) will be disallowed on any RHS.
  4892 Command 'print_theory' outputs the normalized system of recursive
  4893 equations, see section "definitions".
  4894 
  4895 * Configuration options are maintained within the theory or proof
  4896 context (with name and type bool/int/string), providing a very simple
  4897 interface to a poor-man's version of general context data.  Tools may
  4898 declare options in ML (e.g. using Attrib.config_int) and then refer to
  4899 these values using Config.get etc.  Users may change options via an
  4900 associated attribute of the same name.  This form of context
  4901 declaration works particularly well with commands 'declare' or
  4902 'using', for example ``declare [[foo = 42]]''.  Thus it has become
  4903 very easy to avoid global references, which would not observe Isar
  4904 toplevel undo/redo and fail to work with multithreading.
  4905 
  4906 Various global ML references of Pure and HOL have been turned into
  4907 configuration options:
  4908 
  4909   Unify.search_bound		unify_search_bound
  4910   Unify.trace_bound		unify_trace_bound
  4911   Unify.trace_simp		unify_trace_simp
  4912   Unify.trace_types		unify_trace_types
  4913   Simplifier.simp_depth_limit	simp_depth_limit
  4914   Blast.depth_limit		blast_depth_limit
  4915   DatatypeProp.dtK		datatype_distinctness_limit
  4916   fast_arith_neq_limit  	fast_arith_neq_limit
  4917   fast_arith_split_limit	fast_arith_split_limit
  4918 
  4919 * Named collections of theorems may be easily installed as context
  4920 data using the functor NamedThmsFun (see also
  4921 src/Pure/Tools/named_thms.ML).  The user may add or delete facts via
  4922 attributes; there is also a toplevel print command.  This facility is
  4923 just a common case of general context data, which is the preferred way
  4924 for anything more complex than just a list of facts in canonical
  4925 order.
  4926 
  4927 * Isar: command 'declaration' augments a local theory by generic
  4928 declaration functions written in ML.  This enables arbitrary content
  4929 being added to the context, depending on a morphism that tells the
  4930 difference of the original declaration context wrt. the application
  4931 context encountered later on.
  4932 
  4933 * Isar: proper interfaces for simplification procedures.  Command
  4934 'simproc_setup' declares named simprocs (with match patterns, and body
  4935 text in ML).  Attribute "simproc" adds/deletes simprocs in the current
  4936 context.  ML antiquotation @{simproc name} retrieves named simprocs.
  4937 
  4938 * Isar: an extra pair of brackets around attribute declarations
  4939 abbreviates a theorem reference involving an internal dummy fact,
  4940 which will be ignored later --- only the effect of the attribute on
  4941 the background context will persist.  This form of in-place
  4942 declarations is particularly useful with commands like 'declare' and
  4943 'using', for example ``have A using [[simproc a]] by simp''.
  4944 
  4945 * Isar: method "assumption" (and implicit closing of subproofs) now
  4946 takes simple non-atomic goal assumptions into account: after applying
  4947 an assumption as a rule the resulting subgoals are solved by atomic
  4948 assumption steps.  This is particularly useful to finish 'obtain'
  4949 goals, such as "!!x. (!!x. P x ==> thesis) ==> P x ==> thesis",
  4950 without referring to the original premise "!!x. P x ==> thesis" in the
  4951 Isar proof context.  POTENTIAL INCOMPATIBILITY: method "assumption" is
  4952 more permissive.
  4953 
  4954 * Isar: implicit use of prems from the Isar proof context is
  4955 considered a legacy feature.  Common applications like ``have A .''
  4956 may be replaced by ``have A by fact'' or ``note `A`''.  In general,
  4957 referencing facts explicitly here improves readability and
  4958 maintainability of proof texts.
  4959 
  4960 * Isar: improper proof element 'guess' is like 'obtain', but derives
  4961 the obtained context from the course of reasoning!  For example:
  4962 
  4963   assume "EX x y. A x & B y"   -- "any previous fact"
  4964   then guess x and y by clarify
  4965 
  4966 This technique is potentially adventurous, depending on the facts and
  4967 proof tools being involved here.
  4968 
  4969 * Isar: known facts from the proof context may be specified as literal
  4970 propositions, using ASCII back-quote syntax.  This works wherever
  4971 named facts used to be allowed so far, in proof commands, proof
  4972 methods, attributes etc.  Literal facts are retrieved from the context
  4973 according to unification of type and term parameters.  For example,
  4974 provided that "A" and "A ==> B" and "!!x. P x ==> Q x" are known
  4975 theorems in the current context, then these are valid literal facts:
  4976 `A` and `A ==> B` and `!!x. P x ==> Q x" as well as `P a ==> Q a` etc.
  4977 
  4978 There is also a proof method "fact" which does the same composition
  4979 for explicit goal states, e.g. the following proof texts coincide with
  4980 certain special cases of literal facts:
  4981 
  4982   have "A" by fact                 ==  note `A`
  4983   have "A ==> B" by fact           ==  note `A ==> B`
  4984   have "!!x. P x ==> Q x" by fact  ==  note `!!x. P x ==> Q x`
  4985   have "P a ==> Q a" by fact       ==  note `P a ==> Q a`
  4986 
  4987 * Isar: ":" (colon) is no longer a symbolic identifier character in
  4988 outer syntax.  Thus symbolic identifiers may be used without
  4989 additional white space in declarations like this: ``assume *: A''.
  4990 
  4991 * Isar: 'print_facts' prints all local facts of the current context,
  4992 both named and unnamed ones.
  4993 
  4994 * Isar: 'def' now admits simultaneous definitions, e.g.:
  4995 
  4996   def x == "t" and y == "u"
  4997 
  4998 * Isar: added command 'unfolding', which is structurally similar to
  4999 'using', but affects both the goal state and facts by unfolding given
  5000 rewrite rules.  Thus many occurrences of the 'unfold' method or
  5001 'unfolded' attribute may be replaced by first-class proof text.
  5002 
  5003 * Isar: methods 'unfold' / 'fold', attributes 'unfolded' / 'folded',
  5004 and command 'unfolding' now all support object-level equalities
  5005 (potentially conditional).  The underlying notion of rewrite rule is
  5006 analogous to the 'rule_format' attribute, but *not* that of the
  5007 Simplifier (which is usually more generous).
  5008 
  5009 * Isar: the new attribute [rotated n] (default n = 1) rotates the
  5010 premises of a theorem by n. Useful in conjunction with drule.
  5011 
  5012 * Isar: the goal restriction operator [N] (default N = 1) evaluates a
  5013 method expression within a sandbox consisting of the first N
  5014 sub-goals, which need to exist.  For example, ``simp_all [3]''
  5015 simplifies the first three sub-goals, while (rule foo, simp_all)[]
  5016 simplifies all new goals that emerge from applying rule foo to the
  5017 originally first one.
  5018 
  5019 * Isar: schematic goals are no longer restricted to higher-order
  5020 patterns; e.g. ``lemma "?P(?x)" by (rule TrueI)'' now works as
  5021 expected.
  5022 
  5023 * Isar: the conclusion of a long theorem statement is now either
  5024 'shows' (a simultaneous conjunction, as before), or 'obtains'
  5025 (essentially a disjunction of cases with local parameters and
  5026 assumptions).  The latter allows to express general elimination rules
  5027 adequately; in this notation common elimination rules look like this:
  5028 
  5029   lemma exE:    -- "EX x. P x ==> (!!x. P x ==> thesis) ==> thesis"
  5030     assumes "EX x. P x"
  5031     obtains x where "P x"
  5032 
  5033   lemma conjE:  -- "A & B ==> (A ==> B ==> thesis) ==> thesis"
  5034     assumes "A & B"
  5035     obtains A and B
  5036 
  5037   lemma disjE:  -- "A | B ==> (A ==> thesis) ==> (B ==> thesis) ==> thesis"
  5038     assumes "A | B"
  5039     obtains
  5040       A
  5041     | B
  5042 
  5043 The subsequent classical rules even refer to the formal "thesis"
  5044 explicitly:
  5045 
  5046   lemma classical:     -- "(~ thesis ==> thesis) ==> thesis"
  5047     obtains "~ thesis"
  5048 
  5049   lemma Peirce's_Law:  -- "((thesis ==> something) ==> thesis) ==> thesis"
  5050     obtains "thesis ==> something"
  5051 
  5052 The actual proof of an 'obtains' statement is analogous to that of the
  5053 Isar proof element 'obtain', only that there may be several cases.
  5054 Optional case names may be specified in parentheses; these will be
  5055 available both in the present proof and as annotations in the
  5056 resulting rule, for later use with the 'cases' method (cf. attribute
  5057 case_names).
  5058 
  5059 * Isar: the assumptions of a long theorem statement are available as
  5060 "assms" fact in the proof context.  This is more appropriate than the
  5061 (historical) "prems", which refers to all assumptions of the current
  5062 context, including those from the target locale, proof body etc.
  5063 
  5064 * Isar: 'print_statement' prints theorems from the current theory or
  5065 proof context in long statement form, according to the syntax of a
  5066 top-level lemma.
  5067 
  5068 * Isar: 'obtain' takes an optional case name for the local context
  5069 introduction rule (default "that").
  5070 
  5071 * Isar: removed obsolete 'concl is' patterns.  INCOMPATIBILITY, use
  5072 explicit (is "_ ==> ?foo") in the rare cases where this still happens
  5073 to occur.
  5074 
  5075 * Pure: syntax "CONST name" produces a fully internalized constant
  5076 according to the current context.  This is particularly useful for
  5077 syntax translations that should refer to internal constant
  5078 representations independently of name spaces.
  5079 
  5080 * Pure: syntax constant for foo (binder "FOO ") is called "foo_binder"
  5081 instead of "FOO ". This allows multiple binder declarations to coexist
  5082 in the same context.  INCOMPATIBILITY.
  5083 
  5084 * Isar/locales: 'notation' provides a robust interface to the 'syntax'
  5085 primitive that also works in a locale context (both for constants and
  5086 fixed variables). Type declaration and internal syntactic representation
  5087 of given constants retrieved from the context. Likewise, the
  5088 'no_notation' command allows to remove given syntax annotations from the
  5089 current context.
  5090 
  5091 * Isar/locales: new derived specification elements 'axiomatization',
  5092 'definition', 'abbreviation', which support type-inference, admit
  5093 object-level specifications (equality, equivalence).  See also the
  5094 isar-ref manual.  Examples:
  5095 
  5096   axiomatization
  5097     eq  (infix "===" 50) where
  5098     eq_refl: "x === x" and eq_subst: "x === y ==> P x ==> P y"
  5099 
  5100   definition "f x y = x + y + 1"
  5101   definition g where "g x = f x x"
  5102 
  5103   abbreviation
  5104     neq  (infix "=!=" 50) where
  5105     "x =!= y == ~ (x === y)"
  5106 
  5107 These specifications may be also used in a locale context.  Then the
  5108 constants being introduced depend on certain fixed parameters, and the
  5109 constant name is qualified by the locale base name.  An internal
  5110 abbreviation takes care for convenient input and output, making the
  5111 parameters implicit and using the original short name.  See also
  5112 src/HOL/ex/Abstract_NAT.thy for an example of deriving polymorphic
  5113 entities from a monomorphic theory.
  5114 
  5115 Presently, abbreviations are only available 'in' a target locale, but
  5116 not inherited by general import expressions.  Also note that
  5117 'abbreviation' may be used as a type-safe replacement for 'syntax' +
  5118 'translations' in common applications.  The "no_abbrevs" print mode
  5119 prevents folding of abbreviations in term output.
  5120 
  5121 Concrete syntax is attached to specified constants in internal form,
  5122 independently of name spaces.  The parse tree representation is
  5123 slightly different -- use 'notation' instead of raw 'syntax', and
  5124 'translations' with explicit "CONST" markup to accommodate this.
  5125 
  5126 * Pure/Isar: unified syntax for new-style specification mechanisms
  5127 (e.g.  'definition', 'abbreviation', or 'inductive' in HOL) admits
  5128 full type inference and dummy patterns ("_").  For example:
  5129 
  5130   definition "K x _ = x"
  5131 
  5132   inductive conj for A B
  5133   where "A ==> B ==> conj A B"
  5134 
  5135 * Pure: command 'print_abbrevs' prints all constant abbreviations of
  5136 the current context.  Print mode "no_abbrevs" prevents inversion of
  5137 abbreviations on output.
  5138 
  5139 * Isar/locales: improved parameter handling: use of locales "var" and
  5140 "struct" no longer necessary; - parameter renamings are no longer
  5141 required to be injective.  For example, this allows to define
  5142 endomorphisms as locale endom = homom mult mult h.
  5143 
  5144 * Isar/locales: changed the way locales with predicates are defined.
  5145 Instead of accumulating the specification, the imported expression is
  5146 now an interpretation.  INCOMPATIBILITY: different normal form of
  5147 locale expressions.  In particular, in interpretations of locales with
  5148 predicates, goals repesenting already interpreted fragments are not
  5149 removed automatically.  Use methods `intro_locales' and
  5150 `unfold_locales'; see below.
  5151 
  5152 * Isar/locales: new methods `intro_locales' and `unfold_locales'
  5153 provide backward reasoning on locales predicates.  The methods are
  5154 aware of interpretations and discharge corresponding goals.
  5155 `intro_locales' is less aggressive then `unfold_locales' and does not
  5156 unfold predicates to assumptions.
  5157 
  5158 * Isar/locales: the order in which locale fragments are accumulated
  5159 has changed.  This enables to override declarations from fragments due
  5160 to interpretations -- for example, unwanted simp rules.
  5161 
  5162 * Isar/locales: interpretation in theories and proof contexts has been
  5163 extended.  One may now specify (and prove) equations, which are
  5164 unfolded in interpreted theorems.  This is useful for replacing
  5165 defined concepts (constants depending on locale parameters) by
  5166 concepts already existing in the target context.  Example:
  5167 
  5168   interpretation partial_order ["op <= :: [int, int] => bool"]
  5169     where "partial_order.less (op <=) (x::int) y = (x < y)"
  5170 
  5171 Typically, the constant `partial_order.less' is created by a
  5172 definition specification element in the context of locale
  5173 partial_order.
  5174 
  5175 * Method "induct": improved internal context management to support
  5176 local fixes and defines on-the-fly. Thus explicit meta-level
  5177 connectives !!  and ==> are rarely required anymore in inductive goals
  5178 (using object-logic connectives for this purpose has been long
  5179 obsolete anyway). Common proof patterns are explained in
  5180 src/HOL/Induct/Common_Patterns.thy, see also
  5181 src/HOL/Isar_examples/Puzzle.thy and src/HOL/Lambda for realistic
  5182 examples.
  5183 
  5184 * Method "induct": improved handling of simultaneous goals. Instead of
  5185 introducing object-level conjunction, the statement is now split into
  5186 several conclusions, while the corresponding symbolic cases are nested
  5187 accordingly. INCOMPATIBILITY, proofs need to be structured explicitly,
  5188 see src/HOL/Induct/Common_Patterns.thy, for example.
  5189 
  5190 * Method "induct": mutual induction rules are now specified as a list
  5191 of rule sharing the same induction cases. HOL packages usually provide
  5192 foo_bar.inducts for mutually defined items foo and bar (e.g. inductive
  5193 predicates/sets or datatypes). INCOMPATIBILITY, users need to specify
  5194 mutual induction rules differently, i.e. like this:
  5195 
  5196   (induct rule: foo_bar.inducts)
  5197   (induct set: foo bar)
  5198   (induct pred: foo bar)
  5199   (induct type: foo bar)
  5200 
  5201 The ML function ProjectRule.projections turns old-style rules into the
  5202 new format.
  5203 
  5204 * Method "coinduct": dual of induction, see
  5205 src/HOL/Library/Coinductive_List.thy for various examples.
  5206 
  5207 * Method "cases", "induct", "coinduct": the ``(open)'' option is
  5208 considered a legacy feature.
  5209 
  5210 * Attribute "symmetric" produces result with standardized schematic
  5211 variables (index 0).  Potential INCOMPATIBILITY.
  5212 
  5213 * Simplifier: by default the simplifier trace only shows top level
  5214 rewrites now. That is, trace_simp_depth_limit is set to 1 by
  5215 default. Thus there is less danger of being flooded by the trace. The
  5216 trace indicates where parts have been suppressed.
  5217   
  5218 * Provers/classical: removed obsolete classical version of elim_format
  5219 attribute; classical elim/dest rules are now treated uniformly when
  5220 manipulating the claset.
  5221 
  5222 * Provers/classical: stricter checks to ensure that supplied intro,
  5223 dest and elim rules are well-formed; dest and elim rules must have at
  5224 least one premise.
  5225 
  5226 * Provers/classical: attributes dest/elim/intro take an optional
  5227 weight argument for the rule (just as the Pure versions).  Weights are
  5228 ignored by automated tools, but determine the search order of single
  5229 rule steps.
  5230 
  5231 * Syntax: input syntax now supports dummy variable binding "%_. b",
  5232 where the body does not mention the bound variable.  Note that dummy
  5233 patterns implicitly depend on their context of bounds, which makes
  5234 "{_. _}" match any set comprehension as expected.  Potential
  5235 INCOMPATIBILITY -- parse translations need to cope with syntactic
  5236 constant "_idtdummy" in the binding position.
  5237 
  5238 * Syntax: removed obsolete syntactic constant "_K" and its associated
  5239 parse translation.  INCOMPATIBILITY -- use dummy abstraction instead,
  5240 for example "A -> B" => "Pi A (%_. B)".
  5241 
  5242 * Pure: 'class_deps' command visualizes the subclass relation, using
  5243 the graph browser tool.
  5244 
  5245 * Pure: 'print_theory' now suppresses certain internal declarations by
  5246 default; use '!' option for full details.
  5247 
  5248 
  5249 *** HOL ***
  5250 
  5251 * Method "metis" proves goals by applying the Metis general-purpose
  5252 resolution prover (see also http://gilith.com/software/metis/).
  5253 Examples are in the directory MetisExamples.  WARNING: the
  5254 Isabelle/HOL-Metis integration does not yet work properly with
  5255 multi-threading.
  5256   
  5257 * Command 'sledgehammer' invokes external automatic theorem provers as
  5258 background processes.  It generates calls to the "metis" method if
  5259 successful. These can be pasted into the proof.  Users do not have to
  5260 wait for the automatic provers to return.  WARNING: does not really
  5261 work with multi-threading.
  5262 
  5263 * New "auto_quickcheck" feature tests outermost goal statements for
  5264 potential counter-examples.  Controlled by ML references
  5265 auto_quickcheck (default true) and auto_quickcheck_time_limit (default
  5266 5000 milliseconds).  Fails silently if statements is outside of
  5267 executable fragment, or any other codgenerator problem occurs.
  5268 
  5269 * New constant "undefined" with axiom "undefined x = undefined".
  5270 
  5271 * Added class "HOL.eq", allowing for code generation with polymorphic
  5272 equality.
  5273 
  5274 * Some renaming of class constants due to canonical name prefixing in
  5275 the new 'class' package:
  5276 
  5277     HOL.abs ~> HOL.abs_class.abs
  5278     HOL.divide ~> HOL.divide_class.divide
  5279     0 ~> HOL.zero_class.zero
  5280     1 ~> HOL.one_class.one
  5281     op + ~> HOL.plus_class.plus
  5282     op - ~> HOL.minus_class.minus
  5283     uminus ~> HOL.minus_class.uminus
  5284     op * ~> HOL.times_class.times
  5285     op < ~> HOL.ord_class.less
  5286     op <= > HOL.ord_class.less_eq
  5287     Nat.power ~> Power.power_class.power
  5288     Nat.size ~> Nat.size_class.size
  5289     Numeral.number_of ~> Numeral.number_class.number_of
  5290     FixedPoint.Inf ~> Lattices.complete_lattice_class.Inf
  5291     FixedPoint.Sup ~> Lattices.complete_lattice_class.Sup
  5292     Orderings.min ~> Orderings.ord_class.min
  5293     Orderings.max ~> Orderings.ord_class.max
  5294     Divides.op div ~> Divides.div_class.div
  5295     Divides.op mod ~> Divides.div_class.mod
  5296     Divides.op dvd ~> Divides.div_class.dvd
  5297 
  5298 INCOMPATIBILITY.  Adaptions may be required in the following cases:
  5299 
  5300 a) User-defined constants using any of the names "plus", "minus",
  5301 "times", "less" or "less_eq". The standard syntax translations for
  5302 "+", "-" and "*" may go wrong.  INCOMPATIBILITY: use more specific
  5303 names.
  5304 
  5305 b) Variables named "plus", "minus", "times", "less", "less_eq"
  5306 INCOMPATIBILITY: use more specific names.
  5307 
  5308 c) Permutative equations (e.g. "a + b = b + a")
  5309 Since the change of names also changes the order of terms, permutative
  5310 rewrite rules may get applied in a different order. Experience shows
  5311 that this is rarely the case (only two adaptions in the whole Isabelle
  5312 distribution).  INCOMPATIBILITY: rewrite proofs
  5313 
  5314 d) ML code directly refering to constant names
  5315 This in general only affects hand-written proof tactics, simprocs and
  5316 so on.  INCOMPATIBILITY: grep your sourcecode and replace names.
  5317 Consider using @{const_name} antiquotation.
  5318 
  5319 * New class "default" with associated constant "default".
  5320 
  5321 * Function "sgn" is now overloaded and available on int, real, complex
  5322 (and other numeric types), using class "sgn".  Two possible defs of
  5323 sgn are given as equational assumptions in the classes sgn_if and
  5324 sgn_div_norm; ordered_idom now also inherits from sgn_if.
  5325 INCOMPATIBILITY.
  5326 
  5327 * Locale "partial_order" now unified with class "order" (cf. theory
  5328 Orderings), added parameter "less".  INCOMPATIBILITY.
  5329 
  5330 * Renamings in classes "order" and "linorder": facts "refl", "trans" and
  5331 "cases" to "order_refl", "order_trans" and "linorder_cases", to avoid
  5332 clashes with HOL "refl" and "trans".  INCOMPATIBILITY.
  5333 
  5334 * Classes "order" and "linorder": potential INCOMPATIBILITY due to
  5335 changed order of proof goals in instance proofs.
  5336 
  5337 * The transitivity reasoner for partial and linear orders is set up
  5338 for classes "order" and "linorder".  Instances of the reasoner are available
  5339 in all contexts importing or interpreting the corresponding locales.
  5340 Method "order" invokes the reasoner separately; the reasoner
  5341 is also integrated with the Simplifier as a solver.  Diagnostic
  5342 command 'print_orders' shows the available instances of the reasoner
  5343 in the current context.
  5344 
  5345 * Localized monotonicity predicate in theory "Orderings"; integrated
  5346 lemmas max_of_mono and min_of_mono with this predicate.
  5347 INCOMPATIBILITY.
  5348 
  5349 * Formulation of theorem "dense" changed slightly due to integration
  5350 with new class dense_linear_order.
  5351 
  5352 * Uniform lattice theory development in HOL.
  5353 
  5354     constants "meet" and "join" now named "inf" and "sup"
  5355     constant "Meet" now named "Inf"
  5356 
  5357     classes "meet_semilorder" and "join_semilorder" now named
  5358       "lower_semilattice" and "upper_semilattice"
  5359     class "lorder" now named "lattice"
  5360     class "comp_lat" now named "complete_lattice"
  5361 
  5362     Instantiation of lattice classes allows explicit definitions
  5363     for "inf" and "sup" operations (or "Inf" and "Sup" for complete lattices).
  5364 
  5365   INCOMPATIBILITY.  Theorem renames:
  5366 
  5367     meet_left_le            ~> inf_le1
  5368     meet_right_le           ~> inf_le2
  5369     join_left_le            ~> sup_ge1
  5370     join_right_le           ~> sup_ge2
  5371     meet_join_le            ~> inf_sup_ord
  5372     le_meetI                ~> le_infI
  5373     join_leI                ~> le_supI
  5374     le_meet                 ~> le_inf_iff
  5375     le_join                 ~> ge_sup_conv
  5376     meet_idempotent         ~> inf_idem
  5377     join_idempotent         ~> sup_idem
  5378     meet_comm               ~> inf_commute
  5379     join_comm               ~> sup_commute
  5380     meet_leI1               ~> le_infI1
  5381     meet_leI2               ~> le_infI2
  5382     le_joinI1               ~> le_supI1
  5383     le_joinI2               ~> le_supI2
  5384     meet_assoc              ~> inf_assoc
  5385     join_assoc              ~> sup_assoc
  5386     meet_left_comm          ~> inf_left_commute
  5387     meet_left_idempotent    ~> inf_left_idem
  5388     join_left_comm          ~> sup_left_commute
  5389     join_left_idempotent    ~> sup_left_idem
  5390     meet_aci                ~> inf_aci
  5391     join_aci                ~> sup_aci
  5392     le_def_meet             ~> le_iff_inf
  5393     le_def_join             ~> le_iff_sup
  5394     join_absorp2            ~> sup_absorb2
  5395     join_absorp1            ~> sup_absorb1
  5396     meet_absorp1            ~> inf_absorb1
  5397     meet_absorp2            ~> inf_absorb2
  5398     meet_join_absorp        ~> inf_sup_absorb
  5399     join_meet_absorp        ~> sup_inf_absorb
  5400     distrib_join_le         ~> distrib_sup_le
  5401     distrib_meet_le         ~> distrib_inf_le
  5402 
  5403     add_meet_distrib_left   ~> add_inf_distrib_left
  5404     add_join_distrib_left   ~> add_sup_distrib_left
  5405     is_join_neg_meet        ~> is_join_neg_inf
  5406     is_meet_neg_join        ~> is_meet_neg_sup
  5407     add_meet_distrib_right  ~> add_inf_distrib_right
  5408     add_join_distrib_right  ~> add_sup_distrib_right
  5409     add_meet_join_distribs  ~> add_sup_inf_distribs
  5410     join_eq_neg_meet        ~> sup_eq_neg_inf
  5411     meet_eq_neg_join        ~> inf_eq_neg_sup
  5412     add_eq_meet_join        ~> add_eq_inf_sup
  5413     meet_0_imp_0            ~> inf_0_imp_0
  5414     join_0_imp_0            ~> sup_0_imp_0
  5415     meet_0_eq_0             ~> inf_0_eq_0
  5416     join_0_eq_0             ~> sup_0_eq_0
  5417     neg_meet_eq_join        ~> neg_inf_eq_sup
  5418     neg_join_eq_meet        ~> neg_sup_eq_inf
  5419     join_eq_if              ~> sup_eq_if
  5420 
  5421     mono_meet               ~> mono_inf
  5422     mono_join               ~> mono_sup
  5423     meet_bool_eq            ~> inf_bool_eq
  5424     join_bool_eq            ~> sup_bool_eq
  5425     meet_fun_eq             ~> inf_fun_eq
  5426     join_fun_eq             ~> sup_fun_eq
  5427     meet_set_eq             ~> inf_set_eq
  5428     join_set_eq             ~> sup_set_eq
  5429     meet1_iff               ~> inf1_iff
  5430     meet2_iff               ~> inf2_iff
  5431     meet1I                  ~> inf1I
  5432     meet2I                  ~> inf2I
  5433     meet1D1                 ~> inf1D1
  5434     meet2D1                 ~> inf2D1
  5435     meet1D2                 ~> inf1D2
  5436     meet2D2                 ~> inf2D2
  5437     meet1E                  ~> inf1E
  5438     meet2E                  ~> inf2E
  5439     join1_iff               ~> sup1_iff
  5440     join2_iff               ~> sup2_iff
  5441     join1I1                 ~> sup1I1
  5442     join2I1                 ~> sup2I1
  5443     join1I1                 ~> sup1I1
  5444     join2I2                 ~> sup1I2
  5445     join1CI                 ~> sup1CI
  5446     join2CI                 ~> sup2CI
  5447     join1E                  ~> sup1E
  5448     join2E                  ~> sup2E
  5449 
  5450     is_meet_Meet            ~> is_meet_Inf
  5451     Meet_bool_def           ~> Inf_bool_def
  5452     Meet_fun_def            ~> Inf_fun_def
  5453     Meet_greatest           ~> Inf_greatest
  5454     Meet_lower              ~> Inf_lower
  5455     Meet_set_def            ~> Inf_set_def
  5456 
  5457     Sup_def                 ~> Sup_Inf
  5458     Sup_bool_eq             ~> Sup_bool_def
  5459     Sup_fun_eq              ~> Sup_fun_def
  5460     Sup_set_eq              ~> Sup_set_def
  5461 
  5462     listsp_meetI            ~> listsp_infI
  5463     listsp_meet_eq          ~> listsp_inf_eq
  5464 
  5465     meet_min                ~> inf_min
  5466     join_max                ~> sup_max
  5467 
  5468 * Added syntactic class "size"; overloaded constant "size" now has
  5469 type "'a::size ==> bool"
  5470 
  5471 * Internal reorganisation of `size' of datatypes: size theorems
  5472 "foo.size" are no longer subsumed by "foo.simps" (but are still
  5473 simplification rules by default!); theorems "prod.size" now named
  5474 "*.size".
  5475 
  5476 * Class "div" now inherits from class "times" rather than "type".
  5477 INCOMPATIBILITY.
  5478 
  5479 * HOL/Finite_Set: "name-space" locales Lattice, Distrib_lattice,
  5480 Linorder etc.  have disappeared; operations defined in terms of
  5481 fold_set now are named Inf_fin, Sup_fin.  INCOMPATIBILITY.
  5482 
  5483 * HOL/Nat: neq0_conv no longer declared as iff.  INCOMPATIBILITY.
  5484 
  5485 * HOL-Word: New extensive library and type for generic, fixed size
  5486 machine words, with arithmetic, bit-wise, shifting and rotating
  5487 operations, reflection into int, nat, and bool lists, automation for
  5488 linear arithmetic (by automatic reflection into nat or int), including
  5489 lemmas on overflow and monotonicity.  Instantiated to all appropriate
  5490 arithmetic type classes, supporting automatic simplification of
  5491 numerals on all operations.
  5492 
  5493 * Library/Boolean_Algebra: locales for abstract boolean algebras.
  5494 
  5495 * Library/Numeral_Type: numbers as types, e.g. TYPE(32).
  5496 
  5497 * Code generator library theories:
  5498   - Code_Integer represents HOL integers by big integer literals in target
  5499     languages.
  5500   - Code_Char represents HOL characters by character literals in target
  5501     languages.
  5502   - Code_Char_chr like Code_Char, but also offers treatment of character
  5503     codes; includes Code_Integer.
  5504   - Executable_Set allows to generate code for finite sets using lists.
  5505   - Executable_Rat implements rational numbers as triples (sign, enumerator,
  5506     denominator).
  5507   - Executable_Real implements a subset of real numbers, namly those
  5508     representable by rational numbers.
  5509   - Efficient_Nat implements natural numbers by integers, which in general will
  5510     result in higher efficency; pattern matching with 0/Suc is eliminated;
  5511     includes Code_Integer.
  5512   - Code_Index provides an additional datatype index which is mapped to
  5513     target-language built-in integers.
  5514   - Code_Message provides an additional datatype message_string which is isomorphic to
  5515     strings; messages are mapped to target-language strings.
  5516 
  5517 * New package for inductive predicates
  5518 
  5519   An n-ary predicate p with m parameters z_1, ..., z_m can now be defined via
  5520 
  5521     inductive
  5522       p :: "U_1 => ... => U_m => T_1 => ... => T_n => bool"
  5523       for z_1 :: U_1 and ... and z_n :: U_m
  5524     where
  5525       rule_1: "... ==> p z_1 ... z_m t_1_1 ... t_1_n"
  5526     | ...
  5527 
  5528   with full support for type-inference, rather than
  5529 
  5530     consts s :: "U_1 => ... => U_m => (T_1 * ... * T_n) set"
  5531 
  5532     abbreviation p :: "U_1 => ... => U_m => T_1 => ... => T_n => bool"
  5533     where "p z_1 ... z_m x_1 ... x_n == (x_1, ..., x_n) : s z_1 ... z_m"
  5534 
  5535     inductive "s z_1 ... z_m"
  5536     intros
  5537       rule_1: "... ==> (t_1_1, ..., t_1_n) : s z_1 ... z_m"
  5538       ...
  5539 
  5540   For backward compatibility, there is a wrapper allowing inductive
  5541   sets to be defined with the new package via
  5542 
  5543     inductive_set
  5544       s :: "U_1 => ... => U_m => (T_1 * ... * T_n) set"
  5545       for z_1 :: U_1 and ... and z_n :: U_m
  5546     where
  5547       rule_1: "... ==> (t_1_1, ..., t_1_n) : s z_1 ... z_m"
  5548     | ...
  5549 
  5550   or
  5551 
  5552     inductive_set
  5553       s :: "U_1 => ... => U_m => (T_1 * ... * T_n) set"
  5554       and p :: "U_1 => ... => U_m => T_1 => ... => T_n => bool"
  5555       for z_1 :: U_1 and ... and z_n :: U_m
  5556     where
  5557       "p z_1 ... z_m x_1 ... x_n == (x_1, ..., x_n) : s z_1 ... z_m"
  5558     | rule_1: "... ==> p z_1 ... z_m t_1_1 ... t_1_n"
  5559     | ...
  5560 
  5561   if the additional syntax "p ..." is required.
  5562 
  5563   Numerous examples can be found in the subdirectories src/HOL/Auth,
  5564   src/HOL/Bali, src/HOL/Induct, and src/HOL/MicroJava.
  5565 
  5566   INCOMPATIBILITIES:
  5567 
  5568   - Since declaration and definition of inductive sets or predicates
  5569     is no longer separated, abbreviations involving the newly
  5570     introduced sets or predicates must be specified together with the
  5571     introduction rules after the 'where' keyword (see above), rather
  5572     than before the actual inductive definition.
  5573 
  5574   - The variables in induction and elimination rules are now
  5575     quantified in the order of their occurrence in the introduction
  5576     rules, rather than in alphabetical order. Since this may break
  5577     some proofs, these proofs either have to be repaired, e.g. by
  5578     reordering the variables a_i_1 ... a_i_{k_i} in Isar 'case'
  5579     statements of the form
  5580 
  5581       case (rule_i a_i_1 ... a_i_{k_i})
  5582 
  5583     or the old order of quantification has to be restored by explicitly adding
  5584     meta-level quantifiers in the introduction rules, i.e.
  5585 
  5586       | rule_i: "!!a_i_1 ... a_i_{k_i}. ... ==> p z_1 ... z_m t_i_1 ... t_i_n"
  5587 
  5588   - The format of the elimination rules is now
  5589 
  5590       p z_1 ... z_m x_1 ... x_n ==>
  5591         (!!a_1_1 ... a_1_{k_1}. x_1 = t_1_1 ==> ... ==> x_n = t_1_n ==> ... ==> P)
  5592         ==> ... ==> P
  5593 
  5594     for predicates and
  5595 
  5596       (x_1, ..., x_n) : s z_1 ... z_m ==>
  5597         (!!a_1_1 ... a_1_{k_1}. x_1 = t_1_1 ==> ... ==> x_n = t_1_n ==> ... ==> P)
  5598         ==> ... ==> P
  5599 
  5600     for sets rather than
  5601 
  5602       x : s z_1 ... z_m ==>
  5603         (!!a_1_1 ... a_1_{k_1}. x = (t_1_1, ..., t_1_n) ==> ... ==> P)
  5604         ==> ... ==> P
  5605 
  5606     This may require terms in goals to be expanded to n-tuples
  5607     (e.g. using case_tac or simplification with the split_paired_all
  5608     rule) before the above elimination rule is applicable.
  5609 
  5610   - The elimination or case analysis rules for (mutually) inductive
  5611     sets or predicates are now called "p_1.cases" ... "p_k.cases". The
  5612     list of rules "p_1_..._p_k.elims" is no longer available.
  5613 
  5614 * New package "function"/"fun" for general recursive functions,
  5615 supporting mutual and nested recursion, definitions in local contexts,
  5616 more general pattern matching and partiality. See HOL/ex/Fundefs.thy
  5617 for small examples, and the separate tutorial on the function
  5618 package. The old recdef "package" is still available as before, but
  5619 users are encouraged to use the new package.
  5620 
  5621 * Method "lexicographic_order" automatically synthesizes termination
  5622 relations as lexicographic combinations of size measures. 
  5623 
  5624 * Case-expressions allow arbitrary constructor-patterns (including
  5625 "_") and take their order into account, like in functional
  5626 programming.  Internally, this is translated into nested
  5627 case-expressions; missing cases are added and mapped to the predefined
  5628 constant "undefined". In complicated cases printing may no longer show
  5629 the original input but the internal form. Lambda-abstractions allow
  5630 the same form of pattern matching: "% pat1 => e1 | ..." is an
  5631 abbreviation for "%x. case x of pat1 => e1 | ..." where x is a new
  5632 variable.
  5633 
  5634 * IntDef: The constant "int :: nat => int" has been removed; now "int"
  5635 is an abbreviation for "of_nat :: nat => int". The simplification
  5636 rules for "of_nat" have been changed to work like "int" did
  5637 previously.  Potential INCOMPATIBILITY:
  5638   - "of_nat (Suc m)" simplifies to "1 + of_nat m" instead of "of_nat m + 1"
  5639   - of_nat_diff and of_nat_mult are no longer default simp rules
  5640 
  5641 * Method "algebra" solves polynomial equations over (semi)rings using
  5642 Groebner bases. The (semi)ring structure is defined by locales and the
  5643 tool setup depends on that generic context. Installing the method for
  5644 a specific type involves instantiating the locale and possibly adding
  5645 declarations for computation on the coefficients.  The method is
  5646 already instantiated for natural numbers and for the axiomatic class
  5647 of idoms with numerals.  See also the paper by Chaieb and Wenzel at
  5648 CALCULEMUS 2007 for the general principles underlying this
  5649 architecture of context-aware proof-tools.
  5650 
  5651 * Method "ferrack" implements quantifier elimination over
  5652 special-purpose dense linear orders using locales (analogous to
  5653 "algebra"). The method is already installed for class
  5654 {ordered_field,recpower,number_ring} which subsumes real, hyperreal,
  5655 rat, etc.
  5656 
  5657 * Former constant "List.op @" now named "List.append".  Use ML
  5658 antiquotations @{const_name List.append} or @{term " ... @ ... "} to
  5659 circumvent possible incompatibilities when working on ML level.
  5660 
  5661 * primrec: missing cases mapped to "undefined" instead of "arbitrary".
  5662 
  5663 * New function listsum :: 'a list => 'a for arbitrary monoids.
  5664 Special syntax: "SUM x <- xs. f x" (and latex variants)
  5665 
  5666 * New syntax for Haskell-like list comprehension (input only), eg.
  5667 [(x,y). x <- xs, y <- ys, x ~= y], see also src/HOL/List.thy.
  5668 
  5669 * The special syntax for function "filter" has changed from [x :
  5670 xs. P] to [x <- xs. P] to avoid an ambiguity caused by list
  5671 comprehension syntax, and for uniformity.  INCOMPATIBILITY.
  5672 
  5673 * [a..b] is now defined for arbitrary linear orders.  It used to be
  5674 defined on nat only, as an abbreviation for [a..<Suc b]
  5675 INCOMPATIBILITY.
  5676 
  5677 * Renamed lemma "set_take_whileD"  to "set_takeWhileD".
  5678 
  5679 * New functions "sorted" and "sort" in src/HOL/List.thy.
  5680 
  5681 * New lemma collection field_simps (an extension of ring_simps) for
  5682 manipulating (in)equations involving division. Multiplies with all
  5683 denominators that can be proved to be non-zero (in equations) or
  5684 positive/negative (in inequations).
  5685 
  5686 * Lemma collections ring_eq_simps, group_eq_simps and ring_distrib
  5687 have been improved and renamed to ring_simps, group_simps and
  5688 ring_distribs.  Removed lemmas field_xyz in theory Ring_and_Field
  5689 because they were subsumed by lemmas xyz.  INCOMPATIBILITY.
  5690 
  5691 * Theory Library/Commutative_Ring: switched from recdef to function
  5692 package; constants add, mul, pow now curried.  Infix syntax for
  5693 algebraic operations.
  5694 
  5695 * Dropped redundant lemma def_imp_eq in favor of meta_eq_to_obj_eq.
  5696 INCOMPATIBILITY.
  5697 
  5698 * Dropped redundant lemma if_def2 in favor of if_bool_eq_conj.
  5699 INCOMPATIBILITY.
  5700 
  5701 * HOL/records: generalised field-update to take a function on the
  5702 field rather than the new value: r(|A := x|) is translated to A_update
  5703 (K x) r The K-combinator that is internally used is called K_record.
  5704 INCOMPATIBILITY: Usage of the plain update functions has to be
  5705 adapted.
  5706  
  5707 * Class "semiring_0" now contains annihilation axioms x * 0 = 0 and 0
  5708 * x = 0, which are required for a semiring.  Richer structures do not
  5709 inherit from semiring_0 anymore, because this property is a theorem
  5710 there, not an axiom.  INCOMPATIBILITY: In instances of semiring_0,
  5711 there is more to prove, but this is mostly trivial.
  5712 
  5713 * Class "recpower" is generalized to arbitrary monoids, not just
  5714 commutative semirings.  INCOMPATIBILITY: may need to incorporate
  5715 commutativity or semiring properties additionally.
  5716 
  5717 * Constant "List.list_all2" in List.thy now uses authentic syntax.
  5718 INCOMPATIBILITY: translations containing list_all2 may go wrong,
  5719 better use 'abbreviation'.
  5720 
  5721 * Renamed constant "List.op mem" to "List.member".  INCOMPATIBILITY.
  5722 
  5723 * Numeral syntax: type 'bin' which was a mere type copy of 'int' has
  5724 been abandoned in favour of plain 'int'.  INCOMPATIBILITY --
  5725 significant changes for setting up numeral syntax for types:
  5726   - New constants Numeral.pred and Numeral.succ instead
  5727       of former Numeral.bin_pred and Numeral.bin_succ.
  5728   - Use integer operations instead of bin_add, bin_mult and so on.
  5729   - Numeral simplification theorems named Numeral.numeral_simps instead of Bin_simps.
  5730   - ML structure Bin_Simprocs now named Int_Numeral_Base_Simprocs.
  5731 
  5732 See src/HOL/Integ/IntArith.thy for an example setup.
  5733 
  5734 * Command 'normal_form' computes the normal form of a term that may
  5735 contain free variables.  For example ``normal_form "rev [a, b, c]"''
  5736 produces ``[b, c, a]'' (without proof).  This command is suitable for
  5737 heavy-duty computations because the functions are compiled to ML
  5738 first.  Correspondingly, a method "normalization" is provided.  See
  5739 further src/HOL/ex/NormalForm.thy and src/Tools/nbe.ML.
  5740 
  5741 * Alternative iff syntax "A <-> B" for equality on bool (with priority
  5742 25 like -->); output depends on the "iff" print_mode, the default is
  5743 "A = B" (with priority 50).
  5744 
  5745 * Relations less (<) and less_eq (<=) are also available on type bool.
  5746 Modified syntax to disallow nesting without explicit parentheses,
  5747 e.g. "(x < y) < z" or "x < (y < z)", but NOT "x < y < z".  Potential
  5748 INCOMPATIBILITY.
  5749 
  5750 * "LEAST x:A. P" expands to "LEAST x. x:A & P" (input only).
  5751 
  5752 * Relation composition operator "op O" now has precedence 75 and binds
  5753 stronger than union and intersection. INCOMPATIBILITY.
  5754 
  5755 * The old set interval syntax "{m..n(}" (and relatives) has been
  5756 removed.  Use "{m..<n}" (and relatives) instead.
  5757 
  5758 * In the context of the assumption "~(s = t)" the Simplifier rewrites
  5759 "t = s" to False (by simproc "neq").  INCOMPATIBILITY, consider using
  5760 ``declare [[simproc del: neq]]''.
  5761 
  5762 * Simplifier: "m dvd n" where m and n are numbers is evaluated to
  5763 True/False.
  5764 
  5765 * Theorem Cons_eq_map_conv no longer declared as "simp".
  5766 
  5767 * Theorem setsum_mult renamed to setsum_right_distrib.
  5768 
  5769 * Prefer ex1I over ex_ex1I in single-step reasoning, e.g. by the
  5770 ``rule'' method.
  5771 
  5772 * Reimplemented methods "sat" and "satx", with several improvements:
  5773 goals no longer need to be stated as "<prems> ==> False", equivalences
  5774 (i.e. "=" on type bool) are handled, variable names of the form
  5775 "lit_<n>" are no longer reserved, significant speedup.
  5776 
  5777 * Methods "sat" and "satx" can now replay MiniSat proof traces.
  5778 zChaff is still supported as well.
  5779 
  5780 * 'inductive' and 'datatype': provide projections of mutual rules,
  5781 bundled as foo_bar.inducts;
  5782 
  5783 * Library: moved theories Parity, GCD, Binomial, Infinite_Set to
  5784 Library.
  5785 
  5786 * Library: moved theory Accessible_Part to main HOL.
  5787 
  5788 * Library: added theory Coinductive_List of potentially infinite lists
  5789 as greatest fixed-point.
  5790 
  5791 * Library: added theory AssocList which implements (finite) maps as
  5792 association lists.
  5793 
  5794 * Method "evaluation" solves goals (i.e. a boolean expression)
  5795 efficiently by compiling it to ML.  The goal is "proved" (via an
  5796 oracle) if it evaluates to True.
  5797 
  5798 * Linear arithmetic now splits certain operators (e.g. min, max, abs)
  5799 also when invoked by the simplifier.  This results in the Simplifier
  5800 being more powerful on arithmetic goals.  INCOMPATIBILITY.
  5801 Configuration option fast_arith_split_limit=0 recovers the old
  5802 behavior.
  5803 
  5804 * Support for hex (0x20) and binary (0b1001) numerals.
  5805 
  5806 * New method: reify eqs (t), where eqs are equations for an
  5807 interpretation I :: 'a list => 'b => 'c and t::'c is an optional
  5808 parameter, computes a term s::'b and a list xs::'a list and proves the
  5809 theorem I xs s = t. This is also known as reification or quoting. The
  5810 resulting theorem is applied to the subgoal to substitute t with I xs
  5811 s.  If t is omitted, the subgoal itself is reified.
  5812 
  5813 * New method: reflection corr_thm eqs (t). The parameters eqs and (t)
  5814 are as explained above. corr_thm is a theorem for I vs (f t) = I vs t,
  5815 where f is supposed to be a computable function (in the sense of code
  5816 generattion). The method uses reify to compute s and xs as above then
  5817 applies corr_thm and uses normalization by evaluation to "prove" f s =
  5818 r and finally gets the theorem t = r, which is again applied to the
  5819 subgoal. An Example is available in src/HOL/ex/ReflectionEx.thy.
  5820 
  5821 * Reflection: Automatic reification now handels binding, an example is
  5822 available in src/HOL/ex/ReflectionEx.thy
  5823 
  5824 * HOL-Statespace: ``State Spaces: The Locale Way'' introduces a
  5825 command 'statespace' that is similar to 'record', but introduces an
  5826 abstract specification based on the locale infrastructure instead of
  5827 HOL types.  This leads to extra flexibility in composing state spaces,
  5828 in particular multiple inheritance and renaming of components.
  5829 
  5830 
  5831 *** HOL-Complex ***
  5832 
  5833 * Hyperreal: Functions root and sqrt are now defined on negative real
  5834 inputs so that root n (- x) = - root n x and sqrt (- x) = - sqrt x.
  5835 Nonnegativity side conditions have been removed from many lemmas, so
  5836 that more subgoals may now be solved by simplification; potential
  5837 INCOMPATIBILITY.
  5838 
  5839 * Real: new type classes formalize real normed vector spaces and
  5840 algebras, using new overloaded constants scaleR :: real => 'a => 'a
  5841 and norm :: 'a => real.
  5842 
  5843 * Real: constant of_real :: real => 'a::real_algebra_1 injects from
  5844 reals into other types. The overloaded constant Reals :: 'a set is now
  5845 defined as range of_real; potential INCOMPATIBILITY.
  5846 
  5847 * Real: proper support for ML code generation, including 'quickcheck'.
  5848 Reals are implemented as arbitrary precision rationals.
  5849 
  5850 * Hyperreal: Several constants that previously worked only for the
  5851 reals have been generalized, so they now work over arbitrary vector
  5852 spaces. Type annotations may need to be added in some cases; potential
  5853 INCOMPATIBILITY.
  5854 
  5855   Infinitesimal  :: ('a::real_normed_vector) star set
  5856   HFinite        :: ('a::real_normed_vector) star set
  5857   HInfinite      :: ('a::real_normed_vector) star set
  5858   approx         :: ('a::real_normed_vector) star => 'a star => bool
  5859   monad          :: ('a::real_normed_vector) star => 'a star set
  5860   galaxy         :: ('a::real_normed_vector) star => 'a star set
  5861   (NS)LIMSEQ     :: [nat => 'a::real_normed_vector, 'a] => bool
  5862   (NS)convergent :: (nat => 'a::real_normed_vector) => bool
  5863   (NS)Bseq       :: (nat => 'a::real_normed_vector) => bool
  5864   (NS)Cauchy     :: (nat => 'a::real_normed_vector) => bool
  5865   (NS)LIM        :: ['a::real_normed_vector => 'b::real_normed_vector, 'a, 'b] => bool
  5866   is(NS)Cont     :: ['a::real_normed_vector => 'b::real_normed_vector, 'a] => bool
  5867   deriv          :: ['a::real_normed_field => 'a, 'a, 'a] => bool
  5868   sgn            :: 'a::real_normed_vector => 'a
  5869   exp            :: 'a::{recpower,real_normed_field,banach} => 'a
  5870 
  5871 * Complex: Some complex-specific constants are now abbreviations for
  5872 overloaded ones: complex_of_real = of_real, cmod = norm, hcmod =
  5873 hnorm.  Other constants have been entirely removed in favor of the
  5874 polymorphic versions (INCOMPATIBILITY):
  5875 
  5876   approx        <-- capprox
  5877   HFinite       <-- CFinite
  5878   HInfinite     <-- CInfinite
  5879   Infinitesimal <-- CInfinitesimal
  5880   monad         <-- cmonad
  5881   galaxy        <-- cgalaxy
  5882   (NS)LIM       <-- (NS)CLIM, (NS)CRLIM
  5883   is(NS)Cont    <-- is(NS)Contc, is(NS)contCR
  5884   (ns)deriv     <-- (ns)cderiv
  5885 
  5886 
  5887 *** HOL-Algebra ***
  5888 
  5889 * Formalisation of ideals and the quotient construction over rings.
  5890 
  5891 * Order and lattice theory no longer based on records.
  5892 INCOMPATIBILITY.
  5893 
  5894 * Renamed lemmas least_carrier -> least_closed and greatest_carrier ->
  5895 greatest_closed.  INCOMPATIBILITY.
  5896 
  5897 * Method algebra is now set up via an attribute.  For examples see
  5898 Ring.thy.  INCOMPATIBILITY: the method is now weaker on combinations
  5899 of algebraic structures.
  5900 
  5901 * Renamed theory CRing to Ring.
  5902 
  5903 
  5904 *** HOL-Nominal ***
  5905 
  5906 * Substantial, yet incomplete support for nominal datatypes (binding
  5907 structures) based on HOL-Nominal logic.  See src/HOL/Nominal and
  5908 src/HOL/Nominal/Examples.  Prospective users should consult
  5909 http://isabelle.in.tum.de/nominal/
  5910 
  5911 
  5912 *** ML ***
  5913 
  5914 * ML basics: just one true type int, which coincides with IntInf.int
  5915 (even on SML/NJ).
  5916 
  5917 * ML within Isar: antiquotations allow to embed statically-checked
  5918 formal entities in the source, referring to the context available at
  5919 compile-time.  For example:
  5920 
  5921 ML {* @{sort "{zero,one}"} *}
  5922 ML {* @{typ "'a => 'b"} *}
  5923 ML {* @{term "%x. x"} *}
  5924 ML {* @{prop "x == y"} *}
  5925 ML {* @{ctyp "'a => 'b"} *}
  5926 ML {* @{cterm "%x. x"} *}
  5927 ML {* @{cprop "x == y"} *}
  5928 ML {* @{thm asm_rl} *}
  5929 ML {* @{thms asm_rl} *}
  5930 ML {* @{type_name c} *}
  5931 ML {* @{type_syntax c} *}
  5932 ML {* @{const_name c} *}
  5933 ML {* @{const_syntax c} *}
  5934 ML {* @{context} *}
  5935 ML {* @{theory} *}
  5936 ML {* @{theory Pure} *}
  5937 ML {* @{theory_ref} *}
  5938 ML {* @{theory_ref Pure} *}
  5939 ML {* @{simpset} *}
  5940 ML {* @{claset} *}
  5941 ML {* @{clasimpset} *}
  5942 
  5943 The same works for sources being ``used'' within an Isar context.
  5944 
  5945 * ML in Isar: improved error reporting; extra verbosity with
  5946 ML_Context.trace enabled.
  5947 
  5948 * Pure/General/table.ML: the join operations now works via exceptions
  5949 DUP/SAME instead of type option. This is simpler in simple cases, and
  5950 admits slightly more efficient complex applications.
  5951 
  5952 * Pure: 'advanced' translation functions (parse_translation etc.) now
  5953 use Context.generic instead of just theory.
  5954 
  5955 * Pure: datatype Context.generic joins theory/Proof.context and
  5956 provides some facilities for code that works in either kind of
  5957 context, notably GenericDataFun for uniform theory and proof data.
  5958 
  5959 * Pure: simplified internal attribute type, which is now always
  5960 Context.generic * thm -> Context.generic * thm. Global (theory) vs.
  5961 local (Proof.context) attributes have been discontinued, while
  5962 minimizing code duplication. Thm.rule_attribute and
  5963 Thm.declaration_attribute build canonical attributes; see also structure
  5964 Context for further operations on Context.generic, notably
  5965 GenericDataFun. INCOMPATIBILITY, need to adapt attribute type
  5966 declarations and definitions.
  5967 
  5968 * Context data interfaces (Theory/Proof/GenericDataFun): removed
  5969 name/print, uninitialized data defaults to ad-hoc copy of empty value,
  5970 init only required for impure data. INCOMPATIBILITY: empty really need
  5971 to be empty (no dependencies on theory content!)
  5972 
  5973 * Pure/kernel: consts certification ignores sort constraints given in
  5974 signature declarations. (This information is not relevant to the
  5975 logic, but only for type inference.) SIGNIFICANT INTERNAL CHANGE,
  5976 potential INCOMPATIBILITY.
  5977 
  5978 * Pure: axiomatic type classes are now purely definitional, with
  5979 explicit proofs of class axioms and super class relations performed
  5980 internally. See Pure/axclass.ML for the main internal interfaces --
  5981 notably AxClass.define_class supercedes AxClass.add_axclass, and
  5982 AxClass.axiomatize_class/classrel/arity supersede
  5983 Sign.add_classes/classrel/arities.
  5984 
  5985 * Pure/Isar: Args/Attrib parsers operate on Context.generic --
  5986 global/local versions on theory vs. Proof.context have been
  5987 discontinued; Attrib.syntax and Method.syntax have been adapted
  5988 accordingly.  INCOMPATIBILITY, need to adapt parser expressions for
  5989 attributes, methods, etc.
  5990 
  5991 * Pure: several functions of signature "... -> theory -> theory * ..."
  5992 have been reoriented to "... -> theory -> ... * theory" in order to
  5993 allow natural usage in combination with the ||>, ||>>, |-> and
  5994 fold_map combinators.
  5995 
  5996 * Pure: official theorem names (closed derivations) and additional
  5997 comments (tags) are now strictly separate.  Name hints -- which are
  5998 maintained as tags -- may be attached any time without affecting the
  5999 derivation.
  6000 
  6001 * Pure: primitive rule lift_rule now takes goal cterm instead of an
  6002 actual goal state (thm).  Use Thm.lift_rule (Thm.cprem_of st i) to
  6003 achieve the old behaviour.
  6004 
  6005 * Pure: the "Goal" constant is now called "prop", supporting a
  6006 slightly more general idea of ``protecting'' meta-level rule
  6007 statements.
  6008 
  6009 * Pure: Logic.(un)varify only works in a global context, which is now
  6010 enforced instead of silently assumed.  INCOMPATIBILITY, may use
  6011 Logic.legacy_(un)varify as temporary workaround.
  6012 
  6013 * Pure: structure Name provides scalable operations for generating
  6014 internal variable names, notably Name.variants etc.  This replaces
  6015 some popular functions from term.ML:
  6016 
  6017   Term.variant		->  Name.variant
  6018   Term.variantlist	->  Name.variant_list
  6019   Term.invent_names	->  Name.invent_list
  6020 
  6021 Note that low-level renaming rarely occurs in new code -- operations
  6022 from structure Variable are used instead (see below).
  6023 
  6024 * Pure: structure Variable provides fundamental operations for proper
  6025 treatment of fixed/schematic variables in a context.  For example,
  6026 Variable.import introduces fixes for schematics of given facts and
  6027 Variable.export reverses the effect (up to renaming) -- this replaces
  6028 various freeze_thaw operations.
  6029 
  6030 * Pure: structure Goal provides simple interfaces for
  6031 init/conclude/finish and tactical prove operations (replacing former
  6032 Tactic.prove).  Goal.prove is the canonical way to prove results
  6033 within a given context; Goal.prove_global is a degraded version for
  6034 theory level goals, including a global Drule.standard.  Note that
  6035 OldGoals.prove_goalw_cterm has long been obsolete, since it is
  6036 ill-behaved in a local proof context (e.g. with local fixes/assumes or
  6037 in a locale context).
  6038 
  6039 * Pure/Syntax: generic interfaces for parsing (Syntax.parse_term etc.)
  6040 and type checking (Syntax.check_term etc.), with common combinations
  6041 (Syntax.read_term etc.). These supersede former Sign.read_term etc.
  6042 which are considered legacy and await removal.
  6043 
  6044 * Pure/Syntax: generic interfaces for type unchecking
  6045 (Syntax.uncheck_terms etc.) and unparsing (Syntax.unparse_term etc.),
  6046 with common combinations (Syntax.pretty_term, Syntax.string_of_term
  6047 etc.).  Former Sign.pretty_term, Sign.string_of_term etc. are still
  6048 available for convenience, but refer to the very same operations using
  6049 a mere theory instead of a full context.
  6050 
  6051 * Isar: simplified treatment of user-level errors, using exception
  6052 ERROR of string uniformly.  Function error now merely raises ERROR,
  6053 without any side effect on output channels.  The Isar toplevel takes
  6054 care of proper display of ERROR exceptions.  ML code may use plain
  6055 handle/can/try; cat_error may be used to concatenate errors like this:
  6056 
  6057   ... handle ERROR msg => cat_error msg "..."
  6058 
  6059 Toplevel ML code (run directly or through the Isar toplevel) may be
  6060 embedded into the Isar toplevel with exception display/debug like
  6061 this:
  6062 
  6063   Isar.toplevel (fn () => ...)
  6064 
  6065 INCOMPATIBILITY, removed special transform_error facilities, removed
  6066 obsolete variants of user-level exceptions (ERROR_MESSAGE,
  6067 Context.PROOF, ProofContext.CONTEXT, Proof.STATE, ProofHistory.FAIL)
  6068 -- use plain ERROR instead.
  6069 
  6070 * Isar: theory setup now has type (theory -> theory), instead of a
  6071 list.  INCOMPATIBILITY, may use #> to compose setup functions.
  6072 
  6073 * Isar: ML toplevel pretty printer for type Proof.context, subject to
  6074 ProofContext.debug/verbose flags.
  6075 
  6076 * Isar: Toplevel.theory_to_proof admits transactions that modify the
  6077 theory before entering a proof state.  Transactions now always see a
  6078 quasi-functional intermediate checkpoint, both in interactive and
  6079 batch mode.
  6080 
  6081 * Isar: simplified interfaces for outer syntax.  Renamed
  6082 OuterSyntax.add_keywords to OuterSyntax.keywords.  Removed
  6083 OuterSyntax.add_parsers -- this functionality is now included in
  6084 OuterSyntax.command etc.  INCOMPATIBILITY.
  6085 
  6086 * Simplifier: the simpset of a running simplification process now
  6087 contains a proof context (cf. Simplifier.the_context), which is the
  6088 very context that the initial simpset has been retrieved from (by
  6089 simpset_of/local_simpset_of).  Consequently, all plug-in components
  6090 (solver, looper etc.) may depend on arbitrary proof data.
  6091 
  6092 * Simplifier.inherit_context inherits the proof context (plus the
  6093 local bounds) of the current simplification process; any simproc
  6094 etc. that calls the Simplifier recursively should do this!  Removed
  6095 former Simplifier.inherit_bounds, which is already included here --
  6096 INCOMPATIBILITY.  Tools based on low-level rewriting may even have to
  6097 specify an explicit context using Simplifier.context/theory_context.
  6098 
  6099 * Simplifier/Classical Reasoner: more abstract interfaces
  6100 change_simpset/claset for modifying the simpset/claset reference of a
  6101 theory; raw versions simpset/claset_ref etc. have been discontinued --
  6102 INCOMPATIBILITY.
  6103 
  6104 * Provers: more generic wrt. syntax of object-logics, avoid hardwired
  6105 "Trueprop" etc.
  6106 
  6107 
  6108 *** System ***
  6109 
  6110 * settings: the default heap location within ISABELLE_HOME_USER now
  6111 includes ISABELLE_IDENTIFIER.  This simplifies use of multiple
  6112 Isabelle installations.
  6113 
  6114 * isabelle-process: option -S (secure mode) disables some critical
  6115 operations, notably runtime compilation and evaluation of ML source
  6116 code.
  6117 
  6118 * Basic Isabelle mode for jEdit, see Isabelle/lib/jedit/.
  6119 
  6120 * Support for parallel execution, using native multicore support of
  6121 Poly/ML 5.1.  The theory loader exploits parallelism when processing
  6122 independent theories, according to the given theory header
  6123 specifications. The maximum number of worker threads is specified via
  6124 usedir option -M or the "max-threads" setting in Proof General. A
  6125 speedup factor of 1.5--3.5 can be expected on a 4-core machine, and up
  6126 to 6 on a 8-core machine.  User-code needs to observe certain
  6127 guidelines for thread-safe programming, see appendix A in the Isar
  6128 Implementation manual.
  6129 
  6130 
  6131 
  6132 New in Isabelle2005 (October 2005)
  6133 ----------------------------------
  6134 
  6135 *** General ***
  6136 
  6137 * Theory headers: the new header syntax for Isar theories is
  6138 
  6139   theory <name>
  6140   imports <theory1> ... <theoryN>
  6141   uses <file1> ... <fileM>
  6142   begin
  6143 
  6144 where the 'uses' part is optional.  The previous syntax
  6145 
  6146   theory <name> = <theory1> + ... + <theoryN>:
  6147 
  6148 will disappear in the next release.  Use isatool fixheaders to convert
  6149 existing theory files.  Note that there is no change in ancient
  6150 non-Isar theories now, but these will disappear soon.
  6151 
  6152 * Theory loader: parent theories can now also be referred to via
  6153 relative and absolute paths.
  6154 
  6155 * Command 'find_theorems' searches for a list of criteria instead of a
  6156 list of constants. Known criteria are: intro, elim, dest, name:string,
  6157 simp:term, and any term. Criteria can be preceded by '-' to select
  6158 theorems that do not match. Intro, elim, dest select theorems that
  6159 match the current goal, name:s selects theorems whose fully qualified
  6160 name contain s, and simp:term selects all simplification rules whose
  6161 lhs match term.  Any other term is interpreted as pattern and selects
  6162 all theorems matching the pattern. Available in ProofGeneral under
  6163 'ProofGeneral -> Find Theorems' or C-c C-f.  Example:
  6164 
  6165   C-c C-f (100) "(_::nat) + _ + _" intro -name: "HOL."
  6166 
  6167 prints the last 100 theorems matching the pattern "(_::nat) + _ + _",
  6168 matching the current goal as introduction rule and not having "HOL."
  6169 in their name (i.e. not being defined in theory HOL).
  6170 
  6171 * Command 'thms_containing' has been discontinued in favour of
  6172 'find_theorems'; INCOMPATIBILITY.
  6173 
  6174 * Communication with Proof General is now 8bit clean, which means that
  6175 Unicode text in UTF-8 encoding may be used within theory texts (both
  6176 formal and informal parts).  Cf. option -U of the Isabelle Proof
  6177 General interface.  Here are some simple examples (cf. src/HOL/ex):
  6178 
  6179   http://isabelle.in.tum.de/library/HOL/ex/Hebrew.html
  6180   http://isabelle.in.tum.de/library/HOL/ex/Chinese.html
  6181 
  6182 * Improved efficiency of the Simplifier and, to a lesser degree, the
  6183 Classical Reasoner.  Typical big applications run around 2 times
  6184 faster.
  6185 
  6186 
  6187 *** Document preparation ***
  6188 
  6189 * Commands 'display_drafts' and 'print_drafts' perform simple output
  6190 of raw sources.  Only those symbols that do not require additional
  6191 LaTeX packages (depending on comments in isabellesym.sty) are
  6192 displayed properly, everything else is left verbatim.  isatool display
  6193 and isatool print are used as front ends (these are subject to the
  6194 DVI/PDF_VIEWER and PRINT_COMMAND settings, respectively).
  6195 
  6196 * Command tags control specific markup of certain regions of text,
  6197 notably folding and hiding.  Predefined tags include "theory" (for
  6198 theory begin and end), "proof" for proof commands, and "ML" for
  6199 commands involving ML code; the additional tags "visible" and
  6200 "invisible" are unused by default.  Users may give explicit tag
  6201 specifications in the text, e.g. ''by %invisible (auto)''.  The
  6202 interpretation of tags is determined by the LaTeX job during document
  6203 preparation: see option -V of isatool usedir, or options -n and -t of
  6204 isatool document, or even the LaTeX macros \isakeeptag, \isafoldtag,
  6205 \isadroptag.
  6206 
  6207 Several document versions may be produced at the same time via isatool
  6208 usedir (the generated index.html will link all of them).  Typical
  6209 specifications include ''-V document=theory,proof,ML'' to present
  6210 theory/proof/ML parts faithfully, ''-V outline=/proof,/ML'' to fold
  6211 proof and ML commands, and ''-V mutilated=-theory,-proof,-ML'' to omit
  6212 these parts without any formal replacement text.  The Isabelle site
  6213 default settings produce ''document'' and ''outline'' versions as
  6214 specified above.
  6215 
  6216 * Several new antiquotations:
  6217 
  6218   @{term_type term} prints a term with its type annotated;
  6219 
  6220   @{typeof term} prints the type of a term;
  6221 
  6222   @{const const} is the same as @{term const}, but checks that the
  6223   argument is a known logical constant;
  6224 
  6225   @{term_style style term} and @{thm_style style thm} print a term or
  6226   theorem applying a "style" to it
  6227 
  6228   @{ML text}
  6229 
  6230 Predefined styles are 'lhs' and 'rhs' printing the lhs/rhs of
  6231 definitions, equations, inequations etc., 'concl' printing only the
  6232 conclusion of a meta-logical statement theorem, and 'prem1' .. 'prem19'
  6233 to print the specified premise.  TermStyle.add_style provides an ML
  6234 interface for introducing further styles.  See also the "LaTeX Sugar"
  6235 document practical applications.  The ML antiquotation prints