NEWS
author wenzelm
Fri Apr 27 21:02:34 2012 +0200 (2012-04-27)
changeset 47806 7e009f4e9f47
parent 47751 f98bbb445c06
child 47807 befe55c8bbdc
permissions -rw-r--r--
tuned;
     1 Isabelle NEWS -- history user-relevant changes
     2 ==============================================
     3 
     4 New in Isabelle2012 (May 2012)
     5 ------------------------------
     6 
     7 *** General ***
     8 
     9 * Prover IDE (PIDE) improvements:
    10 
    11   - more robust Sledgehammer integration (as before the sledgehammer
    12     command-line needs to be typed into the source buffer)
    13   - markup for bound variables
    14   - markup for types of term variables (displayed as tooltips)
    15   - support for user-defined Isar commands within the running session
    16   - improved support for Unicode outside original 16bit range
    17     e.g. glyph for \<A> (thanks to jEdit 4.5.1)
    18 
    19 * Forward declaration of outer syntax keywords within the theory
    20 header -- minor INCOMPATIBILITY for user-defined commands.  Allow new
    21 commands to be used in the same theory where defined.
    22 
    23 * Rule attributes in local theory declarations (e.g. locale or class)
    24 are now statically evaluated: the resulting theorem is stored instead
    25 of the original expression.  INCOMPATIBILITY in rare situations, where
    26 the historic accident of dynamic re-evaluation in interpretations
    27 etc. was exploited.
    28 
    29 * Commands 'lemmas' and 'theorems' allow local variables using 'for'
    30 declaration, and results are standardized before being stored.  Thus
    31 old-style "standard" after instantiation or composition of facts
    32 becomes obsolete.  Minor INCOMPATIBILITY, due to potential change of
    33 indices of schematic variables.
    34 
    35 * Simplified configuration options for syntax ambiguity: see
    36 "syntax_ambiguity_warning" and "syntax_ambiguity_limit" in isar-ref
    37 manual.  Minor INCOMPATIBILITY.
    38 
    39 * Updated and extended reference manuals: "isar-ref" and
    40 "implementation"; reduced remaining material in old "ref" manual.
    41 
    42 
    43 *** Pure ***
    44 
    45 * Auxiliary contexts indicate block structure for specifications with
    46 additional parameters and assumptions.  Such unnamed contexts may be
    47 nested within other targets, like 'theory', 'locale', 'class',
    48 'instantiation' etc.  Results from the local context are generalized
    49 accordingly and applied to the enclosing target context.  Example:
    50 
    51   context
    52     fixes x y z :: 'a
    53     assumes xy: "x = y" and yz: "y = z"
    54   begin
    55 
    56   lemma my_trans: "x = z" using xy yz by simp
    57 
    58   end
    59 
    60   thm my_trans
    61 
    62 The most basic application is to factor-out context elements of
    63 several fixes/assumes/shows theorem statements, e.g. see
    64 ~~/src/HOL/Isar_Examples/Group_Context.thy
    65 
    66 Any other local theory specification element works within the "context
    67 ... begin ... end" block as well.
    68 
    69 * Bundled declarations associate attributed fact expressions with a
    70 given name in the context.  These may be later included in other
    71 contexts.  This allows to manage context extensions casually, without
    72 the logical dependencies of locales and locale interpretation.
    73 
    74 See commands 'bundle', 'include', 'including' etc. in the isar-ref
    75 manual.
    76 
    77 * Rule composition via attribute "OF" (or ML functions OF/MRS) is more
    78 tolerant against multiple unifiers, as long as the final result is
    79 unique.  (As before, rules are composed in canonical right-to-left
    80 order to accommodate newly introduced premises.)
    81 
    82 * Command 'definition' no longer exports the foundational "raw_def"
    83 into the user context.  Minor INCOMPATIBILITY, may use the regular
    84 "def" result with attribute "abs_def" to imitate the old version.
    85 
    86 * Renamed some inner syntax categories:
    87 
    88     num ~> num_token
    89     xnum ~> xnum_token
    90     xstr ~> str_token
    91 
    92 Minor INCOMPATIBILITY.  Note that in practice "num_const" or
    93 "num_position" etc. are mainly used instead (which also include
    94 position information via constraints).
    95 
    96 * Attribute "abs_def" turns an equation of the form "f x y == t" into
    97 "f == %x y. t", which ensures that "simp" or "unfold" steps always
    98 expand it.  This also works for object-logic equality.  (Formerly
    99 undocumented feature.)
   100 
   101 * Discontinued old "prems" fact, which used to refer to the accidental
   102 collection of foundational premises in the context (already marked as
   103 legacy since Isabelle2011).
   104 
   105 * Obsolete command 'types' has been discontinued.  Use 'type_synonym'
   106 instead.  INCOMPATIBILITY.
   107 
   108 * Old code generator for SML and its commands 'code_module',
   109 'code_library', 'consts_code', 'types_code' have been discontinued.
   110 Use commands of the generic code generator instead.  INCOMPATIBILITY.
   111 
   112 * Redundant attribute "code_inline" has been discontinued. Use
   113 "code_unfold" instead.  INCOMPATIBILITY.
   114 
   115 * Dropped attribute "code_unfold_post" in favor of the its dual
   116 "code_abbrev", which yields a common pattern in definitions like
   117 
   118   definition [code_abbrev]: "f = t"
   119 
   120 INCOMPATIBILITY.
   121 
   122 * Sort constraints are now propagated in simultaneous statements, just
   123 like type constraints.  INCOMPATIBILITY in rare situations, where
   124 distinct sorts used to be assigned accidentally.  For example:
   125 
   126   lemma "P (x::'a::foo)" and "Q (y::'a::bar)"  -- "now illegal"
   127 
   128   lemma "P (x::'a)" and "Q (y::'a::bar)"
   129     -- "now uniform 'a::bar instead of default sort for first occurrence (!)"
   130 
   131 
   132 *** HOL ***
   133 
   134 * New tutorial "Programming and Proving in Isabelle/HOL"
   135 ("prog-prove").  It completely supersedes "A Tutorial Introduction to
   136 Structured Isar Proofs" ("isar-overview"), which has been removed.  It
   137 also supersedes "Isabelle/HOL, A Proof Assistant for Higher-Order
   138 Logic" as the recommended beginners tutorial, but does not cover all
   139 of the material of that old tutorial.
   140 
   141 * Type 'a set is now a proper type constructor (just as before
   142 Isabelle2008).  Definitions mem_def and Collect_def have disappeared.
   143 Non-trivial INCOMPATIBILITY.  For developments keeping predicates and
   144 sets separate, it is often sufficient to rephrase sets S accidentally
   145 used as predicates by "%x. x : S" and predicates P accidentally used
   146 as sets by "{x. P x}".  Corresponding proofs in a first step should be
   147 pruned from any tinkering with former theorems mem_def and Collect_def
   148 as far as possible.
   149 
   150 For developments which deliberately mixed predicates and sets, a
   151 planning step is necessary to determine what should become a predicate
   152 and what a set.  It can be helpful to carry out that step in
   153 Isabelle2011-1 before jumping right into the current release.
   154 
   155 * The representation of numerals has changed.  Datatype "num"
   156 represents strictly positive binary numerals, along with functions
   157 "numeral :: num => 'a" and "neg_numeral :: num => 'a" to represent
   158 positive and negated numeric literals, respectively. (See definitions
   159 in ~~/src/HOL/Num.thy.) Potential INCOMPATIBILITY, some user theories
   160 may require adaptations as follows:
   161 
   162   - Theorems with number_ring or number_semiring constraints: These
   163     classes are gone; use comm_ring_1 or comm_semiring_1 instead.
   164 
   165   - Theories defining numeric types: Remove number, number_semiring,
   166     and number_ring instances. Defer all theorems about numerals until
   167     after classes one and semigroup_add have been instantiated.
   168 
   169   - Numeral-only simp rules: Replace each rule having a "number_of v"
   170     pattern with two copies, one for numeral and one for neg_numeral.
   171 
   172   - Theorems about subclasses of semiring_1 or ring_1: These classes
   173     automatically support numerals now, so more simp rules and
   174     simprocs may now apply within the proof.
   175 
   176   - Definitions and theorems using old constructors Pls/Min/Bit0/Bit1:
   177     Redefine using other integer operations.
   178 
   179 * Code generation by default implements sets as container type rather
   180 than predicates.  INCOMPATIBILITY.
   181 
   182 * New proof import from HOL Light: Faster, simpler, and more scalable.
   183 Requires a proof bundle, which is available as an external component.
   184 Discontinued old (and mostly dead) Importer for HOL4 and HOL Light.
   185 INCOMPATIBILITY.
   186 
   187 * New type synonym 'a rel = ('a * 'a) set
   188 
   189 * Typedef with implicit set definition is considered legacy.  Use
   190 "typedef (open)" form instead, which will eventually become the
   191 default.
   192 
   193 * More default pred/set conversions on a couple of relation operations
   194 and predicates.  Added powers of predicate relations.  Consolidation
   195 of some relation theorems:
   196 
   197   converse_def ~> converse_unfold
   198   rel_comp_def ~> relcomp_unfold
   199   symp_def ~> (dropped, use symp_def and sym_def instead)
   200   transp_def ~> transp_trans
   201   Domain_def ~> Domain_unfold
   202   Range_def ~> Domain_converse [symmetric]
   203 
   204 Generalized theorems INF_INT_eq, INF_INT_eq2, SUP_UN_eq, SUP_UN_eq2.
   205 
   206 See theory "Relation" for examples for making use of pred/set
   207 conversions by means of attributes "to_set" and "to_pred".
   208 
   209 INCOMPATIBILITY.
   210 
   211 * Renamed facts about the power operation on relations, i.e., relpow
   212 to match the constant's name:
   213 
   214   rel_pow_1 ~> relpow_1
   215   rel_pow_0_I ~> relpow_0_I
   216   rel_pow_Suc_I ~> relpow_Suc_I
   217   rel_pow_Suc_I2 ~> relpow_Suc_I2
   218   rel_pow_0_E ~> relpow_0_E
   219   rel_pow_Suc_E ~> relpow_Suc_E
   220   rel_pow_E ~> relpow_E
   221   rel_pow_Suc_D2 ~> relpow_Suc_D2
   222   rel_pow_Suc_E2 ~> relpow_Suc_E2
   223   rel_pow_Suc_D2' ~> relpow_Suc_D2'
   224   rel_pow_E2 ~> relpow_E2
   225   rel_pow_add ~> relpow_add
   226   rel_pow_commute ~> relpow
   227   rel_pow_empty ~> relpow_empty:
   228   rtrancl_imp_UN_rel_pow ~> rtrancl_imp_UN_relpow
   229   rel_pow_imp_rtrancl ~> relpow_imp_rtrancl
   230   rtrancl_is_UN_rel_pow ~> rtrancl_is_UN_relpow
   231   rtrancl_imp_rel_pow ~> rtrancl_imp_relpow
   232   rel_pow_fun_conv ~> relpow_fun_conv
   233   rel_pow_finite_bounded1 ~> relpow_finite_bounded1
   234   rel_pow_finite_bounded ~> relpow_finite_bounded
   235   rtrancl_finite_eq_rel_pow ~> rtrancl_finite_eq_relpow
   236   trancl_finite_eq_rel_pow ~> trancl_finite_eq_relpow
   237   single_valued_rel_pow ~> single_valued_relpow
   238 
   239 INCOMPATIBILITY.
   240 
   241 * Theory Relation: Consolidated constant name for relation composition
   242 and corresponding theorem names:
   243 
   244   - Renamed constant rel_comp to relcomp.
   245 
   246   - Dropped abbreviation pred_comp. Use relcompp instead.
   247 
   248   - Renamed theorems:
   249 
   250     rel_compI ~> relcompI
   251     rel_compEpair ~> relcompEpair
   252     rel_compE ~> relcompE
   253     pred_comp_rel_comp_eq ~> relcompp_relcomp_eq
   254     rel_comp_empty1 ~> relcomp_empty1
   255     rel_comp_mono ~> relcomp_mono
   256     rel_comp_subset_Sigma ~> relcomp_subset_Sigma
   257     rel_comp_distrib ~> relcomp_distrib
   258     rel_comp_distrib2 ~> relcomp_distrib2
   259     rel_comp_UNION_distrib ~> relcomp_UNION_distrib
   260     rel_comp_UNION_distrib2 ~> relcomp_UNION_distrib2
   261     single_valued_rel_comp ~> single_valued_relcomp
   262     rel_comp_def ~> relcomp_unfold
   263     converse_rel_comp ~> converse_relcomp
   264     pred_compI ~> relcomppI
   265     pred_compE ~> relcomppE
   266     pred_comp_bot1 ~> relcompp_bot1
   267     pred_comp_bot2 ~> relcompp_bot2
   268     transp_pred_comp_less_eq ~> transp_relcompp_less_eq
   269     pred_comp_mono ~> relcompp_mono
   270     pred_comp_distrib ~> relcompp_distrib
   271     pred_comp_distrib2 ~> relcompp_distrib2
   272     converse_pred_comp ~> converse_relcompp
   273 
   274     finite_rel_comp ~> finite_relcomp
   275 
   276     set_rel_comp ~> set_relcomp
   277 
   278 INCOMPATIBILITY.
   279 
   280 * Theory Divides: Discontinued redundant theorems about div and mod.
   281 INCOMPATIBILITY, use the corresponding generic theorems instead.
   282 
   283   DIVISION_BY_ZERO ~> div_by_0, mod_by_0
   284   zdiv_self ~> div_self
   285   zmod_self ~> mod_self
   286   zdiv_zero ~> div_0
   287   zmod_zero ~> mod_0
   288   zdiv_zmod_equality ~> div_mod_equality2
   289   zdiv_zmod_equality2 ~> div_mod_equality
   290   zmod_zdiv_trivial ~> mod_div_trivial
   291   zdiv_zminus_zminus ~> div_minus_minus
   292   zmod_zminus_zminus ~> mod_minus_minus
   293   zdiv_zminus2 ~> div_minus_right
   294   zmod_zminus2 ~> mod_minus_right
   295   zdiv_minus1_right ~> div_minus1_right
   296   zmod_minus1_right ~> mod_minus1_right
   297   zdvd_mult_div_cancel ~> dvd_mult_div_cancel
   298   zmod_zmult1_eq ~> mod_mult_right_eq
   299   zpower_zmod ~> power_mod
   300   zdvd_zmod ~> dvd_mod
   301   zdvd_zmod_imp_zdvd ~> dvd_mod_imp_dvd
   302   mod_mult_distrib ~> mult_mod_left
   303   mod_mult_distrib2 ~> mult_mod_right
   304 
   305 * Removed redundant theorems nat_mult_2 and nat_mult_2_right; use
   306 generic mult_2 and mult_2_right instead. INCOMPATIBILITY.
   307 
   308 * Finite_Set.fold now qualified.  INCOMPATIBILITY.
   309 
   310 * Consolidated theorem names concerning fold combinators:
   311 
   312   inf_INFI_fold_inf ~> inf_INF_fold_inf
   313   sup_SUPR_fold_sup ~> sup_SUP_fold_sup
   314   INFI_fold_inf ~> INF_fold_inf
   315   SUPR_fold_sup ~> SUP_fold_sup
   316   union_set ~> union_set_fold
   317   minus_set ~> minus_set_fold
   318   INFI_set_fold ~> INF_set_fold
   319   SUPR_set_fold ~> SUP_set_fold
   320   INF_code ~> INF_set_foldr
   321   SUP_code ~> SUP_set_foldr
   322   foldr.simps ~> foldr.simps (in point-free formulation)
   323   foldr_fold_rev ~> foldr_conv_fold
   324   foldl_fold ~> foldl_conv_fold
   325   foldr_foldr ~> foldr_conv_foldl
   326   foldl_foldr ~> foldl_conv_foldr
   327   fold_set_remdups ~> fold_set_fold_remdups
   328   fold_set ~> fold_set_fold
   329   fold1_set ~> fold1_set_fold
   330 
   331 INCOMPATIBILITY.
   332 
   333 * Dropped rarely useful theorems concerning fold combinators:
   334 foldl_apply, foldl_fun_comm, foldl_rev, fold_weak_invariant,
   335 rev_foldl_cons, fold_set_remdups, fold_set, fold_set1,
   336 concat_conv_foldl, foldl_weak_invariant, foldl_invariant,
   337 foldr_invariant, foldl_absorb0, foldl_foldr1_lemma, foldl_foldr1,
   338 listsum_conv_fold, listsum_foldl, sort_foldl_insort, foldl_assoc,
   339 foldr_conv_foldl, start_le_sum, elem_le_sum, sum_eq_0_conv.
   340 INCOMPATIBILITY.  For the common phrases "%xs. List.foldr plus xs 0"
   341 and "List.foldl plus 0", prefer "List.listsum".  Otherwise it can be
   342 useful to boil down "List.foldr" and "List.foldl" to "List.fold" by
   343 unfolding "foldr_conv_fold" and "foldl_conv_fold".
   344 
   345 * Dropped lemmas minus_set_foldr, union_set_foldr, union_coset_foldr,
   346 inter_coset_foldr, Inf_fin_set_foldr, Sup_fin_set_foldr,
   347 Min_fin_set_foldr, Max_fin_set_foldr, Inf_set_foldr, Sup_set_foldr,
   348 INF_set_foldr, SUP_set_foldr.  INCOMPATIBILITY.  Prefer corresponding
   349 lemmas over fold rather than foldr, or make use of lemmas
   350 fold_conv_foldr and fold_rev.
   351 
   352 * Congruence rules Option.map_cong and Option.bind_cong for recursion
   353 through option types.
   354 
   355 * Concrete syntax for case expressions includes constraints for source
   356 positions, and thus produces Prover IDE markup for its bindings.
   357 INCOMPATIBILITY for old-style syntax translations that augment the
   358 pattern notation; e.g. see src/HOL/HOLCF/One.thy for translations of
   359 one_case.
   360 
   361 * Discontinued configuration option "syntax_positions": atomic terms
   362 in parse trees are always annotated by position constraints.
   363 
   364 * HOL/Library/Set_Algebras.thy: Addition and multiplication on sets
   365 are expressed via type classes again. The special syntax
   366 \<oplus>/\<otimes> has been replaced by plain +/*. Removed constant
   367 setsum_set, which is now subsumed by Big_Operators.setsum.
   368 INCOMPATIBILITY.
   369 
   370 * New theory HOL/Library/DAList provides an abstract type for
   371 association lists with distinct keys.
   372 
   373 * 'datatype' specifications allow explicit sort constraints.
   374 
   375 * Theory HOL/Library/Diagonalize has been removed. INCOMPATIBILITY,
   376 use theory HOL/Library/Nat_Bijection instead.
   377 
   378 * Theory HOL/Library/RBT_Impl: Backing implementation of red-black
   379 trees is now inside a type class context.  Names of affected
   380 operations and lemmas have been prefixed by rbt_.  INCOMPATIBILITY for
   381 theories working directly with raw red-black trees, adapt the names as
   382 follows:
   383 
   384   Operations:
   385   bulkload -> rbt_bulkload
   386   del_from_left -> rbt_del_from_left
   387   del_from_right -> rbt_del_from_right
   388   del -> rbt_del
   389   delete -> rbt_delete
   390   ins -> rbt_ins
   391   insert -> rbt_insert
   392   insertw -> rbt_insert_with
   393   insert_with_key -> rbt_insert_with_key
   394   map_entry -> rbt_map_entry
   395   lookup -> rbt_lookup
   396   sorted -> rbt_sorted
   397   tree_greater -> rbt_greater
   398   tree_less -> rbt_less
   399   tree_less_symbol -> rbt_less_symbol
   400   union -> rbt_union
   401   union_with -> rbt_union_with
   402   union_with_key -> rbt_union_with_key
   403 
   404   Lemmas:
   405   balance_left_sorted -> balance_left_rbt_sorted
   406   balance_left_tree_greater -> balance_left_rbt_greater
   407   balance_left_tree_less -> balance_left_rbt_less
   408   balance_right_sorted -> balance_right_rbt_sorted
   409   balance_right_tree_greater -> balance_right_rbt_greater
   410   balance_right_tree_less -> balance_right_rbt_less
   411   balance_sorted -> balance_rbt_sorted
   412   balance_tree_greater -> balance_rbt_greater
   413   balance_tree_less -> balance_rbt_less
   414   bulkload_is_rbt -> rbt_bulkload_is_rbt
   415   combine_sorted -> combine_rbt_sorted
   416   combine_tree_greater -> combine_rbt_greater
   417   combine_tree_less -> combine_rbt_less
   418   delete_in_tree -> rbt_delete_in_tree
   419   delete_is_rbt -> rbt_delete_is_rbt
   420   del_from_left_tree_greater -> rbt_del_from_left_rbt_greater
   421   del_from_left_tree_less -> rbt_del_from_left_rbt_less
   422   del_from_right_tree_greater -> rbt_del_from_right_rbt_greater
   423   del_from_right_tree_less -> rbt_del_from_right_rbt_less
   424   del_in_tree -> rbt_del_in_tree
   425   del_inv1_inv2 -> rbt_del_inv1_inv2
   426   del_sorted -> rbt_del_rbt_sorted
   427   del_tree_greater -> rbt_del_rbt_greater
   428   del_tree_less -> rbt_del_rbt_less
   429   dom_lookup_Branch -> dom_rbt_lookup_Branch
   430   entries_lookup -> entries_rbt_lookup
   431   finite_dom_lookup -> finite_dom_rbt_lookup
   432   insert_sorted -> rbt_insert_rbt_sorted
   433   insertw_is_rbt -> rbt_insertw_is_rbt
   434   insertwk_is_rbt -> rbt_insertwk_is_rbt
   435   insertwk_sorted -> rbt_insertwk_rbt_sorted
   436   insertw_sorted -> rbt_insertw_rbt_sorted
   437   ins_sorted -> ins_rbt_sorted
   438   ins_tree_greater -> ins_rbt_greater
   439   ins_tree_less -> ins_rbt_less
   440   is_rbt_sorted -> is_rbt_rbt_sorted
   441   lookup_balance -> rbt_lookup_balance
   442   lookup_bulkload -> rbt_lookup_rbt_bulkload
   443   lookup_delete -> rbt_lookup_rbt_delete
   444   lookup_Empty -> rbt_lookup_Empty
   445   lookup_from_in_tree -> rbt_lookup_from_in_tree
   446   lookup_in_tree -> rbt_lookup_in_tree
   447   lookup_ins -> rbt_lookup_ins
   448   lookup_insert -> rbt_lookup_rbt_insert
   449   lookup_insertw -> rbt_lookup_rbt_insertw
   450   lookup_insertwk -> rbt_lookup_rbt_insertwk
   451   lookup_keys -> rbt_lookup_keys
   452   lookup_map -> rbt_lookup_map
   453   lookup_map_entry -> rbt_lookup_rbt_map_entry
   454   lookup_tree_greater -> rbt_lookup_rbt_greater
   455   lookup_tree_less -> rbt_lookup_rbt_less
   456   lookup_union -> rbt_lookup_rbt_union
   457   map_entry_color_of -> rbt_map_entry_color_of
   458   map_entry_inv1 -> rbt_map_entry_inv1
   459   map_entry_inv2 -> rbt_map_entry_inv2
   460   map_entry_is_rbt -> rbt_map_entry_is_rbt
   461   map_entry_sorted -> rbt_map_entry_rbt_sorted
   462   map_entry_tree_greater -> rbt_map_entry_rbt_greater
   463   map_entry_tree_less -> rbt_map_entry_rbt_less
   464   map_tree_greater -> map_rbt_greater
   465   map_tree_less -> map_rbt_less
   466   map_sorted -> map_rbt_sorted
   467   paint_sorted -> paint_rbt_sorted
   468   paint_lookup -> paint_rbt_lookup
   469   paint_tree_greater -> paint_rbt_greater
   470   paint_tree_less -> paint_rbt_less
   471   sorted_entries -> rbt_sorted_entries
   472   tree_greater_eq_trans -> rbt_greater_eq_trans
   473   tree_greater_nit -> rbt_greater_nit
   474   tree_greater_prop -> rbt_greater_prop
   475   tree_greater_simps -> rbt_greater_simps
   476   tree_greater_trans -> rbt_greater_trans
   477   tree_less_eq_trans -> rbt_less_eq_trans
   478   tree_less_nit -> rbt_less_nit
   479   tree_less_prop -> rbt_less_prop
   480   tree_less_simps -> rbt_less_simps
   481   tree_less_trans -> rbt_less_trans
   482   tree_ord_props -> rbt_ord_props
   483   union_Branch -> rbt_union_Branch
   484   union_is_rbt -> rbt_union_is_rbt
   485   unionw_is_rbt -> rbt_unionw_is_rbt
   486   unionwk_is_rbt -> rbt_unionwk_is_rbt
   487   unionwk_sorted -> rbt_unionwk_rbt_sorted
   488 
   489 * Theory HOL/Library/Float: Floating point numbers are now defined as a
   490 subset of the real numbers. All operations are defined using the
   491 lifing-framework and proofs use the transfer method.
   492 INCOMPATIBILITY.
   493 
   494   Changed Operations:
   495   float_abs -> abs
   496   float_nprt -> nprt
   497   float_pprt -> pprt
   498   pow2 -> use powr
   499   round_down -> float_round_down
   500   round_up -> float_round_up
   501   scale -> exponent
   502 
   503   Removed Operations:
   504   ceiling_fl, lb_mult, lb_mod, ub_mult, ub_mod
   505 
   506   Renamed Lemmas:
   507   abs_float_def -> Float.compute_float_abs
   508   bitlen_ge0 -> bitlen_nonneg
   509   bitlen.simps -> Float.compute_bitlen
   510   float_components -> Float_mantissa_exponent
   511   float_divl.simps -> Float.compute_float_divl
   512   float_divr.simps -> Float.compute_float_divr
   513   float_eq_odd -> mult_powr_eq_mult_powr_iff
   514   float_power -> real_of_float_power
   515   lapprox_posrat_def -> Float.compute_lapprox_posrat
   516   lapprox_rat.simps -> Float.compute_lapprox_rat
   517   le_float_def' -> Float.compute_float_le
   518   le_float_def -> less_eq_float.rep_eq
   519   less_float_def' -> Float.compute_float_less
   520   less_float_def -> less_float.rep_eq
   521   normfloat_def -> Float.compute_normfloat
   522   normfloat_imp_odd_or_zero -> mantissa_not_dvd and mantissa_noteq_0
   523   normfloat -> normfloat_def
   524   normfloat_unique -> use normfloat_def
   525   number_of_float_Float -> Float.compute_float_numeral, Float.compute_float_neg_numeral
   526   one_float_def -> Float.compute_float_one
   527   plus_float_def -> Float.compute_float_plus
   528   rapprox_posrat_def -> Float.compute_rapprox_posrat
   529   rapprox_rat.simps -> Float.compute_rapprox_rat
   530   real_of_float_0 -> zero_float.rep_eq
   531   real_of_float_1 -> one_float.rep_eq
   532   real_of_float_abs -> abs_float.rep_eq
   533   real_of_float_add -> plus_float.rep_eq
   534   real_of_float_minus -> uminus_float.rep_eq
   535   real_of_float_mult -> times_float.rep_eq
   536   real_of_float_simp -> Float.rep_eq
   537   real_of_float_sub -> minus_float.rep_eq
   538   round_down.simps -> Float.compute_float_round_down
   539   round_up.simps -> Float.compute_float_round_up
   540   times_float_def -> Float.compute_float_times
   541   uminus_float_def -> Float.compute_float_uminus
   542   zero_float_def -> Float.compute_float_zero
   543 
   544   Lemmas not necessary anymore, use the transfer method:
   545   bitlen_B0, bitlen_B1, bitlen_ge1, bitlen_Min, bitlen_Pls, float_divl,
   546   float_divr, float_le_simp, float_less1_mantissa_bound,
   547   float_less_simp, float_less_zero, float_le_zero,
   548   float_pos_less1_e_neg, float_pos_m_pos, float_split, float_split2,
   549   floor_pos_exp, lapprox_posrat, lapprox_posrat_bottom, lapprox_rat,
   550   lapprox_rat_bottom, normalized_float, rapprox_posrat,
   551   rapprox_posrat_le1, rapprox_rat, real_of_float_ge0_exp,
   552   real_of_float_neg_exp, real_of_float_nge0_exp, round_down floor_fl,
   553   round_up, zero_le_float, zero_less_float
   554 
   555 * Session HOL-Word: Discontinued many redundant theorems specific to
   556 type 'a word. INCOMPATIBILITY, use the corresponding generic theorems
   557 instead.
   558 
   559   word_sub_alt ~> word_sub_wi
   560   word_add_alt ~> word_add_def
   561   word_mult_alt ~> word_mult_def
   562   word_minus_alt ~> word_minus_def
   563   word_0_alt ~> word_0_wi
   564   word_1_alt ~> word_1_wi
   565   word_add_0 ~> add_0_left
   566   word_add_0_right ~> add_0_right
   567   word_mult_1 ~> mult_1_left
   568   word_mult_1_right ~> mult_1_right
   569   word_add_commute ~> add_commute
   570   word_add_assoc ~> add_assoc
   571   word_add_left_commute ~> add_left_commute
   572   word_mult_commute ~> mult_commute
   573   word_mult_assoc ~> mult_assoc
   574   word_mult_left_commute ~> mult_left_commute
   575   word_left_distrib ~> left_distrib
   576   word_right_distrib ~> right_distrib
   577   word_left_minus ~> left_minus
   578   word_diff_0_right ~> diff_0_right
   579   word_diff_self ~> diff_self
   580   word_sub_def ~> diff_minus
   581   word_diff_minus ~> diff_minus
   582   word_add_ac ~> add_ac
   583   word_mult_ac ~> mult_ac
   584   word_plus_ac0 ~> add_0_left add_0_right add_ac
   585   word_times_ac1 ~> mult_1_left mult_1_right mult_ac
   586   word_order_trans ~> order_trans
   587   word_order_refl ~> order_refl
   588   word_order_antisym ~> order_antisym
   589   word_order_linear ~> linorder_linear
   590   lenw1_zero_neq_one ~> zero_neq_one
   591   word_number_of_eq ~> number_of_eq
   592   word_of_int_add_hom ~> wi_hom_add
   593   word_of_int_sub_hom ~> wi_hom_sub
   594   word_of_int_mult_hom ~> wi_hom_mult
   595   word_of_int_minus_hom ~> wi_hom_neg
   596   word_of_int_succ_hom ~> wi_hom_succ
   597   word_of_int_pred_hom ~> wi_hom_pred
   598   word_of_int_0_hom ~> word_0_wi
   599   word_of_int_1_hom ~> word_1_wi
   600 
   601 * New tactic "word_bitwise" for splitting machine word equalities and
   602 inequalities into logical circuits. Requires theory "WordBitwise" from HOL-Word
   603 session. Supports addition, subtraction, multiplication, shifting by
   604 constants, bitwise operators and numeric constants. Requires fixed-length word
   605 types, cannot operate on 'a word. Solves many standard word identies outright
   606 and converts more into first order problems amenable to blast or similar. See
   607 HOL/Word/WordBitwise.thy and examples in HOL/Word/Examples/WordExamples.thy.
   608 
   609 * Clarified attribute "mono_set": pure declaration without modifying
   610 the result of the fact expression.
   611 
   612 * "Transitive_Closure.ntrancl": bounded transitive closure on
   613 relations.
   614 
   615 * Constant "Set.not_member" now qualified.  INCOMPATIBILITY.
   616 
   617 * Theory Int: Discontinued many legacy theorems specific to type int.
   618 INCOMPATIBILITY, use the corresponding generic theorems instead.
   619 
   620   zminus_zminus ~> minus_minus
   621   zminus_0 ~> minus_zero
   622   zminus_zadd_distrib ~> minus_add_distrib
   623   zadd_commute ~> add_commute
   624   zadd_assoc ~> add_assoc
   625   zadd_left_commute ~> add_left_commute
   626   zadd_ac ~> add_ac
   627   zmult_ac ~> mult_ac
   628   zadd_0 ~> add_0_left
   629   zadd_0_right ~> add_0_right
   630   zadd_zminus_inverse2 ~> left_minus
   631   zmult_zminus ~> mult_minus_left
   632   zmult_commute ~> mult_commute
   633   zmult_assoc ~> mult_assoc
   634   zadd_zmult_distrib ~> left_distrib
   635   zadd_zmult_distrib2 ~> right_distrib
   636   zdiff_zmult_distrib ~> left_diff_distrib
   637   zdiff_zmult_distrib2 ~> right_diff_distrib
   638   zmult_1 ~> mult_1_left
   639   zmult_1_right ~> mult_1_right
   640   zle_refl ~> order_refl
   641   zle_trans ~> order_trans
   642   zle_antisym ~> order_antisym
   643   zle_linear ~> linorder_linear
   644   zless_linear ~> linorder_less_linear
   645   zadd_left_mono ~> add_left_mono
   646   zadd_strict_right_mono ~> add_strict_right_mono
   647   zadd_zless_mono ~> add_less_le_mono
   648   int_0_less_1 ~> zero_less_one
   649   int_0_neq_1 ~> zero_neq_one
   650   zless_le ~> less_le
   651   zpower_zadd_distrib ~> power_add
   652   zero_less_zpower_abs_iff ~> zero_less_power_abs_iff
   653   zero_le_zpower_abs ~> zero_le_power_abs
   654 
   655 * Theory Deriv: Renamed
   656 
   657   DERIV_nonneg_imp_nonincreasing ~> DERIV_nonneg_imp_nondecreasing
   658 
   659 * Theory Library/Multiset: Improved code generation of multisets.
   660 
   661 * Session HOL-Probability: Introduced the type "'a measure" to represent
   662 measures, this replaces the records 'a algebra and 'a measure_space. The
   663 locales based on subset_class now have two locale-parameters the space
   664 \<Omega> and the set of measurables sets M. The product of probability spaces
   665 uses now the same constant as the finite product of sigma-finite measure
   666 spaces "PiM :: ('i => 'a) measure". Most constants are defined now
   667 outside of locales and gain an additional parameter, like null_sets,
   668 almost_eventually or \<mu>'. Measure space constructions for distributions
   669 and densities now got their own constants distr and density. Instead of
   670 using locales to describe measure spaces with a finite space, the
   671 measure count_space and point_measure is introduced. INCOMPATIBILITY.
   672 
   673   Renamed constants:
   674   measure -> emeasure
   675   finite_measure.\<mu>' -> measure
   676   product_algebra_generator -> prod_algebra
   677   product_prob_space.emb -> prod_emb
   678   product_prob_space.infprod_algebra -> PiM
   679 
   680   Removed locales:
   681   completeable_measure_space
   682   finite_measure_space
   683   finite_prob_space
   684   finite_product_finite_prob_space
   685   finite_product_sigma_algebra
   686   finite_sigma_algebra
   687   measure_space
   688   pair_finite_prob_space
   689   pair_finite_sigma_algebra
   690   pair_finite_space
   691   pair_sigma_algebra
   692   product_sigma_algebra
   693 
   694   Removed constants:
   695   conditional_space
   696   distribution -> use distr measure, or distributed predicate
   697   image_space
   698   joint_distribution -> use distr measure, or distributed predicate
   699   pair_measure_generator
   700   product_prob_space.infprod_algebra -> use PiM
   701   subvimage
   702 
   703   Replacement theorems:
   704   finite_additivity_sufficient -> ring_of_sets.countably_additiveI_finite
   705   finite_measure.empty_measure -> measure_empty
   706   finite_measure.finite_continuity_from_above -> finite_measure.finite_Lim_measure_decseq
   707   finite_measure.finite_continuity_from_below -> finite_measure.finite_Lim_measure_incseq
   708   finite_measure.finite_measure_countably_subadditive -> finite_measure.finite_measure_subadditive_countably
   709   finite_measure.finite_measure_eq -> finite_measure.emeasure_eq_measure
   710   finite_measure.finite_measure -> finite_measure.emeasure_finite
   711   finite_measure.finite_measure_finite_singleton -> finite_measure.finite_measure_eq_setsum_singleton
   712   finite_measure.positive_measure' -> measure_nonneg
   713   finite_measure.real_measure -> finite_measure.emeasure_real
   714   finite_product_prob_space.finite_measure_times -> finite_product_prob_space.finite_measure_PiM_emb
   715   finite_product_sigma_algebra.in_P -> sets_PiM_I_finite
   716   finite_product_sigma_algebra.P_empty -> space_PiM_empty, sets_PiM_empty
   717   information_space.conditional_entropy_eq -> information_space.conditional_entropy_simple_distributed
   718   information_space.conditional_entropy_positive -> information_space.conditional_entropy_nonneg_simple
   719   information_space.conditional_mutual_information_eq_mutual_information -> information_space.conditional_mutual_information_eq_mutual_information_simple
   720   information_space.conditional_mutual_information_generic_positive -> information_space.conditional_mutual_information_nonneg_simple
   721   information_space.conditional_mutual_information_positive -> information_space.conditional_mutual_information_nonneg_simple
   722   information_space.entropy_commute -> information_space.entropy_commute_simple
   723   information_space.entropy_eq -> information_space.entropy_simple_distributed
   724   information_space.entropy_generic_eq -> information_space.entropy_simple_distributed
   725   information_space.entropy_positive -> information_space.entropy_nonneg_simple
   726   information_space.entropy_uniform_max -> information_space.entropy_uniform
   727   information_space.KL_eq_0_imp -> information_space.KL_eq_0_iff_eq
   728   information_space.KL_eq_0 -> information_space.KL_same_eq_0
   729   information_space.KL_ge_0 -> information_space.KL_nonneg
   730   information_space.mutual_information_eq -> information_space.mutual_information_simple_distributed
   731   information_space.mutual_information_positive -> information_space.mutual_information_nonneg_simple
   732   Int_stable_cuboids -> Int_stable_atLeastAtMost
   733   Int_stable_product_algebra_generator -> positive_integral
   734   measure_preserving -> equality "distr M N f = N" "f : measurable M N"
   735   measure_space.additive -> emeasure_additive
   736   measure_space.AE_iff_null_set -> AE_iff_null
   737   measure_space.almost_everywhere_def -> eventually_ae_filter
   738   measure_space.almost_everywhere_vimage -> AE_distrD
   739   measure_space.continuity_from_above -> INF_emeasure_decseq
   740   measure_space.continuity_from_above_Lim -> Lim_emeasure_decseq
   741   measure_space.continuity_from_below_Lim -> Lim_emeasure_incseq
   742   measure_space.continuity_from_below -> SUP_emeasure_incseq
   743   measure_space_density -> emeasure_density
   744   measure_space.density_is_absolutely_continuous -> absolutely_continuousI_density
   745   measure_space.integrable_vimage -> integrable_distr
   746   measure_space.integral_translated_density -> integral_density
   747   measure_space.integral_vimage -> integral_distr
   748   measure_space.measure_additive -> plus_emeasure
   749   measure_space.measure_compl -> emeasure_compl
   750   measure_space.measure_countable_increasing -> emeasure_countable_increasing
   751   measure_space.measure_countably_subadditive -> emeasure_subadditive_countably
   752   measure_space.measure_decseq -> decseq_emeasure
   753   measure_space.measure_Diff -> emeasure_Diff
   754   measure_space.measure_Diff_null_set -> emeasure_Diff_null_set
   755   measure_space.measure_eq_0 -> emeasure_eq_0
   756   measure_space.measure_finitely_subadditive -> emeasure_subadditive_finite
   757   measure_space.measure_finite_singleton -> emeasure_eq_setsum_singleton
   758   measure_space.measure_incseq -> incseq_emeasure
   759   measure_space.measure_insert -> emeasure_insert
   760   measure_space.measure_mono -> emeasure_mono
   761   measure_space.measure_not_negative -> emeasure_not_MInf
   762   measure_space.measure_preserving_Int_stable -> measure_eqI_generator_eq
   763   measure_space.measure_setsum -> setsum_emeasure
   764   measure_space.measure_setsum_split -> setsum_emeasure_cover
   765   measure_space.measure_space_vimage -> emeasure_distr
   766   measure_space.measure_subadditive_finite -> emeasure_subadditive_finite
   767   measure_space.measure_subadditive -> subadditive
   768   measure_space.measure_top -> emeasure_space
   769   measure_space.measure_UN_eq_0 -> emeasure_UN_eq_0
   770   measure_space.measure_Un_null_set -> emeasure_Un_null_set
   771   measure_space.positive_integral_translated_density -> positive_integral_density
   772   measure_space.positive_integral_vimage -> positive_integral_distr
   773   measure_space.real_continuity_from_above -> Lim_measure_decseq
   774   measure_space.real_continuity_from_below -> Lim_measure_incseq
   775   measure_space.real_measure_countably_subadditive -> measure_subadditive_countably
   776   measure_space.real_measure_Diff -> measure_Diff
   777   measure_space.real_measure_finite_Union -> measure_finite_Union
   778   measure_space.real_measure_setsum_singleton -> measure_eq_setsum_singleton
   779   measure_space.real_measure_subadditive -> measure_subadditive
   780   measure_space.real_measure_Union -> measure_Union
   781   measure_space.real_measure_UNION -> measure_UNION
   782   measure_space.simple_function_vimage -> simple_function_comp
   783   measure_space.simple_integral_vimage -> simple_integral_distr
   784   measure_space.simple_integral_vimage -> simple_integral_distr
   785   measure_unique_Int_stable -> measure_eqI_generator_eq
   786   measure_unique_Int_stable_vimage -> measure_eqI_generator_eq
   787   pair_sigma_algebra.measurable_cut_fst -> sets_Pair1
   788   pair_sigma_algebra.measurable_cut_snd -> sets_Pair2
   789   pair_sigma_algebra.measurable_pair_image_fst -> measurable_Pair1
   790   pair_sigma_algebra.measurable_pair_image_snd -> measurable_Pair2
   791   pair_sigma_algebra.measurable_product_swap -> measurable_pair_swap_iff
   792   pair_sigma_algebra.pair_sigma_algebra_measurable -> measurable_pair_swap
   793   pair_sigma_algebra.pair_sigma_algebra_swap_measurable -> measurable_pair_swap'
   794   pair_sigma_algebra.sets_swap -> sets_pair_swap
   795   pair_sigma_finite.measure_cut_measurable_fst -> pair_sigma_finite.measurable_emeasure_Pair1
   796   pair_sigma_finite.measure_cut_measurable_snd -> pair_sigma_finite.measurable_emeasure_Pair2
   797   pair_sigma_finite.measure_preserving_swap -> pair_sigma_finite.distr_pair_swap
   798   pair_sigma_finite.pair_measure_alt2 -> pair_sigma_finite.emeasure_pair_measure_alt2
   799   pair_sigma_finite.pair_measure_alt -> pair_sigma_finite.emeasure_pair_measure_alt
   800   pair_sigma_finite.pair_measure_times -> pair_sigma_finite.emeasure_pair_measure_Times
   801   prob_space.indep_distribution_eq_measure -> prob_space.indep_vars_iff_distr_eq_PiM
   802   prob_space.indep_var_distributionD -> prob_space.indep_var_distribution_eq
   803   prob_space.measure_space_1 -> prob_space.emeasure_space_1
   804   prob_space.prob_space_vimage -> prob_space_distr
   805   prob_space.random_variable_restrict -> measurable_restrict
   806   prob_space_unique_Int_stable -> measure_eqI_prob_space
   807   product_algebraE -> prod_algebraE_all
   808   product_algebra_generator_der -> prod_algebra_eq_finite
   809   product_algebra_generator_into_space -> prod_algebra_sets_into_space
   810   product_algebraI -> sets_PiM_I_finite
   811   product_measure_exists -> product_sigma_finite.sigma_finite
   812   product_prob_space.finite_index_eq_finite_product -> product_prob_space.sets_PiM_generator
   813   product_prob_space.finite_measure_infprod_emb_Pi -> product_prob_space.measure_PiM_emb
   814   product_prob_space.infprod_spec -> product_prob_space.emeasure_PiM_emb_not_empty
   815   product_prob_space.measurable_component -> measurable_component_singleton
   816   product_prob_space.measurable_emb -> measurable_prod_emb
   817   product_prob_space.measurable_into_infprod_algebra -> measurable_PiM_single
   818   product_prob_space.measurable_singleton_infprod -> measurable_component_singleton
   819   product_prob_space.measure_emb -> emeasure_prod_emb
   820   product_prob_space.measure_preserving_restrict -> product_prob_space.distr_restrict
   821   product_sigma_algebra.product_algebra_into_space -> space_closed
   822   product_sigma_finite.measure_fold -> product_sigma_finite.distr_merge
   823   product_sigma_finite.measure_preserving_component_singelton -> product_sigma_finite.distr_singleton
   824   product_sigma_finite.measure_preserving_merge -> product_sigma_finite.distr_merge
   825   sequence_space.measure_infprod -> sequence_space.measure_PiM_countable
   826   sets_product_algebra -> sets_PiM
   827   sigma_algebra.measurable_sigma -> measurable_measure_of
   828   sigma_finite_measure.disjoint_sigma_finite -> sigma_finite_disjoint
   829   sigma_finite_measure.RN_deriv_vimage -> sigma_finite_measure.RN_deriv_distr
   830   sigma_product_algebra_sigma_eq -> sigma_prod_algebra_sigma_eq
   831   space_product_algebra -> space_PiM
   832 
   833 * New "case_product" attribute to generate a case rule doing multiple
   834 case distinctions at the same time.  E.g.
   835 
   836   list.exhaust [case_product nat.exhaust]
   837 
   838 produces a rule which can be used to perform case distinction on both
   839 a list and a nat.
   840 
   841 * New Transfer package:
   842 
   843   - transfer_rule attribute: Maintains a collection of transfer rules,
   844     which relate constants at two different types. Transfer rules may
   845     relate different type instances of the same polymorphic constant,
   846     or they may relate an operation on a raw type to a corresponding
   847     operation on an abstract type (quotient or subtype). For example:
   848 
   849     ((A ===> B) ===> list_all2 A ===> list_all2 B) map map
   850     (cr_int ===> cr_int ===> cr_int) (%(x,y) (u,v). (x+u, y+v)) plus_int
   851 
   852   - transfer method: Replaces a subgoal on abstract types with an
   853     equivalent subgoal on the corresponding raw types. Constants are
   854     replaced with corresponding ones according to the transfer rules.
   855     Goals are generalized over all free variables by default; this is
   856     necessary for variables whose types changes, but can be overridden
   857     for specific variables with e.g. 'transfer fixing: x y z'.
   858     The variant transfer' method allows replacing a subgoal with
   859     one that is logically stronger (rather than equivalent).
   860 
   861   - relator_eq attribute: Collects identity laws for relators of
   862     various type constructors, e.g. "list_all2 (op =) = (op =)". The
   863     transfer method uses these lemmas to infer transfer rules for
   864     non-polymorphic constants on the fly.
   865 
   866   - transfer_prover method: Assists with proving a transfer rule for a
   867     new constant, provided the constant is defined in terms of other
   868     constants that already have transfer rules. It should be applied
   869     after unfolding the constant definitions.
   870 
   871   - HOL/ex/Transfer_Int_Nat.thy: Example theory demonstrating transfer
   872     from type nat to type int.
   873 
   874 * New Lifting package:
   875 
   876   - lift_definition command: Defines operations on an abstract type in
   877     terms of a corresponding operation on a representation type. Example
   878     syntax:
   879 
   880     lift_definition dlist_insert :: "'a => 'a dlist => 'a dlist"
   881       is List.insert
   882 
   883     Users must discharge a respectfulness proof obligation when each
   884     constant is defined. (For a type copy, i.e. a typedef with UNIV,
   885     the proof is discharged automatically.) The obligation is
   886     presented in a user-friendly, readable form; a respectfulness
   887     theorem in the standard format and a transfer rule are generated
   888     by the package.
   889 
   890   - Integration with code_abstype: For typedefs (e.g. subtypes
   891     corresponding to a datatype invariant, such as dlist),
   892     lift_definition generates a code certificate theorem and sets up
   893     code generation for each constant.
   894 
   895   - setup_lifting command: Sets up the Lifting package to work with
   896     a user-defined type. The user must provide either a quotient
   897     theorem or a type_definition theorem. The package configures
   898     transfer rules for equality and quantifiers on the type, and sets
   899     up the lift_definition command to work with the type.
   900 
   901   - Usage examples: See Quotient_Examples/Lift_DList.thy,
   902     Quotient_Examples/Lift_RBT.thy, Word/Word.thy and
   903     Library/Float.thy.
   904 
   905 * Quotient package:
   906 
   907   - The 'quotient_type' command now supports a 'morphisms' option with
   908     rep and abs functions, similar to typedef.
   909 
   910   - 'quotient_type' sets up new types to work with the Lifting and
   911     Transfer packages, as with 'setup_lifting'.
   912 
   913   - The 'quotient_definition' command now requires the user to prove a
   914     respectfulness property at the point where the constant is
   915     defined, similar to lift_definition; INCOMPATIBILITY.
   916 
   917   - Renamed predicate 'Quotient' to 'Quotient3', and renamed theorems
   918     accordingly, INCOMPATIBILITY.
   919 
   920 * New diagnostic command 'find_unused_assms' to find potentially
   921 superfluous assumptions in theorems using Quickcheck.
   922 
   923 * Quickcheck:
   924 
   925   - Quickcheck returns variable assignments as counterexamples, which
   926     allows to reveal the underspecification of functions under test.
   927     For example, refuting "hd xs = x", it presents the variable
   928     assignment xs = [] and x = a1 as a counterexample, assuming that
   929     any property is false whenever "hd []" occurs in it.
   930 
   931     These counterexample are marked as potentially spurious, as
   932     Quickcheck also returns "xs = []" as a counterexample to the
   933     obvious theorem "hd xs = hd xs".
   934 
   935     After finding a potentially spurious counterexample, Quickcheck
   936     continues searching for genuine ones.
   937 
   938     By default, Quickcheck shows potentially spurious and genuine
   939     counterexamples. The option "genuine_only" sets quickcheck to only
   940     show genuine counterexamples.
   941 
   942   - The command 'quickcheck_generator' creates random and exhaustive
   943     value generators for a given type and operations.
   944 
   945     It generates values by using the operations as if they were
   946     constructors of that type.
   947 
   948   - Support for multisets.
   949 
   950   - Added "use_subtype" options.
   951 
   952   - Added "quickcheck_locale" configuration to specify how to process
   953     conjectures in a locale context.
   954 
   955 * Nitpick:
   956   - Fixed infinite loop caused by the 'peephole_optim' option and
   957     affecting 'rat' and 'real'.
   958 
   959 * Sledgehammer:
   960   - Integrated more tightly with SPASS, as described in the ITP 2012 paper "More
   961     SPASS with Isabelle".
   962   - Made it try "smt" as a fallback if "metis" fails or times out.
   963   - Added support for the following provers: Alt-Ergo (via Why3 and TFF1),
   964     iProver, iProver-Eq.
   965   - Replaced remote E-SInE with remote Satallax in the default setup.
   966   - Sped up the minimizer.
   967   - Added "lam_trans", "uncurry_aliases", and "minimize" options.
   968   - Renamed "slicing" ("no_slicing") option to "slice" ("dont_slice").
   969   - Renamed "sound" option to "strict".
   970 
   971 * Metis:
   972   - Added possibility to specify lambda translations scheme as a
   973     parenthesized argument (e.g., "by (metis (lifting) ...)").
   974 
   975 * SMT:
   976   - Renamed "smt_fixed" option to "smt_read_only_certificates".
   977 
   978 * Command 'try0':
   979   - Renamed from 'try_methods'. INCOMPATIBILITY.
   980 
   981 * New "eventually_elim" method as a generalized variant of the
   982 eventually_elim* rules. Supports structured proofs.
   983 
   984 * HOL/TPTP: support to parse and import TPTP problems (all languages)
   985 into Isabelle/HOL.
   986 
   987 
   988 *** FOL ***
   989 
   990 * New "case_product" attribute (see HOL).
   991 
   992 
   993 *** ZF ***
   994 
   995 * Greater support for structured proofs involving induction or case
   996 analysis.
   997 
   998 * Much greater use of mathematical symbols.
   999 
  1000 * Removal of many ML theorem bindings.  INCOMPATIBILITY.
  1001 
  1002 
  1003 *** ML ***
  1004 
  1005 * Antiquotation @{keyword "name"} produces a parser for outer syntax
  1006 from a minor keyword introduced via theory header declaration.
  1007 
  1008 * Antiquotation @{command_spec "name"} produces the
  1009 Outer_Syntax.command_spec from a major keyword introduced via theory
  1010 header declaration; it can be passed to Outer_Syntax.command etc.
  1011 
  1012 * Local_Theory.define no longer hard-wires default theorem name
  1013 "foo_def", but retains the binding as given.  If that is Binding.empty
  1014 / Attrib.empty_binding, the result is not registered as user-level
  1015 fact.  The Local_Theory.define_internal variant allows to specify a
  1016 non-empty name (used for the foundation in the background theory),
  1017 while omitting the fact binding in the user-context.  Potential
  1018 INCOMPATIBILITY for derived definitional packages: need to specify
  1019 naming policy for primitive definitions more explicitly.
  1020 
  1021 * Renamed Thm.capply to Thm.apply, and Thm.cabs to Thm.lambda in
  1022 conformance with similar operations in structure Term and Logic.
  1023 
  1024 * Antiquotation @{attributes [...]} embeds attribute source
  1025 representation into the ML text, which is particularly useful with
  1026 declarations like Local_Theory.note.
  1027 
  1028 * Structure Proof_Context follows standard naming scheme.  Old
  1029 ProofContext has been discontinued.  INCOMPATIBILITY.
  1030 
  1031 * Refined Local_Theory.declaration {syntax, pervasive}, with subtle
  1032 change of semantics: update is applied to auxiliary local theory
  1033 context as well.
  1034 
  1035 * Modernized some old-style infix operations:
  1036 
  1037   addeqcongs    ~> Simplifier.add_eqcong
  1038   deleqcongs    ~> Simplifier.del_eqcong
  1039   addcongs      ~> Simplifier.add_cong
  1040   delcongs      ~> Simplifier.del_cong
  1041   setmksimps    ~> Simplifier.set_mksimps
  1042   setmkcong     ~> Simplifier.set_mkcong
  1043   setmksym      ~> Simplifier.set_mksym
  1044   setmkeqTrue   ~> Simplifier.set_mkeqTrue
  1045   settermless   ~> Simplifier.set_termless
  1046   setsubgoaler  ~> Simplifier.set_subgoaler
  1047   addsplits     ~> Splitter.add_split
  1048   delsplits     ~> Splitter.del_split
  1049 
  1050 
  1051 *** System ***
  1052 
  1053 * USER_HOME settings variable points to cross-platform user home
  1054 directory, which coincides with HOME on POSIX systems only.  Likewise,
  1055 the Isabelle path specification "~" now expands to $USER_HOME, instead
  1056 of former $HOME.  A different default for USER_HOME may be set
  1057 explicitly in shell environment, before Isabelle settings are
  1058 evaluated.  Minor INCOMPATIBILITY: need to adapt Isabelle path where
  1059 the generic user home was intended.
  1060 
  1061 * ISABELLE_JDK_HOME settings variable points to JDK with javac and jar
  1062 (not just JRE).
  1063 
  1064 * ISABELLE_HOME_WINDOWS refers to ISABELLE_HOME in windows file name
  1065 notation, which is useful for the jEdit file browser, for example.
  1066 
  1067 
  1068 
  1069 New in Isabelle2011-1 (October 2011)
  1070 ------------------------------------
  1071 
  1072 *** General ***
  1073 
  1074 * Improved Isabelle/jEdit Prover IDE (PIDE), which can be invoked as
  1075 "isabelle jedit" or "ISABELLE_HOME/Isabelle" on the command line.
  1076 
  1077   - Management of multiple theory files directly from the editor
  1078     buffer store -- bypassing the file-system (no requirement to save
  1079     files for checking).
  1080 
  1081   - Markup of formal entities within the text buffer, with semantic
  1082     highlighting, tooltips and hyperlinks to jump to defining source
  1083     positions.
  1084 
  1085   - Improved text rendering, with sub/superscripts in the source
  1086     buffer (including support for copy/paste wrt. output panel, HTML
  1087     theory output and other non-Isabelle text boxes).
  1088 
  1089   - Refined scheduling of proof checking and printing of results,
  1090     based on interactive editor view.  (Note: jEdit folding and
  1091     narrowing allows to restrict buffer perspectives explicitly.)
  1092 
  1093   - Reduced CPU performance requirements, usable on machines with few
  1094     cores.
  1095 
  1096   - Reduced memory requirements due to pruning of unused document
  1097     versions (garbage collection).
  1098 
  1099 See also ~~/src/Tools/jEdit/README.html for further information,
  1100 including some remaining limitations.
  1101 
  1102 * Theory loader: source files are exclusively located via the master
  1103 directory of each theory node (where the .thy file itself resides).
  1104 The global load path (such as src/HOL/Library) has been discontinued.
  1105 Note that the path element ~~ may be used to reference theories in the
  1106 Isabelle home folder -- for instance, "~~/src/HOL/Library/FuncSet".
  1107 INCOMPATIBILITY.
  1108 
  1109 * Theory loader: source files are identified by content via SHA1
  1110 digests.  Discontinued former path/modtime identification and optional
  1111 ISABELLE_FILE_IDENT plugin scripts.
  1112 
  1113 * Parallelization of nested Isar proofs is subject to
  1114 Goal.parallel_proofs_threshold (default 100).  See also isabelle
  1115 usedir option -Q.
  1116 
  1117 * Name space: former unsynchronized references are now proper
  1118 configuration options, with more conventional names:
  1119 
  1120   long_names   ~> names_long
  1121   short_names  ~> names_short
  1122   unique_names ~> names_unique
  1123 
  1124 Minor INCOMPATIBILITY, need to declare options in context like this:
  1125 
  1126   declare [[names_unique = false]]
  1127 
  1128 * Literal facts `prop` may contain dummy patterns, e.g. `_ = _`.  Note
  1129 that the result needs to be unique, which means fact specifications
  1130 may have to be refined after enriching a proof context.
  1131 
  1132 * Attribute "case_names" has been refined: the assumptions in each case
  1133 can be named now by following the case name with [name1 name2 ...].
  1134 
  1135 * Isabelle/Isar reference manual has been updated and extended:
  1136   - "Synopsis" provides a catalog of main Isar language concepts.
  1137   - Formal references in syntax diagrams, via @{rail} antiquotation.
  1138   - Updated material from classic "ref" manual, notably about
  1139     "Classical Reasoner".
  1140 
  1141 
  1142 *** HOL ***
  1143 
  1144 * Class bot and top require underlying partial order rather than
  1145 preorder: uniqueness of bot and top is guaranteed.  INCOMPATIBILITY.
  1146 
  1147 * Class complete_lattice: generalized a couple of lemmas from sets;
  1148 generalized theorems INF_cong and SUP_cong.  New type classes for
  1149 complete boolean algebras and complete linear orders.  Lemmas
  1150 Inf_less_iff, less_Sup_iff, INF_less_iff, less_SUP_iff now reside in
  1151 class complete_linorder.
  1152 
  1153 Changed proposition of lemmas Inf_bool_def, Sup_bool_def, Inf_fun_def,
  1154 Sup_fun_def, Inf_apply, Sup_apply.
  1155 
  1156 Removed redundant lemmas (the right hand side gives hints how to
  1157 replace them for (metis ...), or (simp only: ...) proofs):
  1158 
  1159   Inf_singleton ~> Inf_insert [where A="{}", unfolded Inf_empty inf_top_right]
  1160   Sup_singleton ~> Sup_insert [where A="{}", unfolded Sup_empty sup_bot_right]
  1161   Inf_binary ~> Inf_insert, Inf_empty, and inf_top_right
  1162   Sup_binary ~> Sup_insert, Sup_empty, and sup_bot_right
  1163   Int_eq_Inter ~> Inf_insert, Inf_empty, and inf_top_right
  1164   Un_eq_Union ~> Sup_insert, Sup_empty, and sup_bot_right
  1165   Inter_def ~> INF_def, image_def
  1166   Union_def ~> SUP_def, image_def
  1167   INT_eq ~> INF_def, and image_def
  1168   UN_eq ~> SUP_def, and image_def
  1169   INF_subset ~> INF_superset_mono [OF _ order_refl]
  1170 
  1171 More consistent and comprehensive names:
  1172 
  1173   INTER_eq_Inter_image ~> INF_def
  1174   UNION_eq_Union_image ~> SUP_def
  1175   INFI_def ~> INF_def
  1176   SUPR_def ~> SUP_def
  1177   INF_leI ~> INF_lower
  1178   INF_leI2 ~> INF_lower2
  1179   le_INFI ~> INF_greatest
  1180   le_SUPI ~> SUP_upper
  1181   le_SUPI2 ~> SUP_upper2
  1182   SUP_leI ~> SUP_least
  1183   INFI_bool_eq ~> INF_bool_eq
  1184   SUPR_bool_eq ~> SUP_bool_eq
  1185   INFI_apply ~> INF_apply
  1186   SUPR_apply ~> SUP_apply
  1187   INTER_def ~> INTER_eq
  1188   UNION_def ~> UNION_eq
  1189 
  1190 INCOMPATIBILITY.
  1191 
  1192 * Renamed theory Complete_Lattice to Complete_Lattices.
  1193 INCOMPATIBILITY.
  1194 
  1195 * Theory Complete_Lattices: lemmas Inf_eq_top_iff, INF_eq_top_iff,
  1196 INF_image, Inf_insert, INF_top, Inf_top_conv, INF_top_conv, SUP_bot,
  1197 Sup_bot_conv, SUP_bot_conv, Sup_eq_top_iff, SUP_eq_top_iff, SUP_image,
  1198 Sup_insert are now declared as [simp].  INCOMPATIBILITY.
  1199 
  1200 * Theory Lattice: lemmas compl_inf_bot, compl_le_comp_iff,
  1201 compl_sup_top, inf_idem, inf_left_idem, inf_sup_absorb, sup_idem,
  1202 sup_inf_absob, sup_left_idem are now declared as [simp].  Minor
  1203 INCOMPATIBILITY.
  1204 
  1205 * Added syntactic classes "inf" and "sup" for the respective
  1206 constants.  INCOMPATIBILITY: Changes in the argument order of the
  1207 (mostly internal) locale predicates for some derived classes.
  1208 
  1209 * Theorem collections ball_simps and bex_simps do not contain theorems
  1210 referring to UNION any longer; these have been moved to collection
  1211 UN_ball_bex_simps.  INCOMPATIBILITY.
  1212 
  1213 * Theory Archimedean_Field: floor now is defined as parameter of a
  1214 separate type class floor_ceiling.
  1215 
  1216 * Theory Finite_Set: more coherent development of fold_set locales:
  1217 
  1218     locale fun_left_comm ~> locale comp_fun_commute
  1219     locale fun_left_comm_idem ~> locale comp_fun_idem
  1220 
  1221 Both use point-free characterization; interpretation proofs may need
  1222 adjustment.  INCOMPATIBILITY.
  1223 
  1224 * Theory Limits: Type "'a net" has been renamed to "'a filter", in
  1225 accordance with standard mathematical terminology. INCOMPATIBILITY.
  1226 
  1227 * Theory Complex_Main: The locale interpretations for the
  1228 bounded_linear and bounded_bilinear locales have been removed, in
  1229 order to reduce the number of duplicate lemmas. Users must use the
  1230 original names for distributivity theorems, potential INCOMPATIBILITY.
  1231 
  1232   divide.add ~> add_divide_distrib
  1233   divide.diff ~> diff_divide_distrib
  1234   divide.setsum ~> setsum_divide_distrib
  1235   mult.add_right ~> right_distrib
  1236   mult.diff_right ~> right_diff_distrib
  1237   mult_right.setsum ~> setsum_right_distrib
  1238   mult_left.diff ~> left_diff_distrib
  1239 
  1240 * Theory Complex_Main: Several redundant theorems have been removed or
  1241 replaced by more general versions. INCOMPATIBILITY.
  1242 
  1243   real_diff_def ~> minus_real_def
  1244   real_divide_def ~> divide_real_def
  1245   real_less_def ~> less_le
  1246   real_abs_def ~> abs_real_def
  1247   real_sgn_def ~> sgn_real_def
  1248   real_mult_commute ~> mult_commute
  1249   real_mult_assoc ~> mult_assoc
  1250   real_mult_1 ~> mult_1_left
  1251   real_add_mult_distrib ~> left_distrib
  1252   real_zero_not_eq_one ~> zero_neq_one
  1253   real_mult_inverse_left ~> left_inverse
  1254   INVERSE_ZERO ~> inverse_zero
  1255   real_le_refl ~> order_refl
  1256   real_le_antisym ~> order_antisym
  1257   real_le_trans ~> order_trans
  1258   real_le_linear ~> linear
  1259   real_le_eq_diff ~> le_iff_diff_le_0
  1260   real_add_left_mono ~> add_left_mono
  1261   real_mult_order ~> mult_pos_pos
  1262   real_mult_less_mono2 ~> mult_strict_left_mono
  1263   real_of_int_real_of_nat ~> real_of_int_of_nat_eq
  1264   real_0_le_divide_iff ~> zero_le_divide_iff
  1265   realpow_two_disj ~> power2_eq_iff
  1266   real_squared_diff_one_factored ~> square_diff_one_factored
  1267   realpow_two_diff ~> square_diff_square_factored
  1268   reals_complete2 ~> complete_real
  1269   real_sum_squared_expand ~> power2_sum
  1270   exp_ln_eq ~> ln_unique
  1271   expi_add ~> exp_add
  1272   expi_zero ~> exp_zero
  1273   lemma_DERIV_subst ~> DERIV_cong
  1274   LIMSEQ_Zfun_iff ~> tendsto_Zfun_iff
  1275   LIMSEQ_const ~> tendsto_const
  1276   LIMSEQ_norm ~> tendsto_norm
  1277   LIMSEQ_add ~> tendsto_add
  1278   LIMSEQ_minus ~> tendsto_minus
  1279   LIMSEQ_minus_cancel ~> tendsto_minus_cancel
  1280   LIMSEQ_diff ~> tendsto_diff
  1281   bounded_linear.LIMSEQ ~> bounded_linear.tendsto
  1282   bounded_bilinear.LIMSEQ ~> bounded_bilinear.tendsto
  1283   LIMSEQ_mult ~> tendsto_mult
  1284   LIMSEQ_inverse ~> tendsto_inverse
  1285   LIMSEQ_divide ~> tendsto_divide
  1286   LIMSEQ_pow ~> tendsto_power
  1287   LIMSEQ_setsum ~> tendsto_setsum
  1288   LIMSEQ_setprod ~> tendsto_setprod
  1289   LIMSEQ_norm_zero ~> tendsto_norm_zero_iff
  1290   LIMSEQ_rabs_zero ~> tendsto_rabs_zero_iff
  1291   LIMSEQ_imp_rabs ~> tendsto_rabs
  1292   LIMSEQ_add_minus ~> tendsto_add [OF _ tendsto_minus]
  1293   LIMSEQ_add_const ~> tendsto_add [OF _ tendsto_const]
  1294   LIMSEQ_diff_const ~> tendsto_diff [OF _ tendsto_const]
  1295   LIMSEQ_Complex ~> tendsto_Complex
  1296   LIM_ident ~> tendsto_ident_at
  1297   LIM_const ~> tendsto_const
  1298   LIM_add ~> tendsto_add
  1299   LIM_add_zero ~> tendsto_add_zero
  1300   LIM_minus ~> tendsto_minus
  1301   LIM_diff ~> tendsto_diff
  1302   LIM_norm ~> tendsto_norm
  1303   LIM_norm_zero ~> tendsto_norm_zero
  1304   LIM_norm_zero_cancel ~> tendsto_norm_zero_cancel
  1305   LIM_norm_zero_iff ~> tendsto_norm_zero_iff
  1306   LIM_rabs ~> tendsto_rabs
  1307   LIM_rabs_zero ~> tendsto_rabs_zero
  1308   LIM_rabs_zero_cancel ~> tendsto_rabs_zero_cancel
  1309   LIM_rabs_zero_iff ~> tendsto_rabs_zero_iff
  1310   LIM_compose ~> tendsto_compose
  1311   LIM_mult ~> tendsto_mult
  1312   LIM_scaleR ~> tendsto_scaleR
  1313   LIM_of_real ~> tendsto_of_real
  1314   LIM_power ~> tendsto_power
  1315   LIM_inverse ~> tendsto_inverse
  1316   LIM_sgn ~> tendsto_sgn
  1317   isCont_LIM_compose ~> isCont_tendsto_compose
  1318   bounded_linear.LIM ~> bounded_linear.tendsto
  1319   bounded_linear.LIM_zero ~> bounded_linear.tendsto_zero
  1320   bounded_bilinear.LIM ~> bounded_bilinear.tendsto
  1321   bounded_bilinear.LIM_prod_zero ~> bounded_bilinear.tendsto_zero
  1322   bounded_bilinear.LIM_left_zero ~> bounded_bilinear.tendsto_left_zero
  1323   bounded_bilinear.LIM_right_zero ~> bounded_bilinear.tendsto_right_zero
  1324   LIM_inverse_fun ~> tendsto_inverse [OF tendsto_ident_at]
  1325 
  1326 * Theory Complex_Main: The definition of infinite series was
  1327 generalized.  Now it is defined on the type class {topological_space,
  1328 comm_monoid_add}.  Hence it is useable also for extended real numbers.
  1329 
  1330 * Theory Complex_Main: The complex exponential function "expi" is now
  1331 a type-constrained abbreviation for "exp :: complex => complex"; thus
  1332 several polymorphic lemmas about "exp" are now applicable to "expi".
  1333 
  1334 * Code generation:
  1335 
  1336   - Theory Library/Code_Char_ord provides native ordering of
  1337     characters in the target language.
  1338 
  1339   - Commands code_module and code_library are legacy, use export_code
  1340     instead.
  1341 
  1342   - Method "evaluation" is legacy, use method "eval" instead.
  1343 
  1344   - Legacy evaluator "SML" is deactivated by default.  May be
  1345     reactivated by the following theory command:
  1346 
  1347       setup {* Value.add_evaluator ("SML", Codegen.eval_term) *}
  1348 
  1349 * Declare ext [intro] by default.  Rare INCOMPATIBILITY.
  1350 
  1351 * New proof method "induction" that gives induction hypotheses the
  1352 name "IH", thus distinguishing them from further hypotheses that come
  1353 from rule induction.  The latter are still called "hyps".  Method
  1354 "induction" is a thin wrapper around "induct" and follows the same
  1355 syntax.
  1356 
  1357 * Method "fastsimp" has been renamed to "fastforce", but "fastsimp" is
  1358 still available as a legacy feature for some time.
  1359 
  1360 * Nitpick:
  1361   - Added "need" and "total_consts" options.
  1362   - Reintroduced "show_skolems" option by popular demand.
  1363   - Renamed attribute: nitpick_def ~> nitpick_unfold.
  1364     INCOMPATIBILITY.
  1365 
  1366 * Sledgehammer:
  1367   - Use quasi-sound (and efficient) translations by default.
  1368   - Added support for the following provers: E-ToFoF, LEO-II,
  1369     Satallax, SNARK, Waldmeister, and Z3 with TPTP syntax.
  1370   - Automatically preplay and minimize proofs before showing them if
  1371     this can be done within reasonable time.
  1372   - sledgehammer available_provers ~> sledgehammer supported_provers.
  1373     INCOMPATIBILITY.
  1374   - Added "preplay_timeout", "slicing", "type_enc", "sound",
  1375     "max_mono_iters", and "max_new_mono_instances" options.
  1376   - Removed "explicit_apply" and "full_types" options as well as "Full
  1377     Types" Proof General menu item. INCOMPATIBILITY.
  1378 
  1379 * Metis:
  1380   - Removed "metisF" -- use "metis" instead. INCOMPATIBILITY.
  1381   - Obsoleted "metisFT" -- use "metis (full_types)" instead.
  1382     INCOMPATIBILITY.
  1383 
  1384 * Command 'try':
  1385   - Renamed 'try_methods' and added "simp:", "intro:", "dest:", and
  1386     "elim:" options. INCOMPATIBILITY.
  1387   - Introduced 'try' that not only runs 'try_methods' but also
  1388     'solve_direct', 'sledgehammer', 'quickcheck', and 'nitpick'.
  1389 
  1390 * Quickcheck:
  1391   - Added "eval" option to evaluate terms for the found counterexample
  1392     (currently only supported by the default (exhaustive) tester).
  1393   - Added post-processing of terms to obtain readable counterexamples
  1394     (currently only supported by the default (exhaustive) tester).
  1395   - New counterexample generator quickcheck[narrowing] enables
  1396     narrowing-based testing.  Requires the Glasgow Haskell compiler
  1397     with its installation location defined in the Isabelle settings
  1398     environment as ISABELLE_GHC.
  1399   - Removed quickcheck tester "SML" based on the SML code generator
  1400     (formly in HOL/Library).
  1401 
  1402 * Function package: discontinued option "tailrec".  INCOMPATIBILITY,
  1403 use 'partial_function' instead.
  1404 
  1405 * Theory Library/Extended_Reals replaces now the positive extended
  1406 reals found in probability theory. This file is extended by
  1407 Multivariate_Analysis/Extended_Real_Limits.
  1408 
  1409 * Theory Library/Old_Recdef: old 'recdef' package has been moved here,
  1410 from where it must be imported explicitly if it is really required.
  1411 INCOMPATIBILITY.
  1412 
  1413 * Theory Library/Wfrec: well-founded recursion combinator "wfrec" has
  1414 been moved here.  INCOMPATIBILITY.
  1415 
  1416 * Theory Library/Saturated provides type of numbers with saturated
  1417 arithmetic.
  1418 
  1419 * Theory Library/Product_Lattice defines a pointwise ordering for the
  1420 product type 'a * 'b, and provides instance proofs for various order
  1421 and lattice type classes.
  1422 
  1423 * Theory Library/Countable now provides the "countable_datatype" proof
  1424 method for proving "countable" class instances for datatypes.
  1425 
  1426 * Theory Library/Cset_Monad allows do notation for computable sets
  1427 (cset) via the generic monad ad-hoc overloading facility.
  1428 
  1429 * Library: Theories of common data structures are split into theories
  1430 for implementation, an invariant-ensuring type, and connection to an
  1431 abstract type. INCOMPATIBILITY.
  1432 
  1433   - RBT is split into RBT and RBT_Mapping.
  1434   - AssocList is split and renamed into AList and AList_Mapping.
  1435   - DList is split into DList_Impl, DList, and DList_Cset.
  1436   - Cset is split into Cset and List_Cset.
  1437 
  1438 * Theory Library/Nat_Infinity has been renamed to
  1439 Library/Extended_Nat, with name changes of the following types and
  1440 constants:
  1441 
  1442   type inat   ~> type enat
  1443   Fin         ~> enat
  1444   Infty       ~> infinity (overloaded)
  1445   iSuc        ~> eSuc
  1446   the_Fin     ~> the_enat
  1447 
  1448 Every theorem name containing "inat", "Fin", "Infty", or "iSuc" has
  1449 been renamed accordingly. INCOMPATIBILITY.
  1450 
  1451 * Session Multivariate_Analysis: The euclidean_space type class now
  1452 fixes a constant "Basis :: 'a set" consisting of the standard
  1453 orthonormal basis for the type. Users now have the option of
  1454 quantifying over this set instead of using the "basis" function, e.g.
  1455 "ALL x:Basis. P x" vs "ALL i<DIM('a). P (basis i)".
  1456 
  1457 * Session Multivariate_Analysis: Type "('a, 'b) cart" has been renamed
  1458 to "('a, 'b) vec" (the syntax "'a ^ 'b" remains unaffected). Constants
  1459 "Cart_nth" and "Cart_lambda" have been respectively renamed to
  1460 "vec_nth" and "vec_lambda"; theorems mentioning those names have
  1461 changed to match. Definition theorems for overloaded constants now use
  1462 the standard "foo_vec_def" naming scheme. A few other theorems have
  1463 been renamed as follows (INCOMPATIBILITY):
  1464 
  1465   Cart_eq          ~> vec_eq_iff
  1466   dist_nth_le_cart ~> dist_vec_nth_le
  1467   tendsto_vector   ~> vec_tendstoI
  1468   Cauchy_vector    ~> vec_CauchyI
  1469 
  1470 * Session Multivariate_Analysis: Several duplicate theorems have been
  1471 removed, and other theorems have been renamed or replaced with more
  1472 general versions. INCOMPATIBILITY.
  1473 
  1474   finite_choice ~> finite_set_choice
  1475   eventually_conjI ~> eventually_conj
  1476   eventually_and ~> eventually_conj_iff
  1477   eventually_false ~> eventually_False
  1478   setsum_norm ~> norm_setsum
  1479   Lim_sequentially ~> LIMSEQ_def
  1480   Lim_ident_at ~> LIM_ident
  1481   Lim_const ~> tendsto_const
  1482   Lim_cmul ~> tendsto_scaleR [OF tendsto_const]
  1483   Lim_neg ~> tendsto_minus
  1484   Lim_add ~> tendsto_add
  1485   Lim_sub ~> tendsto_diff
  1486   Lim_mul ~> tendsto_scaleR
  1487   Lim_vmul ~> tendsto_scaleR [OF _ tendsto_const]
  1488   Lim_null_norm ~> tendsto_norm_zero_iff [symmetric]
  1489   Lim_linear ~> bounded_linear.tendsto
  1490   Lim_component ~> tendsto_euclidean_component
  1491   Lim_component_cart ~> tendsto_vec_nth
  1492   Lim_inner ~> tendsto_inner [OF tendsto_const]
  1493   dot_lsum ~> inner_setsum_left
  1494   dot_rsum ~> inner_setsum_right
  1495   continuous_cmul ~> continuous_scaleR [OF continuous_const]
  1496   continuous_neg ~> continuous_minus
  1497   continuous_sub ~> continuous_diff
  1498   continuous_vmul ~> continuous_scaleR [OF _ continuous_const]
  1499   continuous_mul ~> continuous_scaleR
  1500   continuous_inv ~> continuous_inverse
  1501   continuous_at_within_inv ~> continuous_at_within_inverse
  1502   continuous_at_inv ~> continuous_at_inverse
  1503   continuous_at_norm ~> continuous_norm [OF continuous_at_id]
  1504   continuous_at_infnorm ~> continuous_infnorm [OF continuous_at_id]
  1505   continuous_at_component ~> continuous_component [OF continuous_at_id]
  1506   continuous_on_neg ~> continuous_on_minus
  1507   continuous_on_sub ~> continuous_on_diff
  1508   continuous_on_cmul ~> continuous_on_scaleR [OF continuous_on_const]
  1509   continuous_on_vmul ~> continuous_on_scaleR [OF _ continuous_on_const]
  1510   continuous_on_mul ~> continuous_on_scaleR
  1511   continuous_on_mul_real ~> continuous_on_mult
  1512   continuous_on_inner ~> continuous_on_inner [OF continuous_on_const]
  1513   continuous_on_norm ~> continuous_on_norm [OF continuous_on_id]
  1514   continuous_on_inverse ~> continuous_on_inv
  1515   uniformly_continuous_on_neg ~> uniformly_continuous_on_minus
  1516   uniformly_continuous_on_sub ~> uniformly_continuous_on_diff
  1517   subset_interior ~> interior_mono
  1518   subset_closure ~> closure_mono
  1519   closure_univ ~> closure_UNIV
  1520   real_arch_lt ~> reals_Archimedean2
  1521   real_arch ~> reals_Archimedean3
  1522   real_abs_norm ~> abs_norm_cancel
  1523   real_abs_sub_norm ~> norm_triangle_ineq3
  1524   norm_cauchy_schwarz_abs ~> Cauchy_Schwarz_ineq2
  1525 
  1526 * Session HOL-Probability:
  1527   - Caratheodory's extension lemma is now proved for ring_of_sets.
  1528   - Infinite products of probability measures are now available.
  1529   - Sigma closure is independent, if the generator is independent
  1530   - Use extended reals instead of positive extended
  1531     reals. INCOMPATIBILITY.
  1532 
  1533 * Session HOLCF: Discontinued legacy theorem names, INCOMPATIBILITY.
  1534 
  1535   expand_fun_below ~> fun_below_iff
  1536   below_fun_ext ~> fun_belowI
  1537   expand_cfun_eq ~> cfun_eq_iff
  1538   ext_cfun ~> cfun_eqI
  1539   expand_cfun_below ~> cfun_below_iff
  1540   below_cfun_ext ~> cfun_belowI
  1541   monofun_fun_fun ~> fun_belowD
  1542   monofun_fun_arg ~> monofunE
  1543   monofun_lub_fun ~> adm_monofun [THEN admD]
  1544   cont_lub_fun ~> adm_cont [THEN admD]
  1545   cont2cont_Rep_CFun ~> cont2cont_APP
  1546   cont_Rep_CFun_app ~> cont_APP_app
  1547   cont_Rep_CFun_app_app ~> cont_APP_app_app
  1548   cont_cfun_fun ~> cont_Rep_cfun1 [THEN contE]
  1549   cont_cfun_arg ~> cont_Rep_cfun2 [THEN contE]
  1550   contlub_cfun ~> lub_APP [symmetric]
  1551   contlub_LAM ~> lub_LAM [symmetric]
  1552   thelubI ~> lub_eqI
  1553   UU_I ~> bottomI
  1554   lift_distinct1 ~> lift.distinct(1)
  1555   lift_distinct2 ~> lift.distinct(2)
  1556   Def_not_UU ~> lift.distinct(2)
  1557   Def_inject ~> lift.inject
  1558   below_UU_iff ~> below_bottom_iff
  1559   eq_UU_iff ~> eq_bottom_iff
  1560 
  1561 
  1562 *** Document preparation ***
  1563 
  1564 * Antiquotation @{rail} layouts railroad syntax diagrams, see also
  1565 isar-ref manual, both for description and actual application of the
  1566 same.
  1567 
  1568 * Antiquotation @{value} evaluates the given term and presents its
  1569 result.
  1570 
  1571 * Antiquotations: term style "isub" provides ad-hoc conversion of
  1572 variables x1, y23 into subscripted form x\<^isub>1,
  1573 y\<^isub>2\<^isub>3.
  1574 
  1575 * Predefined LaTeX macros for Isabelle symbols \<bind> and \<then>
  1576 (e.g. see ~~/src/HOL/Library/Monad_Syntax.thy).
  1577 
  1578 * Localized \isabellestyle switch can be used within blocks or groups
  1579 like this:
  1580 
  1581   \isabellestyle{it}  %preferred default
  1582   {\isabellestylett @{text "typewriter stuff"}}
  1583 
  1584 * Discontinued special treatment of hard tabulators.  Implicit
  1585 tab-width is now defined as 1.  Potential INCOMPATIBILITY for visual
  1586 layouts.
  1587 
  1588 
  1589 *** ML ***
  1590 
  1591 * The inner syntax of sort/type/term/prop supports inlined YXML
  1592 representations within quoted string tokens.  By encoding logical
  1593 entities via Term_XML (in ML or Scala) concrete syntax can be
  1594 bypassed, which is particularly useful for producing bits of text
  1595 under external program control.
  1596 
  1597 * Antiquotations for ML and document preparation are managed as theory
  1598 data, which requires explicit setup.
  1599 
  1600 * Isabelle_Process.is_active allows tools to check if the official
  1601 process wrapper is running (Isabelle/Scala/jEdit) or the old TTY loop
  1602 (better known as Proof General).
  1603 
  1604 * Structure Proof_Context follows standard naming scheme.  Old
  1605 ProofContext is still available for some time as legacy alias.
  1606 
  1607 * Structure Timing provides various operations for timing; supersedes
  1608 former start_timing/end_timing etc.
  1609 
  1610 * Path.print is the official way to show file-system paths to users
  1611 (including quotes etc.).
  1612 
  1613 * Inner syntax: identifiers in parse trees of generic categories
  1614 "logic", "aprop", "idt" etc. carry position information (disguised as
  1615 type constraints).  Occasional INCOMPATIBILITY with non-compliant
  1616 translations that choke on unexpected type constraints.  Positions can
  1617 be stripped in ML translations via Syntax.strip_positions /
  1618 Syntax.strip_positions_ast, or via the syntax constant
  1619 "_strip_positions" within parse trees.  As last resort, positions can
  1620 be disabled via the configuration option Syntax.positions, which is
  1621 called "syntax_positions" in Isar attribute syntax.
  1622 
  1623 * Discontinued special status of various ML structures that contribute
  1624 to structure Syntax (Ast, Lexicon, Mixfix, Parser, Printer etc.): less
  1625 pervasive content, no inclusion in structure Syntax.  INCOMPATIBILITY,
  1626 refer directly to Ast.Constant, Lexicon.is_identifier,
  1627 Syntax_Trans.mk_binder_tr etc.
  1628 
  1629 * Typed print translation: discontinued show_sorts argument, which is
  1630 already available via context of "advanced" translation.
  1631 
  1632 * Refined PARALLEL_GOALS tactical: degrades gracefully for schematic
  1633 goal states; body tactic needs to address all subgoals uniformly.
  1634 
  1635 * Slightly more special eq_list/eq_set, with shortcut involving
  1636 pointer equality (assumes that eq relation is reflexive).
  1637 
  1638 * Classical tactics use proper Proof.context instead of historic types
  1639 claset/clasimpset.  Old-style declarations like addIs, addEs, addDs
  1640 operate directly on Proof.context.  Raw type claset retains its use as
  1641 snapshot of the classical context, which can be recovered via
  1642 (put_claset HOL_cs) etc.  Type clasimpset has been discontinued.
  1643 INCOMPATIBILITY, classical tactics and derived proof methods require
  1644 proper Proof.context.
  1645 
  1646 
  1647 *** System ***
  1648 
  1649 * Discontinued support for Poly/ML 5.2, which was the last version
  1650 without proper multithreading and TimeLimit implementation.
  1651 
  1652 * Discontinued old lib/scripts/polyml-platform, which has been
  1653 obsolete since Isabelle2009-2.
  1654 
  1655 * Various optional external tools are referenced more robustly and
  1656 uniformly by explicit Isabelle settings as follows:
  1657 
  1658   ISABELLE_CSDP   (formerly CSDP_EXE)
  1659   ISABELLE_GHC    (formerly EXEC_GHC or GHC_PATH)
  1660   ISABELLE_OCAML  (formerly EXEC_OCAML)
  1661   ISABELLE_SWIPL  (formerly EXEC_SWIPL)
  1662   ISABELLE_YAP    (formerly EXEC_YAP)
  1663 
  1664 Note that automated detection from the file-system or search path has
  1665 been discontinued.  INCOMPATIBILITY.
  1666 
  1667 * Scala layer provides JVM method invocation service for static
  1668 methods of type (String)String, see Invoke_Scala.method in ML.  For
  1669 example:
  1670 
  1671   Invoke_Scala.method "java.lang.System.getProperty" "java.home"
  1672 
  1673 Together with YXML.string_of_body/parse_body and XML.Encode/Decode
  1674 this allows to pass structured values between ML and Scala.
  1675 
  1676 * The IsabelleText fonts includes some further glyphs to support the
  1677 Prover IDE.  Potential INCOMPATIBILITY: users who happen to have
  1678 installed a local copy (which is normally *not* required) need to
  1679 delete or update it from ~~/lib/fonts/.
  1680 
  1681 
  1682 
  1683 New in Isabelle2011 (January 2011)
  1684 ----------------------------------
  1685 
  1686 *** General ***
  1687 
  1688 * Experimental Prover IDE based on Isabelle/Scala and jEdit (see
  1689 src/Tools/jEdit).  This also serves as IDE for Isabelle/ML, with
  1690 useful tooltips and hyperlinks produced from its static analysis.  The
  1691 bundled component provides an executable Isabelle tool that can be run
  1692 like this:
  1693 
  1694   Isabelle2011/bin/isabelle jedit
  1695 
  1696 * Significantly improved Isabelle/Isar implementation manual.
  1697 
  1698 * System settings: ISABELLE_HOME_USER now includes ISABELLE_IDENTIFIER
  1699 (and thus refers to something like $HOME/.isabelle/Isabelle2011),
  1700 while the default heap location within that directory lacks that extra
  1701 suffix.  This isolates multiple Isabelle installations from each
  1702 other, avoiding problems with old settings in new versions.
  1703 INCOMPATIBILITY, need to copy/upgrade old user settings manually.
  1704 
  1705 * Source files are always encoded as UTF-8, instead of old-fashioned
  1706 ISO-Latin-1.  INCOMPATIBILITY.  Isabelle LaTeX documents might require
  1707 the following package declarations:
  1708 
  1709   \usepackage[utf8]{inputenc}
  1710   \usepackage{textcomp}
  1711 
  1712 * Explicit treatment of UTF-8 sequences as Isabelle symbols, such that
  1713 a Unicode character is treated as a single symbol, not a sequence of
  1714 non-ASCII bytes as before.  Since Isabelle/ML string literals may
  1715 contain symbols without further backslash escapes, Unicode can now be
  1716 used here as well.  Recall that Symbol.explode in ML provides a
  1717 consistent view on symbols, while raw explode (or String.explode)
  1718 merely give a byte-oriented representation.
  1719 
  1720 * Theory loader: source files are primarily located via the master
  1721 directory of each theory node (where the .thy file itself resides).
  1722 The global load path is still partially available as legacy feature.
  1723 Minor INCOMPATIBILITY due to subtle change in file lookup: use
  1724 explicit paths, relatively to the theory.
  1725 
  1726 * Special treatment of ML file names has been discontinued.
  1727 Historically, optional extensions .ML or .sml were added on demand --
  1728 at the cost of clarity of file dependencies.  Recall that Isabelle/ML
  1729 files exclusively use the .ML extension.  Minor INCOMPATIBILTY.
  1730 
  1731 * Various options that affect pretty printing etc. are now properly
  1732 handled within the context via configuration options, instead of
  1733 unsynchronized references or print modes.  There are both ML Config.T
  1734 entities and Isar declaration attributes to access these.
  1735 
  1736   ML (Config.T)                 Isar (attribute)
  1737 
  1738   eta_contract                  eta_contract
  1739   show_brackets                 show_brackets
  1740   show_sorts                    show_sorts
  1741   show_types                    show_types
  1742   show_question_marks           show_question_marks
  1743   show_consts                   show_consts
  1744   show_abbrevs                  show_abbrevs
  1745 
  1746   Syntax.ast_trace              syntax_ast_trace
  1747   Syntax.ast_stat               syntax_ast_stat
  1748   Syntax.ambiguity_level        syntax_ambiguity_level
  1749 
  1750   Goal_Display.goals_limit      goals_limit
  1751   Goal_Display.show_main_goal   show_main_goal
  1752 
  1753   Method.rule_trace             rule_trace
  1754 
  1755   Thy_Output.display            thy_output_display
  1756   Thy_Output.quotes             thy_output_quotes
  1757   Thy_Output.indent             thy_output_indent
  1758   Thy_Output.source             thy_output_source
  1759   Thy_Output.break              thy_output_break
  1760 
  1761 Note that corresponding "..._default" references in ML may only be
  1762 changed globally at the ROOT session setup, but *not* within a theory.
  1763 The option "show_abbrevs" supersedes the former print mode
  1764 "no_abbrevs" with inverted meaning.
  1765 
  1766 * More systematic naming of some configuration options.
  1767 INCOMPATIBILITY.
  1768 
  1769   trace_simp  ~>  simp_trace
  1770   debug_simp  ~>  simp_debug
  1771 
  1772 * Support for real valued configuration options, using simplistic
  1773 floating-point notation that coincides with the inner syntax for
  1774 float_token.
  1775 
  1776 * Support for real valued preferences (with approximative PGIP type):
  1777 front-ends need to accept "pgint" values in float notation.
  1778 INCOMPATIBILITY.
  1779 
  1780 * The IsabelleText font now includes Cyrillic, Hebrew, Arabic from
  1781 DejaVu Sans.
  1782 
  1783 * Discontinued support for Poly/ML 5.0 and 5.1 versions.
  1784 
  1785 
  1786 *** Pure ***
  1787 
  1788 * Command 'type_synonym' (with single argument) replaces somewhat
  1789 outdated 'types', which is still available as legacy feature for some
  1790 time.
  1791 
  1792 * Command 'nonterminal' (with 'and' separated list of arguments)
  1793 replaces somewhat outdated 'nonterminals'.  INCOMPATIBILITY.
  1794 
  1795 * Command 'notepad' replaces former 'example_proof' for
  1796 experimentation in Isar without any result.  INCOMPATIBILITY.
  1797 
  1798 * Locale interpretation commands 'interpret' and 'sublocale' accept
  1799 lists of equations to map definitions in a locale to appropriate
  1800 entities in the context of the interpretation.  The 'interpretation'
  1801 command already provided this functionality.
  1802 
  1803 * Diagnostic command 'print_dependencies' prints the locale instances
  1804 that would be activated if the specified expression was interpreted in
  1805 the current context.  Variant "print_dependencies!" assumes a context
  1806 without interpretations.
  1807 
  1808 * Diagnostic command 'print_interps' prints interpretations in proofs
  1809 in addition to interpretations in theories.
  1810 
  1811 * Discontinued obsolete 'global' and 'local' commands to manipulate
  1812 the theory name space.  Rare INCOMPATIBILITY.  The ML functions
  1813 Sign.root_path and Sign.local_path may be applied directly where this
  1814 feature is still required for historical reasons.
  1815 
  1816 * Discontinued obsolete 'constdefs' command.  INCOMPATIBILITY, use
  1817 'definition' instead.
  1818 
  1819 * The "prems" fact, which refers to the accidental collection of
  1820 foundational premises in the context, is now explicitly marked as
  1821 legacy feature and will be discontinued soon.  Consider using "assms"
  1822 of the head statement or reference facts by explicit names.
  1823 
  1824 * Document antiquotations @{class} and @{type} print classes and type
  1825 constructors.
  1826 
  1827 * Document antiquotation @{file} checks file/directory entries within
  1828 the local file system.
  1829 
  1830 
  1831 *** HOL ***
  1832 
  1833 * Coercive subtyping: functions can be declared as coercions and type
  1834 inference will add them as necessary upon input of a term.  Theory
  1835 Complex_Main declares real :: nat => real and real :: int => real as
  1836 coercions. A coercion function f is declared like this:
  1837 
  1838   declare [[coercion f]]
  1839 
  1840 To lift coercions through type constructors (e.g. from nat => real to
  1841 nat list => real list), map functions can be declared, e.g.
  1842 
  1843   declare [[coercion_map map]]
  1844 
  1845 Currently coercion inference is activated only in theories including
  1846 real numbers, i.e. descendants of Complex_Main.  This is controlled by
  1847 the configuration option "coercion_enabled", e.g. it can be enabled in
  1848 other theories like this:
  1849 
  1850   declare [[coercion_enabled]]
  1851 
  1852 * Command 'partial_function' provides basic support for recursive
  1853 function definitions over complete partial orders.  Concrete instances
  1854 are provided for i) the option type, ii) tail recursion on arbitrary
  1855 types, and iii) the heap monad of Imperative_HOL.  See
  1856 src/HOL/ex/Fundefs.thy and src/HOL/Imperative_HOL/ex/Linked_Lists.thy
  1857 for examples.
  1858 
  1859 * Function package: f.psimps rules are no longer implicitly declared
  1860 as [simp].  INCOMPATIBILITY.
  1861 
  1862 * Datatype package: theorems generated for executable equality (class
  1863 "eq") carry proper names and are treated as default code equations.
  1864 
  1865 * Inductive package: now offers command 'inductive_simps' to
  1866 automatically derive instantiated and simplified equations for
  1867 inductive predicates, similar to 'inductive_cases'.
  1868 
  1869 * Command 'enriched_type' allows to register properties of the
  1870 functorial structure of types.
  1871 
  1872 * Improved infrastructure for term evaluation using code generator
  1873 techniques, in particular static evaluation conversions.
  1874 
  1875 * Code generator: Scala (2.8 or higher) has been added to the target
  1876 languages.
  1877 
  1878 * Code generator: globbing constant expressions "*" and "Theory.*"
  1879 have been replaced by the more idiomatic "_" and "Theory._".
  1880 INCOMPATIBILITY.
  1881 
  1882 * Code generator: export_code without explicit file declaration prints
  1883 to standard output.  INCOMPATIBILITY.
  1884 
  1885 * Code generator: do not print function definitions for case
  1886 combinators any longer.
  1887 
  1888 * Code generator: simplification with rules determined with
  1889 src/Tools/Code/code_simp.ML and method "code_simp".
  1890 
  1891 * Code generator for records: more idiomatic representation of record
  1892 types.  Warning: records are not covered by ancient SML code
  1893 generation any longer.  INCOMPATIBILITY.  In cases of need, a suitable
  1894 rep_datatype declaration helps to succeed then:
  1895 
  1896   record 'a foo = ...
  1897   ...
  1898   rep_datatype foo_ext ...
  1899 
  1900 * Records: logical foundation type for records does not carry a
  1901 '_type' suffix any longer (obsolete due to authentic syntax).
  1902 INCOMPATIBILITY.
  1903 
  1904 * Quickcheck now by default uses exhaustive testing instead of random
  1905 testing.  Random testing can be invoked by "quickcheck [random]",
  1906 exhaustive testing by "quickcheck [exhaustive]".
  1907 
  1908 * Quickcheck instantiates polymorphic types with small finite
  1909 datatypes by default. This enables a simple execution mechanism to
  1910 handle quantifiers and function equality over the finite datatypes.
  1911 
  1912 * Quickcheck random generator has been renamed from "code" to
  1913 "random".  INCOMPATIBILITY.
  1914 
  1915 * Quickcheck now has a configurable time limit which is set to 30
  1916 seconds by default. This can be changed by adding [timeout = n] to the
  1917 quickcheck command. The time limit for Auto Quickcheck is still set
  1918 independently.
  1919 
  1920 * Quickcheck in locales considers interpretations of that locale for
  1921 counter example search.
  1922 
  1923 * Sledgehammer:
  1924   - Added "smt" and "remote_smt" provers based on the "smt" proof
  1925     method. See the Sledgehammer manual for details ("isabelle doc
  1926     sledgehammer").
  1927   - Renamed commands:
  1928     sledgehammer atp_info ~> sledgehammer running_provers
  1929     sledgehammer atp_kill ~> sledgehammer kill_provers
  1930     sledgehammer available_atps ~> sledgehammer available_provers
  1931     INCOMPATIBILITY.
  1932   - Renamed options:
  1933     sledgehammer [atps = ...] ~> sledgehammer [provers = ...]
  1934     sledgehammer [atp = ...] ~> sledgehammer [prover = ...]
  1935     sledgehammer [timeout = 77 s] ~> sledgehammer [timeout = 77]
  1936     (and "ms" and "min" are no longer supported)
  1937     INCOMPATIBILITY.
  1938 
  1939 * Nitpick:
  1940   - Renamed options:
  1941     nitpick [timeout = 77 s] ~> nitpick [timeout = 77]
  1942     nitpick [tac_timeout = 777 ms] ~> nitpick [tac_timeout = 0.777]
  1943     INCOMPATIBILITY.
  1944   - Added support for partial quotient types.
  1945   - Added local versions of the "Nitpick.register_xxx" functions.
  1946   - Added "whack" option.
  1947   - Allow registration of quotient types as codatatypes.
  1948   - Improved "merge_type_vars" option to merge more types.
  1949   - Removed unsound "fast_descrs" option.
  1950   - Added custom symmetry breaking for datatypes, making it possible to reach
  1951     higher cardinalities.
  1952   - Prevent the expansion of too large definitions.
  1953 
  1954 * Proof methods "metis" and "meson" now have configuration options
  1955 "meson_trace", "metis_trace", and "metis_verbose" that can be enabled
  1956 to diagnose these tools. E.g.
  1957 
  1958     using [[metis_trace = true]]
  1959 
  1960 * Auto Solve: Renamed "Auto Solve Direct".  The tool is now available
  1961 manually as command 'solve_direct'.
  1962 
  1963 * The default SMT solver Z3 must be enabled explicitly (due to
  1964 licensing issues) by setting the environment variable
  1965 Z3_NON_COMMERCIAL in etc/settings of the component, for example.  For
  1966 commercial applications, the SMT solver CVC3 is provided as fall-back;
  1967 changing the SMT solver is done via the configuration option
  1968 "smt_solver".
  1969 
  1970 * Remote SMT solvers need to be referred to by the "remote_" prefix,
  1971 i.e. "remote_cvc3" and "remote_z3".
  1972 
  1973 * Added basic SMT support for datatypes, records, and typedefs using
  1974 the oracle mode (no proofs).  Direct support of pairs has been dropped
  1975 in exchange (pass theorems fst_conv snd_conv pair_collapse to the SMT
  1976 support for a similar behavior).  Minor INCOMPATIBILITY.
  1977 
  1978 * Changed SMT configuration options:
  1979   - Renamed:
  1980     z3_proofs ~> smt_oracle (with inverted meaning)
  1981     z3_trace_assms ~> smt_trace_used_facts
  1982     INCOMPATIBILITY.
  1983   - Added:
  1984     smt_verbose
  1985     smt_random_seed
  1986     smt_datatypes
  1987     smt_infer_triggers
  1988     smt_monomorph_limit
  1989     cvc3_options
  1990     remote_cvc3_options
  1991     remote_z3_options
  1992     yices_options
  1993 
  1994 * Boogie output files (.b2i files) need to be declared in the theory
  1995 header.
  1996 
  1997 * Simplification procedure "list_to_set_comprehension" rewrites list
  1998 comprehensions applied to List.set to set comprehensions.  Occasional
  1999 INCOMPATIBILITY, may be deactivated like this:
  2000 
  2001   declare [[simproc del: list_to_set_comprehension]]
  2002 
  2003 * Removed old version of primrec package.  INCOMPATIBILITY.
  2004 
  2005 * Removed simplifier congruence rule of "prod_case", as has for long
  2006 been the case with "split".  INCOMPATIBILITY.
  2007 
  2008 * String.literal is a type, but not a datatype.  INCOMPATIBILITY.
  2009 
  2010 * Removed [split_format ... and ... and ...] version of
  2011 [split_format].  Potential INCOMPATIBILITY.
  2012 
  2013 * Predicate "sorted" now defined inductively, with nice induction
  2014 rules.  INCOMPATIBILITY: former sorted.simps now named sorted_simps.
  2015 
  2016 * Constant "contents" renamed to "the_elem", to free the generic name
  2017 contents for other uses.  INCOMPATIBILITY.
  2018 
  2019 * Renamed class eq and constant eq (for code generation) to class
  2020 equal and constant equal, plus renaming of related facts and various
  2021 tuning.  INCOMPATIBILITY.
  2022 
  2023 * Dropped type classes mult_mono and mult_mono1.  INCOMPATIBILITY.
  2024 
  2025 * Removed output syntax "'a ~=> 'b" for "'a => 'b option".
  2026 INCOMPATIBILITY.
  2027 
  2028 * Renamed theory Fset to Cset, type Fset.fset to Cset.set, in order to
  2029 avoid confusion with finite sets.  INCOMPATIBILITY.
  2030 
  2031 * Abandoned locales equiv, congruent and congruent2 for equivalence
  2032 relations.  INCOMPATIBILITY: use equivI rather than equiv_intro (same
  2033 for congruent(2)).
  2034 
  2035 * Some previously unqualified names have been qualified:
  2036 
  2037   types
  2038     bool ~> HOL.bool
  2039     nat ~> Nat.nat
  2040 
  2041   constants
  2042     Trueprop ~> HOL.Trueprop
  2043     True ~> HOL.True
  2044     False ~> HOL.False
  2045     op & ~> HOL.conj
  2046     op | ~> HOL.disj
  2047     op --> ~> HOL.implies
  2048     op = ~> HOL.eq
  2049     Not ~> HOL.Not
  2050     The ~> HOL.The
  2051     All ~> HOL.All
  2052     Ex ~> HOL.Ex
  2053     Ex1 ~> HOL.Ex1
  2054     Let ~> HOL.Let
  2055     If ~> HOL.If
  2056     Ball ~> Set.Ball
  2057     Bex ~> Set.Bex
  2058     Suc ~> Nat.Suc
  2059     Pair ~> Product_Type.Pair
  2060     fst ~> Product_Type.fst
  2061     snd ~> Product_Type.snd
  2062     curry ~> Product_Type.curry
  2063     op : ~> Set.member
  2064     Collect ~> Set.Collect
  2065 
  2066 INCOMPATIBILITY.
  2067 
  2068 * More canonical naming convention for some fundamental definitions:
  2069 
  2070     bot_bool_eq ~> bot_bool_def
  2071     top_bool_eq ~> top_bool_def
  2072     inf_bool_eq ~> inf_bool_def
  2073     sup_bool_eq ~> sup_bool_def
  2074     bot_fun_eq  ~> bot_fun_def
  2075     top_fun_eq  ~> top_fun_def
  2076     inf_fun_eq  ~> inf_fun_def
  2077     sup_fun_eq  ~> sup_fun_def
  2078 
  2079 INCOMPATIBILITY.
  2080 
  2081 * More stylized fact names:
  2082 
  2083   expand_fun_eq ~> fun_eq_iff
  2084   expand_set_eq ~> set_eq_iff
  2085   set_ext       ~> set_eqI
  2086   nat_number    ~> eval_nat_numeral
  2087 
  2088 INCOMPATIBILITY.
  2089 
  2090 * Refactoring of code-generation specific operations in theory List:
  2091 
  2092   constants
  2093     null ~> List.null
  2094 
  2095   facts
  2096     mem_iff ~> member_def
  2097     null_empty ~> null_def
  2098 
  2099 INCOMPATIBILITY.  Note that these were not supposed to be used
  2100 regularly unless for striking reasons; their main purpose was code
  2101 generation.
  2102 
  2103 Various operations from the Haskell prelude are used for generating
  2104 Haskell code.
  2105 
  2106 * Term "bij f" is now an abbreviation of "bij_betw f UNIV UNIV".  Term
  2107 "surj f" is now an abbreviation of "range f = UNIV".  The theorems
  2108 bij_def and surj_def are unchanged.  INCOMPATIBILITY.
  2109 
  2110 * Abolished some non-alphabetic type names: "prod" and "sum" replace
  2111 "*" and "+" respectively.  INCOMPATIBILITY.
  2112 
  2113 * Name "Plus" of disjoint sum operator "<+>" is now hidden.  Write
  2114 "Sum_Type.Plus" instead.
  2115 
  2116 * Constant "split" has been merged with constant "prod_case"; names of
  2117 ML functions, facts etc. involving split have been retained so far,
  2118 though.  INCOMPATIBILITY.
  2119 
  2120 * Dropped old infix syntax "_ mem _" for List.member; use "_ : set _"
  2121 instead.  INCOMPATIBILITY.
  2122 
  2123 * Removed lemma "Option.is_none_none" which duplicates "is_none_def".
  2124 INCOMPATIBILITY.
  2125 
  2126 * Former theory Library/Enum is now part of the HOL-Main image.
  2127 INCOMPATIBILITY: all constants of the Enum theory now have to be
  2128 referred to by its qualified name.
  2129 
  2130   enum    ~>  Enum.enum
  2131   nlists  ~>  Enum.nlists
  2132   product ~>  Enum.product
  2133 
  2134 * Theory Library/Monad_Syntax provides do-syntax for monad types.
  2135 Syntax in Library/State_Monad has been changed to avoid ambiguities.
  2136 INCOMPATIBILITY.
  2137 
  2138 * Theory Library/SetsAndFunctions has been split into
  2139 Library/Function_Algebras and Library/Set_Algebras; canonical names
  2140 for instance definitions for functions; various improvements.
  2141 INCOMPATIBILITY.
  2142 
  2143 * Theory Library/Multiset provides stable quicksort implementation of
  2144 sort_key.
  2145 
  2146 * Theory Library/Multiset: renamed empty_idemp ~> empty_neutral.
  2147 INCOMPATIBILITY.
  2148 
  2149 * Session Multivariate_Analysis: introduced a type class for euclidean
  2150 space.  Most theorems are now stated in terms of euclidean spaces
  2151 instead of finite cartesian products.
  2152 
  2153   types
  2154     real ^ 'n ~>  'a::real_vector
  2155               ~>  'a::euclidean_space
  2156               ~>  'a::ordered_euclidean_space
  2157         (depends on your needs)
  2158 
  2159   constants
  2160      _ $ _        ~> _ $$ _
  2161      \<chi> x. _  ~> \<chi>\<chi> x. _
  2162      CARD('n)     ~> DIM('a)
  2163 
  2164 Also note that the indices are now natural numbers and not from some
  2165 finite type. Finite cartesian products of euclidean spaces, products
  2166 of euclidean spaces the real and complex numbers are instantiated to
  2167 be euclidean_spaces.  INCOMPATIBILITY.
  2168 
  2169 * Session Probability: introduced pextreal as positive extended real
  2170 numbers.  Use pextreal as value for measures.  Introduce the
  2171 Radon-Nikodym derivative, product spaces and Fubini's theorem for
  2172 arbitrary sigma finite measures.  Introduces Lebesgue measure based on
  2173 the integral in Multivariate Analysis.  INCOMPATIBILITY.
  2174 
  2175 * Session Imperative_HOL: revamped, corrected dozens of inadequacies.
  2176 INCOMPATIBILITY.
  2177 
  2178 * Session SPARK (with image HOL-SPARK) provides commands to load and
  2179 prove verification conditions generated by the SPARK Ada program
  2180 verifier.  See also src/HOL/SPARK and src/HOL/SPARK/Examples.
  2181 
  2182 
  2183 *** HOL-Algebra ***
  2184 
  2185 * Theorems for additive ring operations (locale abelian_monoid and
  2186 descendants) are generated by interpretation from their multiplicative
  2187 counterparts.  Names (in particular theorem names) have the mandatory
  2188 qualifier 'add'.  Previous theorem names are redeclared for
  2189 compatibility.
  2190 
  2191 * Structure "int_ring" is now an abbreviation (previously a
  2192 definition).  This fits more natural with advanced interpretations.
  2193 
  2194 
  2195 *** HOLCF ***
  2196 
  2197 * The domain package now runs in definitional mode by default: The
  2198 former command 'new_domain' is now called 'domain'.  To use the domain
  2199 package in its original axiomatic mode, use 'domain (unsafe)'.
  2200 INCOMPATIBILITY.
  2201 
  2202 * The new class "domain" is now the default sort.  Class "predomain"
  2203 is an unpointed version of "domain". Theories can be updated by
  2204 replacing sort annotations as shown below.  INCOMPATIBILITY.
  2205 
  2206   'a::type ~> 'a::countable
  2207   'a::cpo  ~> 'a::predomain
  2208   'a::pcpo ~> 'a::domain
  2209 
  2210 * The old type class "rep" has been superseded by class "domain".
  2211 Accordingly, users of the definitional package must remove any
  2212 "default_sort rep" declarations.  INCOMPATIBILITY.
  2213 
  2214 * The domain package (definitional mode) now supports unpointed
  2215 predomain argument types, as long as they are marked 'lazy'. (Strict
  2216 arguments must be in class "domain".) For example, the following
  2217 domain definition now works:
  2218 
  2219   domain natlist = nil | cons (lazy "nat discr") (lazy "natlist")
  2220 
  2221 * Theory HOLCF/Library/HOL_Cpo provides cpo and predomain class
  2222 instances for types from main HOL: bool, nat, int, char, 'a + 'b,
  2223 'a option, and 'a list.  Additionally, it configures fixrec and the
  2224 domain package to work with these types.  For example:
  2225 
  2226   fixrec isInl :: "('a + 'b) u -> tr"
  2227     where "isInl$(up$(Inl x)) = TT" | "isInl$(up$(Inr y)) = FF"
  2228 
  2229   domain V = VFun (lazy "V -> V") | VCon (lazy "nat") (lazy "V list")
  2230 
  2231 * The "(permissive)" option of fixrec has been replaced with a
  2232 per-equation "(unchecked)" option. See
  2233 src/HOL/HOLCF/Tutorial/Fixrec_ex.thy for examples. INCOMPATIBILITY.
  2234 
  2235 * The "bifinite" class no longer fixes a constant "approx"; the class
  2236 now just asserts that such a function exists.  INCOMPATIBILITY.
  2237 
  2238 * Former type "alg_defl" has been renamed to "defl".  HOLCF no longer
  2239 defines an embedding of type 'a defl into udom by default; instances
  2240 of "bifinite" and "domain" classes are available in
  2241 src/HOL/HOLCF/Library/Defl_Bifinite.thy.
  2242 
  2243 * The syntax "REP('a)" has been replaced with "DEFL('a)".
  2244 
  2245 * The predicate "directed" has been removed.  INCOMPATIBILITY.
  2246 
  2247 * The type class "finite_po" has been removed.  INCOMPATIBILITY.
  2248 
  2249 * The function "cprod_map" has been renamed to "prod_map".
  2250 INCOMPATIBILITY.
  2251 
  2252 * The monadic bind operator on each powerdomain has new binder syntax
  2253 similar to sets, e.g. "\<Union>\<sharp>x\<in>xs. t" represents
  2254 "upper_bind\<cdot>xs\<cdot>(\<Lambda> x. t)".
  2255 
  2256 * The infix syntax for binary union on each powerdomain has changed
  2257 from e.g. "+\<sharp>" to "\<union>\<sharp>", for consistency with set
  2258 syntax.  INCOMPATIBILITY.
  2259 
  2260 * The constant "UU" has been renamed to "bottom".  The syntax "UU" is
  2261 still supported as an input translation.
  2262 
  2263 * Renamed some theorems (the original names are also still available).
  2264 
  2265   expand_fun_below   ~> fun_below_iff
  2266   below_fun_ext      ~> fun_belowI
  2267   expand_cfun_eq     ~> cfun_eq_iff
  2268   ext_cfun           ~> cfun_eqI
  2269   expand_cfun_below  ~> cfun_below_iff
  2270   below_cfun_ext     ~> cfun_belowI
  2271   cont2cont_Rep_CFun ~> cont2cont_APP
  2272 
  2273 * The Abs and Rep functions for various types have changed names.
  2274 Related theorem names have also changed to match. INCOMPATIBILITY.
  2275 
  2276   Rep_CFun  ~> Rep_cfun
  2277   Abs_CFun  ~> Abs_cfun
  2278   Rep_Sprod ~> Rep_sprod
  2279   Abs_Sprod ~> Abs_sprod
  2280   Rep_Ssum  ~> Rep_ssum
  2281   Abs_Ssum  ~> Abs_ssum
  2282 
  2283 * Lemmas with names of the form *_defined_iff or *_strict_iff have
  2284 been renamed to *_bottom_iff.  INCOMPATIBILITY.
  2285 
  2286 * Various changes to bisimulation/coinduction with domain package:
  2287 
  2288   - Definitions of "bisim" constants no longer mention definedness.
  2289   - With mutual recursion, "bisim" predicate is now curried.
  2290   - With mutual recursion, each type gets a separate coind theorem.
  2291   - Variable names in bisim_def and coinduct rules have changed.
  2292 
  2293 INCOMPATIBILITY.
  2294 
  2295 * Case combinators generated by the domain package for type "foo" are
  2296 now named "foo_case" instead of "foo_when".  INCOMPATIBILITY.
  2297 
  2298 * Several theorems have been renamed to more accurately reflect the
  2299 names of constants and types involved.  INCOMPATIBILITY.
  2300 
  2301   thelub_const    ~> lub_const
  2302   lub_const       ~> is_lub_const
  2303   thelubI         ~> lub_eqI
  2304   is_lub_lub      ~> is_lubD2
  2305   lubI            ~> is_lub_lub
  2306   unique_lub      ~> is_lub_unique
  2307   is_ub_lub       ~> is_lub_rangeD1
  2308   lub_bin_chain   ~> is_lub_bin_chain
  2309   lub_fun         ~> is_lub_fun
  2310   thelub_fun      ~> lub_fun
  2311   thelub_cfun     ~> lub_cfun
  2312   thelub_Pair     ~> lub_Pair
  2313   lub_cprod       ~> is_lub_prod
  2314   thelub_cprod    ~> lub_prod
  2315   minimal_cprod   ~> minimal_prod
  2316   inst_cprod_pcpo ~> inst_prod_pcpo
  2317   UU_I            ~> bottomI
  2318   compact_UU      ~> compact_bottom
  2319   deflation_UU    ~> deflation_bottom
  2320   finite_deflation_UU ~> finite_deflation_bottom
  2321 
  2322 * Many legacy theorem names have been discontinued.  INCOMPATIBILITY.
  2323 
  2324   sq_ord_less_eq_trans ~> below_eq_trans
  2325   sq_ord_eq_less_trans ~> eq_below_trans
  2326   refl_less            ~> below_refl
  2327   trans_less           ~> below_trans
  2328   antisym_less         ~> below_antisym
  2329   antisym_less_inverse ~> po_eq_conv [THEN iffD1]
  2330   box_less             ~> box_below
  2331   rev_trans_less       ~> rev_below_trans
  2332   not_less2not_eq      ~> not_below2not_eq
  2333   less_UU_iff          ~> below_UU_iff
  2334   flat_less_iff        ~> flat_below_iff
  2335   adm_less             ~> adm_below
  2336   adm_not_less         ~> adm_not_below
  2337   adm_compact_not_less ~> adm_compact_not_below
  2338   less_fun_def         ~> below_fun_def
  2339   expand_fun_less      ~> fun_below_iff
  2340   less_fun_ext         ~> fun_belowI
  2341   less_discr_def       ~> below_discr_def
  2342   discr_less_eq        ~> discr_below_eq
  2343   less_unit_def        ~> below_unit_def
  2344   less_cprod_def       ~> below_prod_def
  2345   prod_lessI           ~> prod_belowI
  2346   Pair_less_iff        ~> Pair_below_iff
  2347   fst_less_iff         ~> fst_below_iff
  2348   snd_less_iff         ~> snd_below_iff
  2349   expand_cfun_less     ~> cfun_below_iff
  2350   less_cfun_ext        ~> cfun_belowI
  2351   injection_less       ~> injection_below
  2352   less_up_def          ~> below_up_def
  2353   not_Iup_less         ~> not_Iup_below
  2354   Iup_less             ~> Iup_below
  2355   up_less              ~> up_below
  2356   Def_inject_less_eq   ~> Def_below_Def
  2357   Def_less_is_eq       ~> Def_below_iff
  2358   spair_less_iff       ~> spair_below_iff
  2359   less_sprod           ~> below_sprod
  2360   spair_less           ~> spair_below
  2361   sfst_less_iff        ~> sfst_below_iff
  2362   ssnd_less_iff        ~> ssnd_below_iff
  2363   fix_least_less       ~> fix_least_below
  2364   dist_less_one        ~> dist_below_one
  2365   less_ONE             ~> below_ONE
  2366   ONE_less_iff         ~> ONE_below_iff
  2367   less_sinlD           ~> below_sinlD
  2368   less_sinrD           ~> below_sinrD
  2369 
  2370 
  2371 *** FOL and ZF ***
  2372 
  2373 * All constant names are now qualified internally and use proper
  2374 identifiers, e.g. "IFOL.eq" instead of "op =".  INCOMPATIBILITY.
  2375 
  2376 
  2377 *** ML ***
  2378 
  2379 * Antiquotation @{assert} inlines a function bool -> unit that raises
  2380 Fail if the argument is false.  Due to inlining the source position of
  2381 failed assertions is included in the error output.
  2382 
  2383 * Discontinued antiquotation @{theory_ref}, which is obsolete since ML
  2384 text is in practice always evaluated with a stable theory checkpoint.
  2385 Minor INCOMPATIBILITY, use (Theory.check_thy @{theory}) instead.
  2386 
  2387 * Antiquotation @{theory A} refers to theory A from the ancestry of
  2388 the current context, not any accidental theory loader state as before.
  2389 Potential INCOMPATIBILITY, subtle change in semantics.
  2390 
  2391 * Syntax.pretty_priority (default 0) configures the required priority
  2392 of pretty-printed output and thus affects insertion of parentheses.
  2393 
  2394 * Syntax.default_root (default "any") configures the inner syntax
  2395 category (nonterminal symbol) for parsing of terms.
  2396 
  2397 * Former exception Library.UnequalLengths now coincides with
  2398 ListPair.UnequalLengths.
  2399 
  2400 * Renamed structure MetaSimplifier to Raw_Simplifier.  Note that the
  2401 main functionality is provided by structure Simplifier.
  2402 
  2403 * Renamed raw "explode" function to "raw_explode" to emphasize its
  2404 meaning.  Note that internally to Isabelle, Symbol.explode is used in
  2405 almost all situations.
  2406 
  2407 * Discontinued obsolete function sys_error and exception SYS_ERROR.
  2408 See implementation manual for further details on exceptions in
  2409 Isabelle/ML.
  2410 
  2411 * Renamed setmp_noncritical to Unsynchronized.setmp to emphasize its
  2412 meaning.
  2413 
  2414 * Renamed structure PureThy to Pure_Thy and moved most of its
  2415 operations to structure Global_Theory, to emphasize that this is
  2416 rarely-used global-only stuff.
  2417 
  2418 * Discontinued Output.debug.  Minor INCOMPATIBILITY, use plain writeln
  2419 instead (or tracing for high-volume output).
  2420 
  2421 * Configuration option show_question_marks only affects regular pretty
  2422 printing of types and terms, not raw Term.string_of_vname.
  2423 
  2424 * ML_Context.thm and ML_Context.thms are no longer pervasive.  Rare
  2425 INCOMPATIBILITY, superseded by static antiquotations @{thm} and
  2426 @{thms} for most purposes.
  2427 
  2428 * ML structure Unsynchronized is never opened, not even in Isar
  2429 interaction mode as before.  Old Unsynchronized.set etc. have been
  2430 discontinued -- use plain := instead.  This should be *rare* anyway,
  2431 since modern tools always work via official context data, notably
  2432 configuration options.
  2433 
  2434 * Parallel and asynchronous execution requires special care concerning
  2435 interrupts.  Structure Exn provides some convenience functions that
  2436 avoid working directly with raw Interrupt.  User code must not absorb
  2437 interrupts -- intermediate handling (for cleanup etc.) needs to be
  2438 followed by re-raising of the original exception.  Another common
  2439 source of mistakes are "handle _" patterns, which make the meaning of
  2440 the program subject to physical effects of the environment.
  2441 
  2442 
  2443 
  2444 New in Isabelle2009-2 (June 2010)
  2445 ---------------------------------
  2446 
  2447 *** General ***
  2448 
  2449 * Authentic syntax for *all* logical entities (type classes, type
  2450 constructors, term constants): provides simple and robust
  2451 correspondence between formal entities and concrete syntax.  Within
  2452 the parse tree / AST representations, "constants" are decorated by
  2453 their category (class, type, const) and spelled out explicitly with
  2454 their full internal name.
  2455 
  2456 Substantial INCOMPATIBILITY concerning low-level syntax declarations
  2457 and translations (translation rules and translation functions in ML).
  2458 Some hints on upgrading:
  2459 
  2460   - Many existing uses of 'syntax' and 'translations' can be replaced
  2461     by more modern 'type_notation', 'notation' and 'abbreviation',
  2462     which are independent of this issue.
  2463 
  2464   - 'translations' require markup within the AST; the term syntax
  2465     provides the following special forms:
  2466 
  2467       CONST c   -- produces syntax version of constant c from context
  2468       XCONST c  -- literally c, checked as constant from context
  2469       c         -- literally c, if declared by 'syntax'
  2470 
  2471     Plain identifiers are treated as AST variables -- occasionally the
  2472     system indicates accidental variables via the error "rhs contains
  2473     extra variables".
  2474 
  2475     Type classes and type constructors are marked according to their
  2476     concrete syntax.  Some old translations rules need to be written
  2477     for the "type" category, using type constructor application
  2478     instead of pseudo-term application of the default category
  2479     "logic".
  2480 
  2481   - 'parse_translation' etc. in ML may use the following
  2482     antiquotations:
  2483 
  2484       @{class_syntax c}   -- type class c within parse tree / AST
  2485       @{term_syntax c}    -- type constructor c within parse tree / AST
  2486       @{const_syntax c}   -- ML version of "CONST c" above
  2487       @{syntax_const c}   -- literally c (checked wrt. 'syntax' declarations)
  2488 
  2489   - Literal types within 'typed_print_translations', i.e. those *not*
  2490     represented as pseudo-terms are represented verbatim.  Use @{class
  2491     c} or @{type_name c} here instead of the above syntax
  2492     antiquotations.
  2493 
  2494 Note that old non-authentic syntax was based on unqualified base
  2495 names, so all of the above "constant" names would coincide.  Recall
  2496 that 'print_syntax' and ML_command "set Syntax.trace_ast" help to
  2497 diagnose syntax problems.
  2498 
  2499 * Type constructors admit general mixfix syntax, not just infix.
  2500 
  2501 * Concrete syntax may be attached to local entities without a proof
  2502 body, too.  This works via regular mixfix annotations for 'fix',
  2503 'def', 'obtain' etc. or via the explicit 'write' command, which is
  2504 similar to the 'notation' command in theory specifications.
  2505 
  2506 * Discontinued unnamed infix syntax (legacy feature for many years) --
  2507 need to specify constant name and syntax separately.  Internal ML
  2508 datatype constructors have been renamed from InfixName to Infix etc.
  2509 Minor INCOMPATIBILITY.
  2510 
  2511 * Schematic theorem statements need to be explicitly markup as such,
  2512 via commands 'schematic_lemma', 'schematic_theorem',
  2513 'schematic_corollary'.  Thus the relevance of the proof is made
  2514 syntactically clear, which impacts performance in a parallel or
  2515 asynchronous interactive environment.  Minor INCOMPATIBILITY.
  2516 
  2517 * Use of cumulative prems via "!" in some proof methods has been
  2518 discontinued (old legacy feature).
  2519 
  2520 * References 'trace_simp' and 'debug_simp' have been replaced by
  2521 configuration options stored in the context. Enabling tracing (the
  2522 case of debugging is similar) in proofs works via
  2523 
  2524   using [[trace_simp = true]]
  2525 
  2526 Tracing is then active for all invocations of the simplifier in
  2527 subsequent goal refinement steps. Tracing may also still be enabled or
  2528 disabled via the ProofGeneral settings menu.
  2529 
  2530 * Separate commands 'hide_class', 'hide_type', 'hide_const',
  2531 'hide_fact' replace the former 'hide' KIND command.  Minor
  2532 INCOMPATIBILITY.
  2533 
  2534 * Improved parallelism of proof term normalization: usedir -p2 -q0 is
  2535 more efficient than combinations with -q1 or -q2.
  2536 
  2537 
  2538 *** Pure ***
  2539 
  2540 * Proofterms record type-class reasoning explicitly, using the
  2541 "unconstrain" operation internally.  This eliminates all sort
  2542 constraints from a theorem and proof, introducing explicit
  2543 OFCLASS-premises.  On the proof term level, this operation is
  2544 automatically applied at theorem boundaries, such that closed proofs
  2545 are always free of sort constraints.  INCOMPATIBILITY for tools that
  2546 inspect proof terms.
  2547 
  2548 * Local theory specifications may depend on extra type variables that
  2549 are not present in the result type -- arguments TYPE('a) :: 'a itself
  2550 are added internally.  For example:
  2551 
  2552   definition unitary :: bool where "unitary = (ALL (x::'a) y. x = y)"
  2553 
  2554 * Predicates of locales introduced by classes carry a mandatory
  2555 "class" prefix.  INCOMPATIBILITY.
  2556 
  2557 * Vacuous class specifications observe default sort.  INCOMPATIBILITY.
  2558 
  2559 * Old 'axclass' command has been discontinued.  INCOMPATIBILITY, use
  2560 'class' instead.
  2561 
  2562 * Command 'code_reflect' allows to incorporate generated ML code into
  2563 runtime environment; replaces immature code_datatype antiquotation.
  2564 INCOMPATIBILITY.
  2565 
  2566 * Code generator: simple concept for abstract datatypes obeying
  2567 invariants.
  2568 
  2569 * Code generator: details of internal data cache have no impact on the
  2570 user space functionality any longer.
  2571 
  2572 * Methods "unfold_locales" and "intro_locales" ignore non-locale
  2573 subgoals.  This is more appropriate for interpretations with 'where'.
  2574 INCOMPATIBILITY.
  2575 
  2576 * Command 'example_proof' opens an empty proof body.  This allows to
  2577 experiment with Isar, without producing any persistent result.
  2578 
  2579 * Commands 'type_notation' and 'no_type_notation' declare type syntax
  2580 within a local theory context, with explicit checking of the
  2581 constructors involved (in contrast to the raw 'syntax' versions).
  2582 
  2583 * Commands 'types' and 'typedecl' now work within a local theory
  2584 context -- without introducing dependencies on parameters or
  2585 assumptions, which is not possible in Isabelle/Pure.
  2586 
  2587 * Command 'defaultsort' has been renamed to 'default_sort', it works
  2588 within a local theory context.  Minor INCOMPATIBILITY.
  2589 
  2590 
  2591 *** HOL ***
  2592 
  2593 * Command 'typedef' now works within a local theory context -- without
  2594 introducing dependencies on parameters or assumptions, which is not
  2595 possible in Isabelle/Pure/HOL.  Note that the logical environment may
  2596 contain multiple interpretations of local typedefs (with different
  2597 non-emptiness proofs), even in a global theory context.
  2598 
  2599 * New package for quotient types.  Commands 'quotient_type' and
  2600 'quotient_definition' may be used for defining types and constants by
  2601 quotient constructions.  An example is the type of integers created by
  2602 quotienting pairs of natural numbers:
  2603 
  2604   fun
  2605     intrel :: "(nat * nat) => (nat * nat) => bool"
  2606   where
  2607     "intrel (x, y) (u, v) = (x + v = u + y)"
  2608 
  2609   quotient_type int = "nat * nat" / intrel
  2610     by (auto simp add: equivp_def expand_fun_eq)
  2611 
  2612   quotient_definition
  2613     "0::int" is "(0::nat, 0::nat)"
  2614 
  2615 The method "lifting" can be used to lift of theorems from the
  2616 underlying "raw" type to the quotient type.  The example
  2617 src/HOL/Quotient_Examples/FSet.thy includes such a quotient
  2618 construction and provides a reasoning infrastructure for finite sets.
  2619 
  2620 * Renamed Library/Quotient.thy to Library/Quotient_Type.thy to avoid
  2621 clash with new theory Quotient in Main HOL.
  2622 
  2623 * Moved the SMT binding into the main HOL session, eliminating
  2624 separate HOL-SMT session.
  2625 
  2626 * List membership infix mem operation is only an input abbreviation.
  2627 INCOMPATIBILITY.
  2628 
  2629 * Theory Library/Word.thy has been removed.  Use library Word/Word.thy
  2630 for future developements; former Library/Word.thy is still present in
  2631 the AFP entry RSAPPS.
  2632 
  2633 * Theorem Int.int_induct renamed to Int.int_of_nat_induct and is no
  2634 longer shadowed.  INCOMPATIBILITY.
  2635 
  2636 * Dropped theorem duplicate comp_arith; use semiring_norm instead.
  2637 INCOMPATIBILITY.
  2638 
  2639 * Dropped theorem RealPow.real_sq_order; use power2_le_imp_le instead.
  2640 INCOMPATIBILITY.
  2641 
  2642 * Dropped normalizing_semiring etc; use the facts in semiring classes
  2643 instead.  INCOMPATIBILITY.
  2644 
  2645 * Dropped several real-specific versions of lemmas about floor and
  2646 ceiling; use the generic lemmas from theory "Archimedean_Field"
  2647 instead.  INCOMPATIBILITY.
  2648 
  2649   floor_number_of_eq         ~> floor_number_of
  2650   le_floor_eq_number_of      ~> number_of_le_floor
  2651   le_floor_eq_zero           ~> zero_le_floor
  2652   le_floor_eq_one            ~> one_le_floor
  2653   floor_less_eq_number_of    ~> floor_less_number_of
  2654   floor_less_eq_zero         ~> floor_less_zero
  2655   floor_less_eq_one          ~> floor_less_one
  2656   less_floor_eq_number_of    ~> number_of_less_floor
  2657   less_floor_eq_zero         ~> zero_less_floor
  2658   less_floor_eq_one          ~> one_less_floor
  2659   floor_le_eq_number_of      ~> floor_le_number_of
  2660   floor_le_eq_zero           ~> floor_le_zero
  2661   floor_le_eq_one            ~> floor_le_one
  2662   floor_subtract_number_of   ~> floor_diff_number_of
  2663   floor_subtract_one         ~> floor_diff_one
  2664   ceiling_number_of_eq       ~> ceiling_number_of
  2665   ceiling_le_eq_number_of    ~> ceiling_le_number_of
  2666   ceiling_le_zero_eq         ~> ceiling_le_zero
  2667   ceiling_le_eq_one          ~> ceiling_le_one
  2668   less_ceiling_eq_number_of  ~> number_of_less_ceiling
  2669   less_ceiling_eq_zero       ~> zero_less_ceiling
  2670   less_ceiling_eq_one        ~> one_less_ceiling
  2671   ceiling_less_eq_number_of  ~> ceiling_less_number_of
  2672   ceiling_less_eq_zero       ~> ceiling_less_zero
  2673   ceiling_less_eq_one        ~> ceiling_less_one
  2674   le_ceiling_eq_number_of    ~> number_of_le_ceiling
  2675   le_ceiling_eq_zero         ~> zero_le_ceiling
  2676   le_ceiling_eq_one          ~> one_le_ceiling
  2677   ceiling_subtract_number_of ~> ceiling_diff_number_of
  2678   ceiling_subtract_one       ~> ceiling_diff_one
  2679 
  2680 * Theory "Finite_Set": various folding_XXX locales facilitate the
  2681 application of the various fold combinators on finite sets.
  2682 
  2683 * Library theory "RBT" renamed to "RBT_Impl"; new library theory "RBT"
  2684 provides abstract red-black tree type which is backed by "RBT_Impl" as
  2685 implementation.  INCOMPATIBILTY.
  2686 
  2687 * Theory Library/Coinductive_List has been removed -- superseded by
  2688 AFP/thys/Coinductive.
  2689 
  2690 * Theory PReal, including the type "preal" and related operations, has
  2691 been removed.  INCOMPATIBILITY.
  2692 
  2693 * Real: new development using Cauchy Sequences.
  2694 
  2695 * Split off theory "Big_Operators" containing setsum, setprod,
  2696 Inf_fin, Sup_fin, Min, Max from theory Finite_Set.  INCOMPATIBILITY.
  2697 
  2698 * Theory "Rational" renamed to "Rat", for consistency with "Nat",
  2699 "Int" etc.  INCOMPATIBILITY.
  2700 
  2701 * Constant Rat.normalize needs to be qualified.  INCOMPATIBILITY.
  2702 
  2703 * New set of rules "ac_simps" provides combined assoc / commute
  2704 rewrites for all interpretations of the appropriate generic locales.
  2705 
  2706 * Renamed theory "OrderedGroup" to "Groups" and split theory
  2707 "Ring_and_Field" into theories "Rings" and "Fields"; for more
  2708 appropriate and more consistent names suitable for name prefixes
  2709 within the HOL theories.  INCOMPATIBILITY.
  2710 
  2711 * Some generic constants have been put to appropriate theories:
  2712   - less_eq, less: Orderings
  2713   - zero, one, plus, minus, uminus, times, abs, sgn: Groups
  2714   - inverse, divide: Rings
  2715 INCOMPATIBILITY.
  2716 
  2717 * More consistent naming of type classes involving orderings (and
  2718 lattices):
  2719 
  2720     lower_semilattice                   ~> semilattice_inf
  2721     upper_semilattice                   ~> semilattice_sup
  2722 
  2723     dense_linear_order                  ~> dense_linorder
  2724 
  2725     pordered_ab_group_add               ~> ordered_ab_group_add
  2726     pordered_ab_group_add_abs           ~> ordered_ab_group_add_abs
  2727     pordered_ab_semigroup_add           ~> ordered_ab_semigroup_add
  2728     pordered_ab_semigroup_add_imp_le    ~> ordered_ab_semigroup_add_imp_le
  2729     pordered_cancel_ab_semigroup_add    ~> ordered_cancel_ab_semigroup_add
  2730     pordered_cancel_comm_semiring       ~> ordered_cancel_comm_semiring
  2731     pordered_cancel_semiring            ~> ordered_cancel_semiring
  2732     pordered_comm_monoid_add            ~> ordered_comm_monoid_add
  2733     pordered_comm_ring                  ~> ordered_comm_ring
  2734     pordered_comm_semiring              ~> ordered_comm_semiring
  2735     pordered_ring                       ~> ordered_ring
  2736     pordered_ring_abs                   ~> ordered_ring_abs
  2737     pordered_semiring                   ~> ordered_semiring
  2738 
  2739     ordered_ab_group_add                ~> linordered_ab_group_add
  2740     ordered_ab_semigroup_add            ~> linordered_ab_semigroup_add
  2741     ordered_cancel_ab_semigroup_add     ~> linordered_cancel_ab_semigroup_add
  2742     ordered_comm_semiring_strict        ~> linordered_comm_semiring_strict
  2743     ordered_field                       ~> linordered_field
  2744     ordered_field_no_lb                 ~> linordered_field_no_lb
  2745     ordered_field_no_ub                 ~> linordered_field_no_ub
  2746     ordered_field_dense_linear_order    ~> dense_linordered_field
  2747     ordered_idom                        ~> linordered_idom
  2748     ordered_ring                        ~> linordered_ring
  2749     ordered_ring_le_cancel_factor       ~> linordered_ring_le_cancel_factor
  2750     ordered_ring_less_cancel_factor     ~> linordered_ring_less_cancel_factor
  2751     ordered_ring_strict                 ~> linordered_ring_strict
  2752     ordered_semidom                     ~> linordered_semidom
  2753     ordered_semiring                    ~> linordered_semiring
  2754     ordered_semiring_1                  ~> linordered_semiring_1
  2755     ordered_semiring_1_strict           ~> linordered_semiring_1_strict
  2756     ordered_semiring_strict             ~> linordered_semiring_strict
  2757 
  2758   The following slightly odd type classes have been moved to a
  2759   separate theory Library/Lattice_Algebras:
  2760 
  2761     lordered_ab_group_add               ~> lattice_ab_group_add
  2762     lordered_ab_group_add_abs           ~> lattice_ab_group_add_abs
  2763     lordered_ab_group_add_meet          ~> semilattice_inf_ab_group_add
  2764     lordered_ab_group_add_join          ~> semilattice_sup_ab_group_add
  2765     lordered_ring                       ~> lattice_ring
  2766 
  2767 INCOMPATIBILITY.
  2768 
  2769 * Refined field classes:
  2770   - classes division_ring_inverse_zero, field_inverse_zero,
  2771     linordered_field_inverse_zero include rule inverse 0 = 0 --
  2772     subsumes former division_by_zero class;
  2773   - numerous lemmas have been ported from field to division_ring.
  2774 INCOMPATIBILITY.
  2775 
  2776 * Refined algebra theorem collections:
  2777   - dropped theorem group group_simps, use algebra_simps instead;
  2778   - dropped theorem group ring_simps, use field_simps instead;
  2779   - proper theorem collection field_simps subsumes former theorem
  2780     groups field_eq_simps and field_simps;
  2781   - dropped lemma eq_minus_self_iff which is a duplicate for
  2782     equal_neg_zero.
  2783 INCOMPATIBILITY.
  2784 
  2785 * Theory Finite_Set and List: some lemmas have been generalized from
  2786 sets to lattices:
  2787 
  2788   fun_left_comm_idem_inter      ~> fun_left_comm_idem_inf
  2789   fun_left_comm_idem_union      ~> fun_left_comm_idem_sup
  2790   inter_Inter_fold_inter        ~> inf_Inf_fold_inf
  2791   union_Union_fold_union        ~> sup_Sup_fold_sup
  2792   Inter_fold_inter              ~> Inf_fold_inf
  2793   Union_fold_union              ~> Sup_fold_sup
  2794   inter_INTER_fold_inter        ~> inf_INFI_fold_inf
  2795   union_UNION_fold_union        ~> sup_SUPR_fold_sup
  2796   INTER_fold_inter              ~> INFI_fold_inf
  2797   UNION_fold_union              ~> SUPR_fold_sup
  2798 
  2799 * Theory "Complete_Lattice": lemmas top_def and bot_def have been
  2800 replaced by the more convenient lemmas Inf_empty and Sup_empty.
  2801 Dropped lemmas Inf_insert_simp and Sup_insert_simp, which are subsumed
  2802 by Inf_insert and Sup_insert.  Lemmas Inf_UNIV and Sup_UNIV replace
  2803 former Inf_Univ and Sup_Univ.  Lemmas inf_top_right and sup_bot_right
  2804 subsume inf_top and sup_bot respectively.  INCOMPATIBILITY.
  2805 
  2806 * Reorganized theory Multiset: swapped notation of pointwise and
  2807 multiset order:
  2808 
  2809   - pointwise ordering is instance of class order with standard syntax
  2810     <= and <;
  2811   - multiset ordering has syntax <=# and <#; partial order properties
  2812     are provided by means of interpretation with prefix
  2813     multiset_order;
  2814   - less duplication, less historical organization of sections,
  2815     conversion from associations lists to multisets, rudimentary code
  2816     generation;
  2817   - use insert_DiffM2 [symmetric] instead of elem_imp_eq_diff_union,
  2818     if needed.
  2819 
  2820 Renamed:
  2821 
  2822   multiset_eq_conv_count_eq  ~>  multiset_ext_iff
  2823   multi_count_ext  ~>  multiset_ext
  2824   diff_union_inverse2  ~>  diff_union_cancelR
  2825 
  2826 INCOMPATIBILITY.
  2827 
  2828 * Theory Permutation: replaced local "remove" by List.remove1.
  2829 
  2830 * Code generation: ML and OCaml code is decorated with signatures.
  2831 
  2832 * Theory List: added transpose.
  2833 
  2834 * Library/Nat_Bijection.thy is a collection of bijective functions
  2835 between nat and other types, which supersedes the older libraries
  2836 Library/Nat_Int_Bij.thy and HOLCF/NatIso.thy.  INCOMPATIBILITY.
  2837 
  2838   Constants:
  2839   Nat_Int_Bij.nat2_to_nat         ~> prod_encode
  2840   Nat_Int_Bij.nat_to_nat2         ~> prod_decode
  2841   Nat_Int_Bij.int_to_nat_bij      ~> int_encode
  2842   Nat_Int_Bij.nat_to_int_bij      ~> int_decode
  2843   Countable.pair_encode           ~> prod_encode
  2844   NatIso.prod2nat                 ~> prod_encode
  2845   NatIso.nat2prod                 ~> prod_decode
  2846   NatIso.sum2nat                  ~> sum_encode
  2847   NatIso.nat2sum                  ~> sum_decode
  2848   NatIso.list2nat                 ~> list_encode
  2849   NatIso.nat2list                 ~> list_decode
  2850   NatIso.set2nat                  ~> set_encode
  2851   NatIso.nat2set                  ~> set_decode
  2852 
  2853   Lemmas:
  2854   Nat_Int_Bij.bij_nat_to_int_bij  ~> bij_int_decode
  2855   Nat_Int_Bij.nat2_to_nat_inj     ~> inj_prod_encode
  2856   Nat_Int_Bij.nat2_to_nat_surj    ~> surj_prod_encode
  2857   Nat_Int_Bij.nat_to_nat2_inj     ~> inj_prod_decode
  2858   Nat_Int_Bij.nat_to_nat2_surj    ~> surj_prod_decode
  2859   Nat_Int_Bij.i2n_n2i_id          ~> int_encode_inverse
  2860   Nat_Int_Bij.n2i_i2n_id          ~> int_decode_inverse
  2861   Nat_Int_Bij.surj_nat_to_int_bij ~> surj_int_encode
  2862   Nat_Int_Bij.surj_int_to_nat_bij ~> surj_int_decode
  2863   Nat_Int_Bij.inj_nat_to_int_bij  ~> inj_int_encode
  2864   Nat_Int_Bij.inj_int_to_nat_bij  ~> inj_int_decode
  2865   Nat_Int_Bij.bij_nat_to_int_bij  ~> bij_int_encode
  2866   Nat_Int_Bij.bij_int_to_nat_bij  ~> bij_int_decode
  2867 
  2868 * Sledgehammer:
  2869   - Renamed ATP commands:
  2870     atp_info     ~> sledgehammer running_atps
  2871     atp_kill     ~> sledgehammer kill_atps
  2872     atp_messages ~> sledgehammer messages
  2873     atp_minimize ~> sledgehammer minimize
  2874     print_atps   ~> sledgehammer available_atps
  2875     INCOMPATIBILITY.
  2876   - Added user's manual ("isabelle doc sledgehammer").
  2877   - Added option syntax and "sledgehammer_params" to customize
  2878     Sledgehammer's behavior.  See the manual for details.
  2879   - Modified the Isar proof reconstruction code so that it produces
  2880     direct proofs rather than proofs by contradiction.  (This feature
  2881     is still experimental.)
  2882   - Made Isar proof reconstruction work for SPASS, remote ATPs, and in
  2883     full-typed mode.
  2884   - Added support for TPTP syntax for SPASS via the "spass_tptp" ATP.
  2885 
  2886 * Nitpick:
  2887   - Added and implemented "binary_ints" and "bits" options.
  2888   - Added "std" option and implemented support for nonstandard models.
  2889   - Added and implemented "finitize" option to improve the precision
  2890     of infinite datatypes based on a monotonicity analysis.
  2891   - Added support for quotient types.
  2892   - Added support for "specification" and "ax_specification"
  2893     constructs.
  2894   - Added support for local definitions (for "function" and
  2895     "termination" proofs).
  2896   - Added support for term postprocessors.
  2897   - Optimized "Multiset.multiset" and "FinFun.finfun".
  2898   - Improved efficiency of "destroy_constrs" optimization.
  2899   - Fixed soundness bugs related to "destroy_constrs" optimization and
  2900     record getters.
  2901   - Fixed soundness bug related to higher-order constructors.
  2902   - Fixed soundness bug when "full_descrs" is enabled.
  2903   - Improved precision of set constructs.
  2904   - Added "atoms" option.
  2905   - Added cache to speed up repeated Kodkod invocations on the same
  2906     problems.
  2907   - Renamed "MiniSatJNI", "zChaffJNI", "BerkMinAlloy", and
  2908     "SAT4JLight" to "MiniSat_JNI", "zChaff_JNI", "BerkMin_Alloy", and
  2909     "SAT4J_Light".  INCOMPATIBILITY.
  2910   - Removed "skolemize", "uncurry", "sym_break", "flatten_prop",
  2911     "sharing_depth", and "show_skolems" options.  INCOMPATIBILITY.
  2912   - Removed "nitpick_intro" attribute.  INCOMPATIBILITY.
  2913 
  2914 * Method "induct" now takes instantiations of the form t, where t is not
  2915   a variable, as a shorthand for "x == t", where x is a fresh variable.
  2916   If this is not intended, t has to be enclosed in parentheses.
  2917   By default, the equalities generated by definitional instantiations
  2918   are pre-simplified, which may cause parameters of inductive cases
  2919   to disappear, or may even delete some of the inductive cases.
  2920   Use "induct (no_simp)" instead of "induct" to restore the old
  2921   behaviour. The (no_simp) option is also understood by the "cases"
  2922   and "nominal_induct" methods, which now perform pre-simplification, too.
  2923   INCOMPATIBILITY.
  2924 
  2925 
  2926 *** HOLCF ***
  2927 
  2928 * Variable names in lemmas generated by the domain package have
  2929 changed; the naming scheme is now consistent with the HOL datatype
  2930 package.  Some proof scripts may be affected, INCOMPATIBILITY.
  2931 
  2932 * The domain package no longer defines the function "foo_copy" for
  2933 recursive domain "foo".  The reach lemma is now stated directly in
  2934 terms of "foo_take".  Lemmas and proofs that mention "foo_copy" must
  2935 be reformulated in terms of "foo_take", INCOMPATIBILITY.
  2936 
  2937 * Most definedness lemmas generated by the domain package (previously
  2938 of the form "x ~= UU ==> foo$x ~= UU") now have an if-and-only-if form
  2939 like "foo$x = UU <-> x = UU", which works better as a simp rule.
  2940 Proofs that used definedness lemmas as intro rules may break,
  2941 potential INCOMPATIBILITY.
  2942 
  2943 * Induction and casedist rules generated by the domain package now
  2944 declare proper case_names (one called "bottom", and one named for each
  2945 constructor).  INCOMPATIBILITY.
  2946 
  2947 * For mutually-recursive domains, separate "reach" and "take_lemma"
  2948 rules are generated for each domain, INCOMPATIBILITY.
  2949 
  2950   foo_bar.reach       ~> foo.reach  bar.reach
  2951   foo_bar.take_lemmas ~> foo.take_lemma  bar.take_lemma
  2952 
  2953 * Some lemmas generated by the domain package have been renamed for
  2954 consistency with the datatype package, INCOMPATIBILITY.
  2955 
  2956   foo.ind        ~> foo.induct
  2957   foo.finite_ind ~> foo.finite_induct
  2958   foo.coind      ~> foo.coinduct
  2959   foo.casedist   ~> foo.exhaust
  2960   foo.exhaust    ~> foo.nchotomy
  2961 
  2962 * For consistency with other definition packages, the fixrec package
  2963 now generates qualified theorem names, INCOMPATIBILITY.
  2964 
  2965   foo_simps  ~> foo.simps
  2966   foo_unfold ~> foo.unfold
  2967   foo_induct ~> foo.induct
  2968 
  2969 * The "fixrec_simp" attribute has been removed.  The "fixrec_simp"
  2970 method and internal fixrec proofs now use the default simpset instead.
  2971 INCOMPATIBILITY.
  2972 
  2973 * The "contlub" predicate has been removed.  Proof scripts should use
  2974 lemma contI2 in place of monocontlub2cont, INCOMPATIBILITY.
  2975 
  2976 * The "admw" predicate has been removed, INCOMPATIBILITY.
  2977 
  2978 * The constants cpair, cfst, and csnd have been removed in favor of
  2979 Pair, fst, and snd from Isabelle/HOL, INCOMPATIBILITY.
  2980 
  2981 
  2982 *** ML ***
  2983 
  2984 * Antiquotations for basic formal entities:
  2985 
  2986     @{class NAME}         -- type class
  2987     @{class_syntax NAME}  -- syntax representation of the above
  2988 
  2989     @{type_name NAME}     -- logical type
  2990     @{type_abbrev NAME}   -- type abbreviation
  2991     @{nonterminal NAME}   -- type of concrete syntactic category
  2992     @{type_syntax NAME}   -- syntax representation of any of the above
  2993 
  2994     @{const_name NAME}    -- logical constant (INCOMPATIBILITY)
  2995     @{const_abbrev NAME}  -- abbreviated constant
  2996     @{const_syntax NAME}  -- syntax representation of any of the above
  2997 
  2998 * Antiquotation @{syntax_const NAME} ensures that NAME refers to a raw
  2999 syntax constant (cf. 'syntax' command).
  3000 
  3001 * Antiquotation @{make_string} inlines a function to print arbitrary
  3002 values similar to the ML toplevel.  The result is compiler dependent
  3003 and may fall back on "?" in certain situations.
  3004 
  3005 * Diagnostic commands 'ML_val' and 'ML_command' may refer to
  3006 antiquotations @{Isar.state} and @{Isar.goal}.  This replaces impure
  3007 Isar.state() and Isar.goal(), which belong to the old TTY loop and do
  3008 not work with the asynchronous Isar document model.
  3009 
  3010 * Configuration options now admit dynamic default values, depending on
  3011 the context or even global references.
  3012 
  3013 * SHA1.digest digests strings according to SHA-1 (see RFC 3174).  It
  3014 uses an efficient external library if available (for Poly/ML).
  3015 
  3016 * Renamed some important ML structures, while keeping the old names
  3017 for some time as aliases within the structure Legacy:
  3018 
  3019   OuterKeyword  ~>  Keyword
  3020   OuterLex      ~>  Token
  3021   OuterParse    ~>  Parse
  3022   OuterSyntax   ~>  Outer_Syntax
  3023   PrintMode     ~>  Print_Mode
  3024   SpecParse     ~>  Parse_Spec
  3025   ThyInfo       ~>  Thy_Info
  3026   ThyLoad       ~>  Thy_Load
  3027   ThyOutput     ~>  Thy_Output
  3028   TypeInfer     ~>  Type_Infer
  3029 
  3030 Note that "open Legacy" simplifies porting of sources, but forgetting
  3031 to remove it again will complicate porting again in the future.
  3032 
  3033 * Most operations that refer to a global context are named
  3034 accordingly, e.g. Simplifier.global_context or
  3035 ProofContext.init_global.  There are some situations where a global
  3036 context actually works, but under normal circumstances one needs to
  3037 pass the proper local context through the code!
  3038 
  3039 * Discontinued old TheoryDataFun with its copy/init operation -- data
  3040 needs to be pure.  Functor Theory_Data_PP retains the traditional
  3041 Pretty.pp argument to merge, which is absent in the standard
  3042 Theory_Data version.
  3043 
  3044 * Sorts.certify_sort and derived "cert" operations for types and terms
  3045 no longer minimize sorts.  Thus certification at the boundary of the
  3046 inference kernel becomes invariant under addition of class relations,
  3047 which is an important monotonicity principle.  Sorts are now minimized
  3048 in the syntax layer only, at the boundary between the end-user and the
  3049 system.  Subtle INCOMPATIBILITY, may have to use Sign.minimize_sort
  3050 explicitly in rare situations.
  3051 
  3052 * Renamed old-style Drule.standard to Drule.export_without_context, to
  3053 emphasize that this is in no way a standard operation.
  3054 INCOMPATIBILITY.
  3055 
  3056 * Subgoal.FOCUS (and variants): resulting goal state is normalized as
  3057 usual for resolution.  Rare INCOMPATIBILITY.
  3058 
  3059 * Renamed varify/unvarify operations to varify_global/unvarify_global
  3060 to emphasize that these only work in a global situation (which is
  3061 quite rare).
  3062 
  3063 * Curried take and drop in library.ML; negative length is interpreted
  3064 as infinity (as in chop).  Subtle INCOMPATIBILITY.
  3065 
  3066 * Proof terms: type substitutions on proof constants now use canonical
  3067 order of type variables.  INCOMPATIBILITY for tools working with proof
  3068 terms.
  3069 
  3070 * Raw axioms/defs may no longer carry sort constraints, and raw defs
  3071 may no longer carry premises.  User-level specifications are
  3072 transformed accordingly by Thm.add_axiom/add_def.
  3073 
  3074 
  3075 *** System ***
  3076 
  3077 * Discontinued special HOL_USEDIR_OPTIONS for the main HOL image;
  3078 ISABELLE_USEDIR_OPTIONS applies uniformly to all sessions.  Note that
  3079 proof terms are enabled unconditionally in the new HOL-Proofs image.
  3080 
  3081 * Discontinued old ISABELLE and ISATOOL environment settings (legacy
  3082 feature since Isabelle2009).  Use ISABELLE_PROCESS and ISABELLE_TOOL,
  3083 respectively.
  3084 
  3085 * Old lib/scripts/polyml-platform is superseded by the
  3086 ISABELLE_PLATFORM setting variable, which defaults to the 32 bit
  3087 variant, even on a 64 bit machine.  The following example setting
  3088 prefers 64 bit if available:
  3089 
  3090   ML_PLATFORM="${ISABELLE_PLATFORM64:-$ISABELLE_PLATFORM}"
  3091 
  3092 * The preliminary Isabelle/jEdit application demonstrates the emerging
  3093 Isabelle/Scala layer for advanced prover interaction and integration.
  3094 See src/Tools/jEdit or "isabelle jedit" provided by the properly built
  3095 component.
  3096 
  3097 * "IsabelleText" is a Unicode font derived from Bitstream Vera Mono
  3098 and Bluesky TeX fonts.  It provides the usual Isabelle symbols,
  3099 similar to the default assignment of the document preparation system
  3100 (cf. isabellesym.sty).  The Isabelle/Scala class Isabelle_System
  3101 provides some operations for direct access to the font without asking
  3102 the user for manual installation.
  3103 
  3104 
  3105 
  3106 New in Isabelle2009-1 (December 2009)
  3107 -------------------------------------
  3108 
  3109 *** General ***
  3110 
  3111 * Discontinued old form of "escaped symbols" such as \\<forall>.  Only
  3112 one backslash should be used, even in ML sources.
  3113 
  3114 
  3115 *** Pure ***
  3116 
  3117 * Locale interpretation propagates mixins along the locale hierarchy.
  3118 The currently only available mixins are the equations used to map
  3119 local definitions to terms of the target domain of an interpretation.
  3120 
  3121 * Reactivated diagnostic command 'print_interps'.  Use "print_interps
  3122 loc" to print all interpretations of locale "loc" in the theory.
  3123 Interpretations in proofs are not shown.
  3124 
  3125 * Thoroughly revised locales tutorial.  New section on conditional
  3126 interpretation.
  3127 
  3128 * On instantiation of classes, remaining undefined class parameters
  3129 are formally declared.  INCOMPATIBILITY.
  3130 
  3131 
  3132 *** Document preparation ***
  3133 
  3134 * New generalized style concept for printing terms: @{foo (style) ...}
  3135 instead of @{foo_style style ...}  (old form is still retained for
  3136 backward compatibility).  Styles can be also applied for
  3137 antiquotations prop, term_type and typeof.
  3138 
  3139 
  3140 *** HOL ***
  3141 
  3142 * New proof method "smt" for a combination of first-order logic with
  3143 equality, linear and nonlinear (natural/integer/real) arithmetic, and
  3144 fixed-size bitvectors; there is also basic support for higher-order
  3145 features (esp. lambda abstractions).  It is an incomplete decision
  3146 procedure based on external SMT solvers using the oracle mechanism;
  3147 for the SMT solver Z3, this method is proof-producing.  Certificates
  3148 are provided to avoid calling the external solvers solely for
  3149 re-checking proofs.  Due to a remote SMT service there is no need for
  3150 installing SMT solvers locally.  See src/HOL/SMT.
  3151 
  3152 * New commands to load and prove verification conditions generated by
  3153 the Boogie program verifier or derived systems (e.g. the Verifying C
  3154 Compiler (VCC) or Spec#).  See src/HOL/Boogie.
  3155 
  3156 * New counterexample generator tool 'nitpick' based on the Kodkod
  3157 relational model finder.  See src/HOL/Tools/Nitpick and
  3158 src/HOL/Nitpick_Examples.
  3159 
  3160 * New commands 'code_pred' and 'values' to invoke the predicate
  3161 compiler and to enumerate values of inductive predicates.
  3162 
  3163 * A tabled implementation of the reflexive transitive closure.
  3164 
  3165 * New implementation of quickcheck uses generic code generator;
  3166 default generators are provided for all suitable HOL types, records
  3167 and datatypes.  Old quickcheck can be re-activated importing theory
  3168 Library/SML_Quickcheck.
  3169 
  3170 * New testing tool Mirabelle for automated proof tools.  Applies
  3171 several tools and tactics like sledgehammer, metis, or quickcheck, to
  3172 every proof step in a theory.  To be used in batch mode via the
  3173 "mirabelle" utility.
  3174 
  3175 * New proof method "sos" (sum of squares) for nonlinear real
  3176 arithmetic (originally due to John Harison). It requires theory
  3177 Library/Sum_Of_Squares.  It is not a complete decision procedure but
  3178 works well in practice on quantifier-free real arithmetic with +, -,
  3179 *, ^, =, <= and <, i.e. boolean combinations of equalities and
  3180 inequalities between polynomials.  It makes use of external
  3181 semidefinite programming solvers.  Method "sos" generates a
  3182 certificate that can be pasted into the proof thus avoiding the need
  3183 to call an external tool every time the proof is checked.  See
  3184 src/HOL/Library/Sum_Of_Squares.
  3185 
  3186 * New method "linarith" invokes existing linear arithmetic decision
  3187 procedure only.
  3188 
  3189 * New command 'atp_minimal' reduces result produced by Sledgehammer.
  3190 
  3191 * New Sledgehammer option "Full Types" in Proof General settings menu.
  3192 Causes full type information to be output to the ATPs.  This slows
  3193 ATPs down considerably but eliminates a source of unsound "proofs"
  3194 that fail later.
  3195 
  3196 * New method "metisFT": A version of metis that uses full type
  3197 information in order to avoid failures of proof reconstruction.
  3198 
  3199 * New evaluator "approximate" approximates an real valued term using
  3200 the same method as the approximation method.
  3201 
  3202 * Method "approximate" now supports arithmetic expressions as
  3203 boundaries of intervals and implements interval splitting and Taylor
  3204 series expansion.
  3205 
  3206 * ML antiquotation @{code_datatype} inserts definition of a datatype
  3207 generated by the code generator; e.g. see src/HOL/Predicate.thy.
  3208 
  3209 * New theory SupInf of the supremum and infimum operators for sets of
  3210 reals.
  3211 
  3212 * New theory Probability, which contains a development of measure
  3213 theory, eventually leading to Lebesgue integration and probability.
  3214 
  3215 * Extended Multivariate Analysis to include derivation and Brouwer's
  3216 fixpoint theorem.
  3217 
  3218 * Reorganization of number theory, INCOMPATIBILITY:
  3219   - new number theory development for nat and int, in theories Divides
  3220     and GCD as well as in new session Number_Theory
  3221   - some constants and facts now suffixed with _nat and _int
  3222     accordingly
  3223   - former session NumberTheory now named Old_Number_Theory, including
  3224     theories Legacy_GCD and Primes (prefer Number_Theory if possible)
  3225   - moved theory Pocklington from src/HOL/Library to
  3226     src/HOL/Old_Number_Theory
  3227 
  3228 * Theory GCD includes functions Gcd/GCD and Lcm/LCM for the gcd and
  3229 lcm of finite and infinite sets. It is shown that they form a complete
  3230 lattice.
  3231 
  3232 * Class semiring_div requires superclass no_zero_divisors and proof of
  3233 div_mult_mult1; theorems div_mult_mult1, div_mult_mult2,
  3234 div_mult_mult1_if, div_mult_mult1 and div_mult_mult2 have been
  3235 generalized to class semiring_div, subsuming former theorems
  3236 zdiv_zmult_zmult1, zdiv_zmult_zmult1_if, zdiv_zmult_zmult1 and
  3237 zdiv_zmult_zmult2.  div_mult_mult1 is now [simp] by default.
  3238 INCOMPATIBILITY.
  3239 
  3240 * Refinements to lattice classes and sets:
  3241   - less default intro/elim rules in locale variant, more default
  3242     intro/elim rules in class variant: more uniformity
  3243   - lemma ge_sup_conv renamed to le_sup_iff, in accordance with
  3244     le_inf_iff
  3245   - dropped lemma alias inf_ACI for inf_aci (same for sup_ACI and
  3246     sup_aci)
  3247   - renamed ACI to inf_sup_aci
  3248   - new class "boolean_algebra"
  3249   - class "complete_lattice" moved to separate theory
  3250     "Complete_Lattice"; corresponding constants (and abbreviations)
  3251     renamed and with authentic syntax:
  3252     Set.Inf ~>    Complete_Lattice.Inf
  3253     Set.Sup ~>    Complete_Lattice.Sup
  3254     Set.INFI ~>   Complete_Lattice.INFI
  3255     Set.SUPR ~>   Complete_Lattice.SUPR
  3256     Set.Inter ~>  Complete_Lattice.Inter
  3257     Set.Union ~>  Complete_Lattice.Union
  3258     Set.INTER ~>  Complete_Lattice.INTER
  3259     Set.UNION ~>  Complete_Lattice.UNION
  3260   - authentic syntax for
  3261     Set.Pow
  3262     Set.image
  3263   - mere abbreviations:
  3264     Set.empty               (for bot)
  3265     Set.UNIV                (for top)
  3266     Set.inter               (for inf, formerly Set.Int)
  3267     Set.union               (for sup, formerly Set.Un)
  3268     Complete_Lattice.Inter  (for Inf)
  3269     Complete_Lattice.Union  (for Sup)
  3270     Complete_Lattice.INTER  (for INFI)
  3271     Complete_Lattice.UNION  (for SUPR)
  3272   - object-logic definitions as far as appropriate
  3273 
  3274 INCOMPATIBILITY.  Care is required when theorems Int_subset_iff or
  3275 Un_subset_iff are explicitly deleted as default simp rules; then also
  3276 their lattice counterparts le_inf_iff and le_sup_iff have to be
  3277 deleted to achieve the desired effect.
  3278 
  3279 * Rules inf_absorb1, inf_absorb2, sup_absorb1, sup_absorb2 are no simp
  3280 rules by default any longer; the same applies to min_max.inf_absorb1
  3281 etc.  INCOMPATIBILITY.
  3282 
  3283 * Rules sup_Int_eq and sup_Un_eq are no longer declared as
  3284 pred_set_conv by default.  INCOMPATIBILITY.
  3285 
  3286 * Power operations on relations and functions are now one dedicated
  3287 constant "compow" with infix syntax "^^".  Power operation on
  3288 multiplicative monoids retains syntax "^" and is now defined generic
  3289 in class power.  INCOMPATIBILITY.
  3290 
  3291 * Relation composition "R O S" now has a more standard argument order:
  3292 "R O S = {(x, z). EX y. (x, y) : R & (y, z) : S}".  INCOMPATIBILITY,
  3293 rewrite propositions with "S O R" --> "R O S". Proofs may occasionally
  3294 break, since the O_assoc rule was not rewritten like this.  Fix using
  3295 O_assoc[symmetric].  The same applies to the curried version "R OO S".
  3296 
  3297 * Function "Inv" is renamed to "inv_into" and function "inv" is now an
  3298 abbreviation for "inv_into UNIV".  Lemmas are renamed accordingly.
  3299 INCOMPATIBILITY.
  3300 
  3301 * Most rules produced by inductive and datatype package have mandatory
  3302 prefixes.  INCOMPATIBILITY.
  3303 
  3304 * Changed "DERIV_intros" to a dynamic fact, which can be augmented by
  3305 the attribute of the same name.  Each of the theorems in the list
  3306 DERIV_intros assumes composition with an additional function and
  3307 matches a variable to the derivative, which has to be solved by the
  3308 Simplifier.  Hence (auto intro!: DERIV_intros) computes the derivative
  3309 of most elementary terms.  Former Maclauren.DERIV_tac and
  3310 Maclauren.deriv_tac should be replaced by (auto intro!: DERIV_intros).
  3311 INCOMPATIBILITY.
  3312 
  3313 * Code generator attributes follow the usual underscore convention:
  3314     code_unfold     replaces    code unfold
  3315     code_post       replaces    code post
  3316     etc.
  3317   INCOMPATIBILITY.
  3318 
  3319 * Renamed methods:
  3320     sizechange -> size_change
  3321     induct_scheme -> induction_schema
  3322   INCOMPATIBILITY.
  3323 
  3324 * Discontinued abbreviation "arbitrary" of constant "undefined".
  3325 INCOMPATIBILITY, use "undefined" directly.
  3326 
  3327 * Renamed theorems:
  3328     Suc_eq_add_numeral_1 -> Suc_eq_plus1
  3329     Suc_eq_add_numeral_1_left -> Suc_eq_plus1_left
  3330     Suc_plus1 -> Suc_eq_plus1
  3331     *anti_sym -> *antisym*
  3332     vector_less_eq_def -> vector_le_def
  3333   INCOMPATIBILITY.
  3334 
  3335 * Added theorem List.map_map as [simp].  Removed List.map_compose.
  3336 INCOMPATIBILITY.
  3337 
  3338 * Removed predicate "M hassize n" (<--> card M = n & finite M).
  3339 INCOMPATIBILITY.
  3340 
  3341 
  3342 *** HOLCF ***
  3343 
  3344 * Theory Representable defines a class "rep" of domains that are
  3345 representable (via an ep-pair) in the universal domain type "udom".
  3346 Instances are provided for all type constructors defined in HOLCF.
  3347 
  3348 * The 'new_domain' command is a purely definitional version of the
  3349 domain package, for representable domains.  Syntax is identical to the
  3350 old domain package.  The 'new_domain' package also supports indirect
  3351 recursion using previously-defined type constructors.  See
  3352 src/HOLCF/ex/New_Domain.thy for examples.
  3353 
  3354 * Method "fixrec_simp" unfolds one step of a fixrec-defined constant
  3355 on the left-hand side of an equation, and then performs
  3356 simplification.  Rewriting is done using rules declared with the
  3357 "fixrec_simp" attribute.  The "fixrec_simp" method is intended as a
  3358 replacement for "fixpat"; see src/HOLCF/ex/Fixrec_ex.thy for examples.
  3359 
  3360 * The pattern-match compiler in 'fixrec' can now handle constructors
  3361 with HOL function types.  Pattern-match combinators for the Pair
  3362 constructor are pre-configured.
  3363 
  3364 * The 'fixrec' package now produces better fixed-point induction rules
  3365 for mutually-recursive definitions:  Induction rules have conclusions
  3366 of the form "P foo bar" instead of "P <foo, bar>".
  3367 
  3368 * The constant "sq_le" (with infix syntax "<<" or "\<sqsubseteq>") has
  3369 been renamed to "below".  The name "below" now replaces "less" in many
  3370 theorem names.  (Legacy theorem names using "less" are still supported
  3371 as well.)
  3372 
  3373 * The 'fixrec' package now supports "bottom patterns".  Bottom
  3374 patterns can be used to generate strictness rules, or to make
  3375 functions more strict (much like the bang-patterns supported by the
  3376 Glasgow Haskell Compiler).  See src/HOLCF/ex/Fixrec_ex.thy for
  3377 examples.
  3378 
  3379 
  3380 *** ML ***
  3381 
  3382 * Support for Poly/ML 5.3.0, with improved reporting of compiler
  3383 errors and run-time exceptions, including detailed source positions.
  3384 
  3385 * Structure Name_Space (formerly NameSpace) now manages uniquely
  3386 identified entries, with some additional information such as source
  3387 position, logical grouping etc.
  3388 
  3389 * Theory and context data is now introduced by the simplified and
  3390 modernized functors Theory_Data, Proof_Data, Generic_Data.  Data needs
  3391 to be pure, but the old TheoryDataFun for mutable data (with explicit
  3392 copy operation) is still available for some time.
  3393 
  3394 * Structure Synchronized (cf. src/Pure/Concurrent/synchronized.ML)
  3395 provides a high-level programming interface to synchronized state
  3396 variables with atomic update.  This works via pure function
  3397 application within a critical section -- its runtime should be as
  3398 short as possible; beware of deadlocks if critical code is nested,
  3399 either directly or indirectly via other synchronized variables!
  3400 
  3401 * Structure Unsynchronized (cf. src/Pure/ML-Systems/unsynchronized.ML)
  3402 wraps raw ML references, explicitly indicating their non-thread-safe
  3403 behaviour.  The Isar toplevel keeps this structure open, to
  3404 accommodate Proof General as well as quick and dirty interactive
  3405 experiments with references.
  3406 
  3407 * PARALLEL_CHOICE and PARALLEL_GOALS provide basic support for
  3408 parallel tactical reasoning.
  3409 
  3410 * Tacticals Subgoal.FOCUS, Subgoal.FOCUS_PREMS, Subgoal.FOCUS_PARAMS
  3411 are similar to SUBPROOF, but are slightly more flexible: only the
  3412 specified parts of the subgoal are imported into the context, and the
  3413 body tactic may introduce new subgoals and schematic variables.
  3414 
  3415 * Old tactical METAHYPS, which does not observe the proof context, has
  3416 been renamed to Old_Goals.METAHYPS and awaits deletion.  Use SUBPROOF
  3417 or Subgoal.FOCUS etc.
  3418 
  3419 * Renamed functor TableFun to Table, and GraphFun to Graph.  (Since
  3420 functors have their own ML name space there is no point to mark them
  3421 separately.)  Minor INCOMPATIBILITY.
  3422 
  3423 * Renamed NamedThmsFun to Named_Thms.  INCOMPATIBILITY.
  3424 
  3425 * Renamed several structures FooBar to Foo_Bar.  Occasional,
  3426 INCOMPATIBILITY.
  3427 
  3428 * Operations of structure Skip_Proof no longer require quick_and_dirty
  3429 mode, which avoids critical setmp.
  3430 
  3431 * Eliminated old Attrib.add_attributes, Method.add_methods and related
  3432 combinators for "args".  INCOMPATIBILITY, need to use simplified
  3433 Attrib/Method.setup introduced in Isabelle2009.
  3434 
  3435 * Proper context for simpset_of, claset_of, clasimpset_of.  May fall
  3436 back on global_simpset_of, global_claset_of, global_clasimpset_of as
  3437 last resort.  INCOMPATIBILITY.
  3438 
  3439 * Display.pretty_thm now requires a proper context (cf. former
  3440 ProofContext.pretty_thm).  May fall back on Display.pretty_thm_global
  3441 or even Display.pretty_thm_without_context as last resort.
  3442 INCOMPATIBILITY.
  3443 
  3444 * Discontinued Display.pretty_ctyp/cterm etc.  INCOMPATIBILITY, use
  3445 Syntax.pretty_typ/term directly, preferably with proper context
  3446 instead of global theory.
  3447 
  3448 
  3449 *** System ***
  3450 
  3451 * Further fine tuning of parallel proof checking, scales up to 8 cores
  3452 (max. speedup factor 5.0).  See also Goal.parallel_proofs in ML and
  3453 usedir option -q.
  3454 
  3455 * Support for additional "Isabelle components" via etc/components, see
  3456 also the system manual.
  3457 
  3458 * The isabelle makeall tool now operates on all components with
  3459 IsaMakefile, not just hardwired "logics".
  3460 
  3461 * Removed "compress" option from isabelle-process and isabelle usedir;
  3462 this is always enabled.
  3463 
  3464 * Discontinued support for Poly/ML 4.x versions.
  3465 
  3466 * Isabelle tool "wwwfind" provides web interface for 'find_theorems'
  3467 on a given logic image.  This requires the lighttpd webserver and is
  3468 currently supported on Linux only.
  3469 
  3470 
  3471 
  3472 New in Isabelle2009 (April 2009)
  3473 --------------------------------
  3474 
  3475 *** General ***
  3476 
  3477 * Simplified main Isabelle executables, with less surprises on
  3478 case-insensitive file-systems (such as Mac OS).
  3479 
  3480   - The main Isabelle tool wrapper is now called "isabelle" instead of
  3481     "isatool."
  3482 
  3483   - The former "isabelle" alias for "isabelle-process" has been
  3484     removed (should rarely occur to regular users).
  3485 
  3486   - The former "isabelle-interface" and its alias "Isabelle" have been
  3487     removed (interfaces are now regular Isabelle tools).
  3488 
  3489 Within scripts and make files, the Isabelle environment variables
  3490 ISABELLE_TOOL and ISABELLE_PROCESS replace old ISATOOL and ISABELLE,
  3491 respectively.  (The latter are still available as legacy feature.)
  3492 
  3493 The old isabelle-interface wrapper could react in confusing ways if
  3494 the interface was uninstalled or changed otherwise.  Individual
  3495 interface tool configuration is now more explicit, see also the
  3496 Isabelle system manual.  In particular, Proof General is now available
  3497 via "isabelle emacs".
  3498 
  3499 INCOMPATIBILITY, need to adapt derivative scripts.  Users may need to
  3500 purge installed copies of Isabelle executables and re-run "isabelle
  3501 install -p ...", or use symlinks.
  3502 
  3503 * The default for ISABELLE_HOME_USER is now ~/.isabelle instead of the
  3504 old ~/isabelle, which was slightly non-standard and apt to cause
  3505 surprises on case-insensitive file-systems (such as Mac OS).
  3506 
  3507 INCOMPATIBILITY, need to move existing ~/isabelle/etc,
  3508 ~/isabelle/heaps, ~/isabelle/browser_info to the new place.  Special
  3509 care is required when using older releases of Isabelle.  Note that
  3510 ISABELLE_HOME_USER can be changed in Isabelle/etc/settings of any
  3511 Isabelle distribution, in order to use the new ~/.isabelle uniformly.
  3512 
  3513 * Proofs of fully specified statements are run in parallel on
  3514 multi-core systems.  A speedup factor of 2.5 to 3.2 can be expected on
  3515 a regular 4-core machine, if the initial heap space is made reasonably
  3516 large (cf. Poly/ML option -H).  (Requires Poly/ML 5.2.1 or later.)
  3517 
  3518 * The main reference manuals ("isar-ref", "implementation", and
  3519 "system") have been updated and extended.  Formally checked references
  3520 as hyperlinks are now available uniformly.
  3521 
  3522 
  3523 *** Pure ***
  3524 
  3525 * Complete re-implementation of locales.  INCOMPATIBILITY in several
  3526 respects.  The most important changes are listed below.  See the
  3527 Tutorial on Locales ("locales" manual) for details.
  3528 
  3529 - In locale expressions, instantiation replaces renaming.  Parameters
  3530 must be declared in a for clause.  To aid compatibility with previous
  3531 parameter inheritance, in locale declarations, parameters that are not
  3532 'touched' (instantiation position "_" or omitted) are implicitly added
  3533 with their syntax at the beginning of the for clause.
  3534 
  3535 - Syntax from abbreviations and definitions in locales is available in
  3536 locale expressions and context elements.  The latter is particularly
  3537 useful in locale declarations.
  3538 
  3539 - More flexible mechanisms to qualify names generated by locale
  3540 expressions.  Qualifiers (prefixes) may be specified in locale
  3541 expressions, and can be marked as mandatory (syntax: "name!:") or
  3542 optional (syntax "name?:").  The default depends for plain "name:"
  3543 depends on the situation where a locale expression is used: in
  3544 commands 'locale' and 'sublocale' prefixes are optional, in
  3545 'interpretation' and 'interpret' prefixes are mandatory.  The old
  3546 implicit qualifiers derived from the parameter names of a locale are
  3547 no longer generated.
  3548 
  3549 - Command "sublocale l < e" replaces "interpretation l < e".  The
  3550 instantiation clause in "interpretation" and "interpret" (square
  3551 brackets) is no longer available.  Use locale expressions.
  3552 
  3553 - When converting proof scripts, mandatory qualifiers in
  3554 'interpretation' and 'interpret' should be retained by default, even
  3555 if this is an INCOMPATIBILITY compared to former behavior.  In the
  3556 worst case, use the "name?:" form for non-mandatory ones.  Qualifiers
  3557 in locale expressions range over a single locale instance only.
  3558 
  3559 - Dropped locale element "includes".  This is a major INCOMPATIBILITY.
  3560 In existing theorem specifications replace the includes element by the
  3561 respective context elements of the included locale, omitting those
  3562 that are already present in the theorem specification.  Multiple
  3563 assume elements of a locale should be replaced by a single one
  3564 involving the locale predicate.  In the proof body, declarations (most
  3565 notably theorems) may be regained by interpreting the respective
  3566 locales in the proof context as required (command "interpret").
  3567 
  3568 If using "includes" in replacement of a target solely because the
  3569 parameter types in the theorem are not as general as in the target,
  3570 consider declaring a new locale with additional type constraints on
  3571 the parameters (context element "constrains").
  3572 
  3573 - Discontinued "locale (open)".  INCOMPATIBILITY.
  3574 
  3575 - Locale interpretation commands no longer attempt to simplify goal.
  3576 INCOMPATIBILITY: in rare situations the generated goal differs.  Use
  3577 methods intro_locales and unfold_locales to clarify.
  3578 
  3579 - Locale interpretation commands no longer accept interpretation
  3580 attributes.  INCOMPATIBILITY.
  3581 
  3582 * Class declaration: so-called "base sort" must not be given in import
  3583 list any longer, but is inferred from the specification.  Particularly
  3584 in HOL, write
  3585 
  3586     class foo = ...
  3587 
  3588 instead of
  3589 
  3590     class foo = type + ...
  3591 
  3592 * Class target: global versions of theorems stemming do not carry a
  3593 parameter prefix any longer.  INCOMPATIBILITY.
  3594 
  3595 * Class 'instance' command no longer accepts attached definitions.
  3596 INCOMPATIBILITY, use proper 'instantiation' target instead.
  3597 
  3598 * Recovered hiding of consts, which was accidentally broken in
  3599 Isabelle2007.  Potential INCOMPATIBILITY, ``hide const c'' really
  3600 makes c inaccessible; consider using ``hide (open) const c'' instead.
  3601 
  3602 * Slightly more coherent Pure syntax, with updated documentation in
  3603 isar-ref manual.  Removed locales meta_term_syntax and
  3604 meta_conjunction_syntax: TERM and &&& (formerly &&) are now permanent,
  3605 INCOMPATIBILITY in rare situations.  Note that &&& should not be used
  3606 directly in regular applications.
  3607 
  3608 * There is a new syntactic category "float_const" for signed decimal
  3609 fractions (e.g. 123.45 or -123.45).
  3610 
  3611 * Removed exotic 'token_translation' command.  INCOMPATIBILITY, use ML
  3612 interface with 'setup' command instead.
  3613 
  3614 * Command 'local_setup' is similar to 'setup', but operates on a local
  3615 theory context.
  3616 
  3617 * The 'axiomatization' command now only works within a global theory
  3618 context.  INCOMPATIBILITY.
  3619 
  3620 * Goal-directed proof now enforces strict proof irrelevance wrt. sort
  3621 hypotheses.  Sorts required in the course of reasoning need to be
  3622 covered by the constraints in the initial statement, completed by the
  3623 type instance information of the background theory.  Non-trivial sort
  3624 hypotheses, which rarely occur in practice, may be specified via
  3625 vacuous propositions of the form SORT_CONSTRAINT('a::c).  For example:
  3626 
  3627   lemma assumes "SORT_CONSTRAINT('a::empty)" shows False ...
  3628 
  3629 The result contains an implicit sort hypotheses as before --
  3630 SORT_CONSTRAINT premises are eliminated as part of the canonical rule
  3631 normalization.
  3632 
  3633 * Generalized Isar history, with support for linear undo, direct state
  3634 addressing etc.
  3635 
  3636 * Changed defaults for unify configuration options:
  3637 
  3638   unify_trace_bound = 50 (formerly 25)
  3639   unify_search_bound = 60 (formerly 30)
  3640 
  3641 * Different bookkeeping for code equations (INCOMPATIBILITY):
  3642 
  3643   a) On theory merge, the last set of code equations for a particular
  3644      constant is taken (in accordance with the policy applied by other
  3645      parts of the code generator framework).
  3646 
  3647   b) Code equations stemming from explicit declarations (e.g. code
  3648      attribute) gain priority over default code equations stemming
  3649      from definition, primrec, fun etc.
  3650 
  3651 * Keyword 'code_exception' now named 'code_abort'.  INCOMPATIBILITY.
  3652 
  3653 * Unified theorem tables for both code generators.  Thus [code
  3654 func] has disappeared and only [code] remains.  INCOMPATIBILITY.
  3655 
  3656 * Command 'find_consts' searches for constants based on type and name
  3657 patterns, e.g.
  3658 
  3659     find_consts "_ => bool"
  3660 
  3661 By default, matching is against subtypes, but it may be restricted to
  3662 the whole type.  Searching by name is possible.  Multiple queries are
  3663 conjunctive and queries may be negated by prefixing them with a
  3664 hyphen:
  3665 
  3666     find_consts strict: "_ => bool" name: "Int" -"int => int"
  3667 
  3668 * New 'find_theorems' criterion "solves" matches theorems that
  3669 directly solve the current goal (modulo higher-order unification).
  3670 
  3671 * Auto solve feature for main theorem statements: whenever a new goal
  3672 is stated, "find_theorems solves" is called; any theorems that could
  3673 solve the lemma directly are listed as part of the goal state.
  3674 Cf. associated options in Proof General Isabelle settings menu,
  3675 enabled by default, with reasonable timeout for pathological cases of
  3676 higher-order unification.
  3677 
  3678 
  3679 *** Document preparation ***
  3680 
  3681 * Antiquotation @{lemma} now imitates a regular terminal proof,
  3682 demanding keyword 'by' and supporting the full method expression
  3683 syntax just like the Isar command 'by'.
  3684 
  3685 
  3686 *** HOL ***
  3687 
  3688 * Integrated main parts of former image HOL-Complex with HOL.  Entry
  3689 points Main and Complex_Main remain as before.
  3690 
  3691 * Logic image HOL-Plain provides a minimal HOL with the most important
  3692 tools available (inductive, datatype, primrec, ...).  This facilitates
  3693 experimentation and tool development.  Note that user applications
  3694 (and library theories) should never refer to anything below theory
  3695 Main, as before.
  3696 
  3697 * Logic image HOL-Main stops at theory Main, and thus facilitates
  3698 experimentation due to shorter build times.
  3699 
  3700 * Logic image HOL-NSA contains theories of nonstandard analysis which
  3701 were previously part of former HOL-Complex.  Entry point Hyperreal
  3702 remains valid, but theories formerly using Complex_Main should now use
  3703 new entry point Hypercomplex.
  3704 
  3705 * Generic ATP manager for Sledgehammer, based on ML threads instead of
  3706 Posix processes.  Avoids potentially expensive forking of the ML
  3707 process.  New thread-based implementation also works on non-Unix
  3708 platforms (Cygwin).  Provers are no longer hardwired, but defined
  3709 within the theory via plain ML wrapper functions.  Basic Sledgehammer
  3710 commands are covered in the isar-ref manual.
  3711 
  3712 * Wrapper scripts for remote SystemOnTPTP service allows to use
  3713 sledgehammer without local ATP installation (Vampire etc.). Other
  3714 provers may be included via suitable ML wrappers, see also
  3715 src/HOL/ATP_Linkup.thy.
  3716 
  3717 * ATP selection (E/Vampire/Spass) is now via Proof General's settings
  3718 menu.
  3719 
  3720 * The metis method no longer fails because the theorem is too trivial
  3721 (contains the empty clause).
  3722 
  3723 * The metis method now fails in the usual manner, rather than raising
  3724 an exception, if it determines that it cannot prove the theorem.
  3725 
  3726 * Method "coherent" implements a prover for coherent logic (see also
  3727 src/Tools/coherent.ML).
  3728 
  3729 * Constants "undefined" and "default" replace "arbitrary".  Usually
  3730 "undefined" is the right choice to replace "arbitrary", though
  3731 logically there is no difference.  INCOMPATIBILITY.
  3732 
  3733 * Command "value" now integrates different evaluation mechanisms.  The
  3734 result of the first successful evaluation mechanism is printed.  In
  3735 square brackets a particular named evaluation mechanisms may be
  3736 specified (currently, [SML], [code] or [nbe]).  See further
  3737 src/HOL/ex/Eval_Examples.thy.
  3738 
  3739 * Normalization by evaluation now allows non-leftlinear equations.
  3740 Declare with attribute [code nbe].
  3741 
  3742 * Methods "case_tac" and "induct_tac" now refer to the very same rules
  3743 as the structured Isar versions "cases" and "induct", cf. the
  3744 corresponding "cases" and "induct" attributes.  Mutual induction rules
  3745 are now presented as a list of individual projections
  3746 (e.g. foo_bar.inducts for types foo and bar); the old format with
  3747 explicit HOL conjunction is no longer supported.  INCOMPATIBILITY, in
  3748 rare situations a different rule is selected --- notably nested tuple
  3749 elimination instead of former prod.exhaust: use explicit (case_tac t
  3750 rule: prod.exhaust) here.
  3751 
  3752 * Attributes "cases", "induct", "coinduct" support "del" option.
  3753 
  3754 * Removed fact "case_split_thm", which duplicates "case_split".
  3755 
  3756 * The option datatype has been moved to a new theory Option.  Renamed
  3757 option_map to Option.map, and o2s to Option.set, INCOMPATIBILITY.
  3758 
  3759 * New predicate "strict_mono" classifies strict functions on partial
  3760 orders.  With strict functions on linear orders, reasoning about
  3761 (in)equalities is facilitated by theorems "strict_mono_eq",
  3762 "strict_mono_less_eq" and "strict_mono_less".
  3763 
  3764 * Some set operations are now proper qualified constants with
  3765 authentic syntax.  INCOMPATIBILITY:
  3766 
  3767     op Int ~>   Set.Int
  3768     op Un ~>    Set.Un
  3769     INTER ~>    Set.INTER
  3770     UNION ~>    Set.UNION
  3771     Inter ~>    Set.Inter
  3772     Union ~>    Set.Union
  3773     {} ~>       Set.empty
  3774     UNIV ~>     Set.UNIV
  3775 
  3776 * Class complete_lattice with operations Inf, Sup, INFI, SUPR now in
  3777 theory Set.
  3778 
  3779 * Auxiliary class "itself" has disappeared -- classes without any
  3780 parameter are treated as expected by the 'class' command.
  3781 
  3782 * Leibnitz's Series for Pi and the arcus tangens and logarithm series.
  3783 
  3784 * Common decision procedures (Cooper, MIR, Ferrack, Approximation,
  3785 Dense_Linear_Order) are now in directory HOL/Decision_Procs.
  3786 
  3787 * Theory src/HOL/Decision_Procs/Approximation provides the new proof
  3788 method "approximation".  It proves formulas on real values by using
  3789 interval arithmetic.  In the formulas are also the transcendental
  3790 functions sin, cos, tan, atan, ln, exp and the constant pi are
  3791 allowed. For examples see
  3792 src/HOL/Descision_Procs/ex/Approximation_Ex.thy.
  3793 
  3794 * Theory "Reflection" now resides in HOL/Library.
  3795 
  3796 * Entry point to Word library now simply named "Word".
  3797 INCOMPATIBILITY.
  3798 
  3799 * Made source layout more coherent with logical distribution
  3800 structure:
  3801 
  3802     src/HOL/Library/RType.thy ~> src/HOL/Typerep.thy
  3803     src/HOL/Library/Code_Message.thy ~> src/HOL/
  3804     src/HOL/Library/GCD.thy ~> src/HOL/
  3805     src/HOL/Library/Order_Relation.thy ~> src/HOL/
  3806     src/HOL/Library/Parity.thy ~> src/HOL/
  3807     src/HOL/Library/Univ_Poly.thy ~> src/HOL/
  3808     src/HOL/Real/ContNotDenum.thy ~> src/HOL/Library/
  3809     src/HOL/Real/Lubs.thy ~> src/HOL/
  3810     src/HOL/Real/PReal.thy ~> src/HOL/
  3811     src/HOL/Real/Rational.thy ~> src/HOL/
  3812     src/HOL/Real/RComplete.thy ~> src/HOL/
  3813     src/HOL/Real/RealDef.thy ~> src/HOL/
  3814     src/HOL/Real/RealPow.thy ~> src/HOL/
  3815     src/HOL/Real/Real.thy ~> src/HOL/
  3816     src/HOL/Complex/Complex_Main.thy ~> src/HOL/
  3817     src/HOL/Complex/Complex.thy ~> src/HOL/
  3818     src/HOL/Complex/FrechetDeriv.thy ~> src/HOL/Library/
  3819     src/HOL/Complex/Fundamental_Theorem_Algebra.thy ~> src/HOL/Library/
  3820     src/HOL/Hyperreal/Deriv.thy ~> src/HOL/
  3821     src/HOL/Hyperreal/Fact.thy ~> src/HOL/
  3822     src/HOL/Hyperreal/Integration.thy ~> src/HOL/
  3823     src/HOL/Hyperreal/Lim.thy ~> src/HOL/
  3824     src/HOL/Hyperreal/Ln.thy ~> src/HOL/
  3825     src/HOL/Hyperreal/Log.thy ~> src/HOL/
  3826     src/HOL/Hyperreal/MacLaurin.thy ~> src/HOL/
  3827     src/HOL/Hyperreal/NthRoot.thy ~> src/HOL/
  3828     src/HOL/Hyperreal/Series.thy ~> src/HOL/
  3829     src/HOL/Hyperreal/SEQ.thy ~> src/HOL/
  3830     src/HOL/Hyperreal/Taylor.thy ~> src/HOL/
  3831     src/HOL/Hyperreal/Transcendental.thy ~> src/HOL/
  3832     src/HOL/Real/Float ~> src/HOL/Library/
  3833     src/HOL/Real/HahnBanach ~> src/HOL/HahnBanach
  3834     src/HOL/Real/RealVector.thy ~> src/HOL/
  3835 
  3836     src/HOL/arith_data.ML ~> src/HOL/Tools
  3837     src/HOL/hologic.ML ~> src/HOL/Tools
  3838     src/HOL/simpdata.ML ~> src/HOL/Tools
  3839     src/HOL/int_arith1.ML ~> src/HOL/Tools/int_arith.ML
  3840     src/HOL/int_factor_simprocs.ML ~> src/HOL/Tools
  3841     src/HOL/nat_simprocs.ML ~> src/HOL/Tools
  3842     src/HOL/Real/float_arith.ML ~> src/HOL/Tools
  3843     src/HOL/Real/float_syntax.ML ~> src/HOL/Tools
  3844     src/HOL/Real/rat_arith.ML ~> src/HOL/Tools
  3845     src/HOL/Real/real_arith.ML ~> src/HOL/Tools
  3846 
  3847     src/HOL/Library/Array.thy ~> src/HOL/Imperative_HOL
  3848     src/HOL/Library/Heap_Monad.thy ~> src/HOL/Imperative_HOL
  3849     src/HOL/Library/Heap.thy ~> src/HOL/Imperative_HOL
  3850     src/HOL/Library/Imperative_HOL.thy ~> src/HOL/Imperative_HOL
  3851     src/HOL/Library/Ref.thy ~> src/HOL/Imperative_HOL
  3852     src/HOL/Library/Relational.thy ~> src/HOL/Imperative_HOL
  3853 
  3854 * If methods "eval" and "evaluation" encounter a structured proof
  3855 state with !!/==>, only the conclusion is evaluated to True (if
  3856 possible), avoiding strange error messages.
  3857 
  3858 * Method "sizechange" automates termination proofs using (a
  3859 modification of) the size-change principle.  Requires SAT solver.  See
  3860 src/HOL/ex/Termination.thy for examples.
  3861 
  3862 * Simplifier: simproc for let expressions now unfolds if bound
  3863 variable occurs at most once in let expression body.  INCOMPATIBILITY.
  3864 
  3865 * Method "arith": Linear arithmetic now ignores all inequalities when
  3866 fast_arith_neq_limit is exceeded, instead of giving up entirely.
  3867 
  3868 * New attribute "arith" for facts that should always be used
  3869 automatically by arithmetic. It is intended to be used locally in
  3870 proofs, e.g.
  3871 
  3872   assumes [arith]: "x > 0"
  3873 
  3874 Global usage is discouraged because of possible performance impact.
  3875 
  3876 * New classes "top" and "bot" with corresponding operations "top" and
  3877 "bot" in theory Orderings; instantiation of class "complete_lattice"
  3878 requires instantiation of classes "top" and "bot".  INCOMPATIBILITY.
  3879 
  3880 * Changed definition lemma "less_fun_def" in order to provide an
  3881 instance for preorders on functions; use lemma "less_le" instead.
  3882 INCOMPATIBILITY.
  3883 
  3884 * Theory Orderings: class "wellorder" moved here, with explicit
  3885 induction rule "less_induct" as assumption.  For instantiation of
  3886 "wellorder" by means of predicate "wf", use rule wf_wellorderI.
  3887 INCOMPATIBILITY.
  3888 
  3889 * Theory Orderings: added class "preorder" as superclass of "order".
  3890 INCOMPATIBILITY: Instantiation proofs for order, linorder
  3891 etc. slightly changed.  Some theorems named order_class.* now named
  3892 preorder_class.*.
  3893 
  3894 * Theory Relation: renamed "refl" to "refl_on", "reflexive" to "refl,
  3895 "diag" to "Id_on".
  3896 
  3897 * Theory Finite_Set: added a new fold combinator of type
  3898 
  3899   ('a => 'b => 'b) => 'b => 'a set => 'b
  3900 
  3901 Occasionally this is more convenient than the old fold combinator
  3902 which is now defined in terms of the new one and renamed to
  3903 fold_image.
  3904 
  3905 * Theories Ring_and_Field and OrderedGroup: The lemmas "group_simps"
  3906 and "ring_simps" have been replaced by "algebra_simps" (which can be
  3907 extended with further lemmas!).  At the moment both still exist but
  3908 the former will disappear at some point.
  3909 
  3910 * Theory Power: Lemma power_Suc is now declared as a simp rule in
  3911 class recpower.  Type-specific simp rules for various recpower types
  3912 have been removed.  INCOMPATIBILITY, rename old lemmas as follows:
  3913 
  3914 rat_power_0    -> power_0
  3915 rat_power_Suc  -> power_Suc
  3916 realpow_0      -> power_0
  3917 realpow_Suc    -> power_Suc
  3918 complexpow_0   -> power_0
  3919 complexpow_Suc -> power_Suc
  3920 power_poly_0   -> power_0
  3921 power_poly_Suc -> power_Suc
  3922 
  3923 * Theories Ring_and_Field and Divides: Definition of "op dvd" has been
  3924 moved to separate class dvd in Ring_and_Field; a couple of lemmas on
  3925 dvd has been generalized to class comm_semiring_1.  Likewise a bunch
  3926 of lemmas from Divides has been generalized from nat to class
  3927 semiring_div.  INCOMPATIBILITY.  This involves the following theorem
  3928 renames resulting from duplicate elimination:
  3929 
  3930     dvd_def_mod ~>          dvd_eq_mod_eq_0
  3931     zero_dvd_iff ~>         dvd_0_left_iff
  3932     dvd_0 ~>                dvd_0_right
  3933     DIVISION_BY_ZERO_DIV ~> div_by_0
  3934     DIVISION_BY_ZERO_MOD ~> mod_by_0
  3935     mult_div ~>             div_mult_self2_is_id
  3936     mult_mod ~>             mod_mult_self2_is_0
  3937 
  3938 * Theory IntDiv: removed many lemmas that are instances of class-based
  3939 generalizations (from Divides and Ring_and_Field).  INCOMPATIBILITY,
  3940 rename old lemmas as follows:
  3941 
  3942 dvd_diff               -> nat_dvd_diff
  3943 dvd_zminus_iff         -> dvd_minus_iff
  3944 mod_add1_eq            -> mod_add_eq
  3945 mod_mult1_eq           -> mod_mult_right_eq
  3946 mod_mult1_eq'          -> mod_mult_left_eq
  3947 mod_mult_distrib_mod   -> mod_mult_eq
  3948 nat_mod_add_left_eq    -> mod_add_left_eq
  3949 nat_mod_add_right_eq   -> mod_add_right_eq
  3950 nat_mod_div_trivial    -> mod_div_trivial
  3951 nat_mod_mod_trivial    -> mod_mod_trivial
  3952 zdiv_zadd_self1        -> div_add_self1
  3953 zdiv_zadd_self2        -> div_add_self2
  3954 zdiv_zmult_self1       -> div_mult_self2_is_id
  3955 zdiv_zmult_self2       -> div_mult_self1_is_id
  3956 zdvd_triv_left         -> dvd_triv_left
  3957 zdvd_triv_right        -> dvd_triv_right
  3958 zdvd_zmult_cancel_disj -> dvd_mult_cancel_left
  3959 zmod_eq0_zdvd_iff      -> dvd_eq_mod_eq_0[symmetric]
  3960 zmod_zadd_left_eq      -> mod_add_left_eq
  3961 zmod_zadd_right_eq     -> mod_add_right_eq
  3962 zmod_zadd_self1        -> mod_add_self1
  3963 zmod_zadd_self2        -> mod_add_self2
  3964 zmod_zadd1_eq          -> mod_add_eq
  3965 zmod_zdiff1_eq         -> mod_diff_eq
  3966 zmod_zdvd_zmod         -> mod_mod_cancel
  3967 zmod_zmod_cancel       -> mod_mod_cancel
  3968 zmod_zmult_self1       -> mod_mult_self2_is_0
  3969 zmod_zmult_self2       -> mod_mult_self1_is_0
  3970 zmod_1                 -> mod_by_1
  3971 zdiv_1                 -> div_by_1
  3972 zdvd_abs1              -> abs_dvd_iff
  3973 zdvd_abs2              -> dvd_abs_iff
  3974 zdvd_refl              -> dvd_refl
  3975 zdvd_trans             -> dvd_trans
  3976 zdvd_zadd              -> dvd_add
  3977 zdvd_zdiff             -> dvd_diff
  3978 zdvd_zminus_iff        -> dvd_minus_iff
  3979 zdvd_zminus2_iff       -> minus_dvd_iff
  3980 zdvd_zmultD            -> dvd_mult_right
  3981 zdvd_zmultD2           -> dvd_mult_left
  3982 zdvd_zmult_mono        -> mult_dvd_mono
  3983 zdvd_0_right           -> dvd_0_right
  3984 zdvd_0_left            -> dvd_0_left_iff
  3985 zdvd_1_left            -> one_dvd
  3986 zminus_dvd_iff         -> minus_dvd_iff
  3987 
  3988 * Theory Rational: 'Fract k 0' now equals '0'.  INCOMPATIBILITY.
  3989 
  3990 * The real numbers offer decimal input syntax: 12.34 is translated
  3991 into 1234/10^2. This translation is not reversed upon output.
  3992 
  3993 * Theory Library/Polynomial defines an abstract type 'a poly of
  3994 univariate polynomials with coefficients of type 'a.  In addition to
  3995 the standard ring operations, it also supports div and mod.  Code
  3996 generation is also supported, using list-style constructors.
  3997 
  3998 * Theory Library/Inner_Product defines a class of real_inner for real
  3999 inner product spaces, with an overloaded operation inner :: 'a => 'a
  4000 => real.  Class real_inner is a subclass of real_normed_vector from
  4001 theory RealVector.
  4002 
  4003 * Theory Library/Product_Vector provides instances for the product
  4004 type 'a * 'b of several classes from RealVector and Inner_Product.
  4005 Definitions of addition, subtraction, scalar multiplication, norms,
  4006 and inner products are included.
  4007 
  4008 * Theory Library/Bit defines the field "bit" of integers modulo 2.  In
  4009 addition to the field operations, numerals and case syntax are also
  4010 supported.
  4011 
  4012 * Theory Library/Diagonalize provides constructive version of Cantor's
  4013 first diagonalization argument.
  4014 
  4015 * Theory Library/GCD: Curried operations gcd, lcm (for nat) and zgcd,
  4016 zlcm (for int); carried together from various gcd/lcm developements in
  4017 the HOL Distribution.  Constants zgcd and zlcm replace former igcd and
  4018 ilcm; corresponding theorems renamed accordingly.  INCOMPATIBILITY,
  4019 may recover tupled syntax as follows:
  4020 
  4021     hide (open) const gcd
  4022     abbreviation gcd where
  4023       "gcd == (%(a, b). GCD.gcd a b)"
  4024     notation (output)
  4025       GCD.gcd ("gcd '(_, _')")
  4026 
  4027 The same works for lcm, zgcd, zlcm.
  4028 
  4029 * Theory Library/Nat_Infinity: added addition, numeral syntax and more
  4030 instantiations for algebraic structures.  Removed some duplicate
  4031 theorems.  Changes in simp rules.  INCOMPATIBILITY.
  4032 
  4033 * ML antiquotation @{code} takes a constant as argument and generates
  4034 corresponding code in background and inserts name of the corresponding
  4035 resulting ML value/function/datatype constructor binding in place.
  4036 All occurrences of @{code} with a single ML block are generated
  4037 simultaneously.  Provides a generic and safe interface for
  4038 instrumentalizing code generation.  See
  4039 src/HOL/Decision_Procs/Ferrack.thy for a more ambitious application.
  4040 In future you ought to refrain from ad-hoc compiling generated SML
  4041 code on the ML toplevel.  Note that (for technical reasons) @{code}
  4042 cannot refer to constants for which user-defined serializations are
  4043 set.  Refer to the corresponding ML counterpart directly in that
  4044 cases.
  4045 
  4046 * Command 'rep_datatype': instead of theorem names the command now
  4047 takes a list of terms denoting the constructors of the type to be
  4048 represented as datatype.  The characteristic theorems have to be
  4049 proven.  INCOMPATIBILITY.  Also observe that the following theorems
  4050 have disappeared in favour of existing ones:
  4051 
  4052     unit_induct                 ~> unit.induct
  4053     prod_induct                 ~> prod.induct
  4054     sum_induct                  ~> sum.induct
  4055     Suc_Suc_eq                  ~> nat.inject
  4056     Suc_not_Zero Zero_not_Suc   ~> nat.distinct
  4057 
  4058 
  4059 *** HOL-Algebra ***
  4060 
  4061 * New locales for orders and lattices where the equivalence relation
  4062 is not restricted to equality.  INCOMPATIBILITY: all order and lattice
  4063 locales use a record structure with field eq for the equivalence.
  4064 
  4065 * New theory of factorial domains.
  4066 
  4067 * Units_l_inv and Units_r_inv are now simp rules by default.
  4068 INCOMPATIBILITY.  Simplifier proof that require deletion of l_inv
  4069 and/or r_inv will now also require deletion of these lemmas.
  4070 
  4071 * Renamed the following theorems, INCOMPATIBILITY:
  4072 
  4073 UpperD ~> Upper_memD
  4074 LowerD ~> Lower_memD
  4075 least_carrier ~> least_closed
  4076 greatest_carrier ~> greatest_closed
  4077 greatest_Lower_above ~> greatest_Lower_below
  4078 one_zero ~> carrier_one_zero
  4079 one_not_zero ~> carrier_one_not_zero  (collision with assumption)
  4080 
  4081 
  4082 *** HOL-Nominal ***
  4083 
  4084 * Nominal datatypes can now contain type-variables.
  4085 
  4086 * Commands 'nominal_inductive' and 'equivariance' work with local
  4087 theory targets.
  4088 
  4089 * Nominal primrec can now works with local theory targets and its
  4090 specification syntax now conforms to the general format as seen in
  4091 'inductive' etc.
  4092 
  4093 * Method "perm_simp" honours the standard simplifier attributes
  4094 (no_asm), (no_asm_use) etc.
  4095 
  4096 * The new predicate #* is defined like freshness, except that on the
  4097 left hand side can be a set or list of atoms.
  4098 
  4099 * Experimental command 'nominal_inductive2' derives strong induction
  4100 principles for inductive definitions.  In contrast to
  4101 'nominal_inductive', which can only deal with a fixed number of
  4102 binders, it can deal with arbitrary expressions standing for sets of
  4103 atoms to be avoided.  The only inductive definition we have at the
  4104 moment that needs this generalisation is the typing rule for Lets in
  4105 the algorithm W:
  4106 
  4107  Gamma |- t1 : T1   (x,close Gamma T1)::Gamma |- t2 : T2   x#Gamma
  4108  -----------------------------------------------------------------
  4109          Gamma |- Let x be t1 in t2 : T2
  4110 
  4111 In this rule one wants to avoid all the binders that are introduced by
  4112 "close Gamma T1".  We are looking for other examples where this
  4113 feature might be useful.  Please let us know.
  4114 
  4115 
  4116 *** HOLCF ***
  4117 
  4118 * Reimplemented the simplification procedure for proving continuity
  4119 subgoals.  The new simproc is extensible; users can declare additional
  4120 continuity introduction rules with the attribute [cont2cont].
  4121 
  4122 * The continuity simproc now uses a different introduction rule for
  4123 solving continuity subgoals on terms with lambda abstractions.  In
  4124 some rare cases the new simproc may fail to solve subgoals that the
  4125 old one could solve, and "simp add: cont2cont_LAM" may be necessary.
  4126 Potential INCOMPATIBILITY.
  4127 
  4128 * Command 'fixrec': specification syntax now conforms to the general
  4129 format as seen in 'inductive' etc.  See src/HOLCF/ex/Fixrec_ex.thy for
  4130 examples.  INCOMPATIBILITY.
  4131 
  4132 
  4133 *** ZF ***
  4134 
  4135 * Proof of Zorn's Lemma for partial orders.
  4136 
  4137 
  4138 *** ML ***
  4139 
  4140 * Multithreading for Poly/ML 5.1/5.2 is no longer supported, only for
  4141 Poly/ML 5.2.1 or later.  Important note: the TimeLimit facility
  4142 depends on multithreading, so timouts will not work before Poly/ML
  4143 5.2.1!
  4144 
  4145 * High-level support for concurrent ML programming, see
  4146 src/Pure/Cuncurrent.  The data-oriented model of "future values" is
  4147 particularly convenient to organize independent functional
  4148 computations.  The concept of "synchronized variables" provides a
  4149 higher-order interface for components with shared state, avoiding the
  4150 delicate details of mutexes and condition variables.  (Requires
  4151 Poly/ML 5.2.1 or later.)
  4152 
  4153 * ML bindings produced via Isar commands are stored within the Isar
  4154 context (theory or proof).  Consequently, commands like 'use' and 'ML'
  4155 become thread-safe and work with undo as expected (concerning
  4156 top-level bindings, not side-effects on global references).
  4157 INCOMPATIBILITY, need to provide proper Isar context when invoking the
  4158 compiler at runtime; really global bindings need to be given outside a
  4159 theory.  (Requires Poly/ML 5.2 or later.)
  4160 
  4161 * Command 'ML_prf' is analogous to 'ML' but works within a proof
  4162 context.  Top-level ML bindings are stored within the proof context in
  4163 a purely sequential fashion, disregarding the nested proof structure.
  4164 ML bindings introduced by 'ML_prf' are discarded at the end of the
  4165 proof.  (Requires Poly/ML 5.2 or later.)
  4166 
  4167 * Simplified ML attribute and method setup, cf. functions Attrib.setup
  4168 and Method.setup, as well as Isar commands 'attribute_setup' and
  4169 'method_setup'.  INCOMPATIBILITY for 'method_setup', need to simplify
  4170 existing code accordingly, or use plain 'setup' together with old
  4171 Method.add_method.
  4172 
  4173 * Simplified ML oracle interface Thm.add_oracle promotes 'a -> cterm
  4174 to 'a -> thm, while results are always tagged with an authentic oracle
  4175 name.  The Isar command 'oracle' is now polymorphic, no argument type
  4176 is specified.  INCOMPATIBILITY, need to simplify existing oracle code
  4177 accordingly.  Note that extra performance may be gained by producing
  4178 the cterm carefully, avoiding slow Thm.cterm_of.
  4179 
  4180 * Simplified interface for defining document antiquotations via
  4181 ThyOutput.antiquotation, ThyOutput.output, and optionally
  4182 ThyOutput.maybe_pretty_source.  INCOMPATIBILITY, need to simplify user
  4183 antiquotations accordingly, see src/Pure/Thy/thy_output.ML for common
  4184 examples.
  4185 
  4186 * More systematic treatment of long names, abstract name bindings, and
  4187 name space operations.  Basic operations on qualified names have been
  4188 move from structure NameSpace to Long_Name, e.g. Long_Name.base_name,
  4189 Long_Name.append.  Old type bstring has been mostly replaced by
  4190 abstract type binding (see structure Binding), which supports precise
  4191 qualification by packages and local theory targets, as well as proper
  4192 tracking of source positions.  INCOMPATIBILITY, need to wrap old
  4193 bstring values into Binding.name, or better pass through abstract
  4194 bindings everywhere.  See further src/Pure/General/long_name.ML,
  4195 src/Pure/General/binding.ML and src/Pure/General/name_space.ML
  4196 
  4197 * Result facts (from PureThy.note_thms, ProofContext.note_thms,
  4198 LocalTheory.note etc.) now refer to the *full* internal name, not the
  4199 bstring as before.  INCOMPATIBILITY, not detected by ML type-checking!
  4200 
  4201 * Disposed old type and term read functions (Sign.read_def_typ,
  4202 Sign.read_typ, Sign.read_def_terms, Sign.read_term,
  4203 Thm.read_def_cterms, Thm.read_cterm etc.).  INCOMPATIBILITY, should
  4204 use regular Syntax.read_typ, Syntax.read_term, Syntax.read_typ_global,
  4205 Syntax.read_term_global etc.; see also OldGoals.read_term as last
  4206 resort for legacy applications.
  4207 
  4208 * Disposed old declarations, tactics, tactic combinators that refer to
  4209 the simpset or claset of an implicit theory (such as Addsimps,
  4210 Simp_tac, SIMPSET).  INCOMPATIBILITY, should use @{simpset} etc. in
  4211 embedded ML text, or local_simpset_of with a proper context passed as
  4212 explicit runtime argument.
  4213 
  4214 * Rules and tactics that read instantiations (read_instantiate,
  4215 res_inst_tac, thin_tac, subgoal_tac etc.) now demand a proper proof
  4216 context, which is required for parsing and type-checking.  Moreover,
  4217 the variables are specified as plain indexnames, not string encodings
  4218 thereof.  INCOMPATIBILITY.
  4219 
  4220 * Generic Toplevel.add_hook interface allows to analyze the result of
  4221 transactions.  E.g. see src/Pure/ProofGeneral/proof_general_pgip.ML
  4222 for theorem dependency output of transactions resulting in a new
  4223 theory state.
  4224 
  4225 * ML antiquotations: block-structured compilation context indicated by
  4226 \<lbrace> ... \<rbrace>; additional antiquotation forms:
  4227 
  4228   @{binding name}                         - basic name binding
  4229   @{let ?pat = term}                      - term abbreviation (HO matching)
  4230   @{note name = fact}                     - fact abbreviation
  4231   @{thm fact}                             - singleton fact (with attributes)
  4232   @{thms fact}                            - general fact (with attributes)
  4233   @{lemma prop by method}                 - singleton goal
  4234   @{lemma prop by meth1 meth2}            - singleton goal
  4235   @{lemma prop1 ... propN by method}      - general goal
  4236   @{lemma prop1 ... propN by meth1 meth2} - general goal
  4237   @{lemma (open) ...}                     - open derivation
  4238 
  4239 
  4240 *** System ***
  4241 
  4242 * The Isabelle "emacs" tool provides a specific interface to invoke
  4243 Proof General / Emacs, with more explicit failure if that is not
  4244 installed (the old isabelle-interface script silently falls back on
  4245 isabelle-process).  The PROOFGENERAL_HOME setting determines the
  4246 installation location of the Proof General distribution.
  4247 
  4248 * Isabelle/lib/classes/Pure.jar provides basic support to integrate
  4249 the Isabelle process into a JVM/Scala application.  See
  4250 Isabelle/lib/jedit/plugin for a minimal example.  (The obsolete Java
  4251 process wrapper has been discontinued.)
  4252 
  4253 * Added homegrown Isabelle font with unicode layout, see lib/fonts.
  4254 
  4255 * Various status messages (with exact source position information) are
  4256 emitted, if proper markup print mode is enabled.  This allows
  4257 user-interface components to provide detailed feedback on internal
  4258 prover operations.
  4259 
  4260 
  4261 
  4262 New in Isabelle2008 (June 2008)
  4263 -------------------------------
  4264 
  4265 *** General ***
  4266 
  4267 * The Isabelle/Isar Reference Manual (isar-ref) has been reorganized
  4268 and updated, with formally checked references as hyperlinks.
  4269 
  4270 * Theory loader: use_thy (and similar operations) no longer set the
  4271 implicit ML context, which was occasionally hard to predict and in
  4272 conflict with concurrency.  INCOMPATIBILITY, use ML within Isar which
  4273 provides a proper context already.
  4274 
  4275 * Theory loader: old-style ML proof scripts being *attached* to a thy
  4276 file are no longer supported.  INCOMPATIBILITY, regular 'uses' and
  4277 'use' within a theory file will do the job.
  4278 
  4279 * Name space merge now observes canonical order, i.e. the second space
  4280 is inserted into the first one, while existing entries in the first
  4281 space take precedence.  INCOMPATIBILITY in rare situations, may try to
  4282 swap theory imports.
  4283 
  4284 * Syntax: symbol \<chi> is now considered a letter.  Potential
  4285 INCOMPATIBILITY in identifier syntax etc.
  4286 
  4287 * Outer syntax: string tokens no longer admit escaped white space,
  4288 which was an accidental (undocumented) feature.  INCOMPATIBILITY, use
  4289 white space without escapes.
  4290 
  4291 * Outer syntax: string tokens may contain arbitrary character codes
  4292 specified via 3 decimal digits (as in SML).  E.g. "foo\095bar" for
  4293 "foo_bar".
  4294 
  4295 
  4296 *** Pure ***
  4297 
  4298 * Context-dependent token translations.  Default setup reverts locally
  4299 fixed variables, and adds hilite markup for undeclared frees.
  4300 
  4301 * Unused theorems can be found using the new command 'unused_thms'.
  4302 There are three ways of invoking it:
  4303 
  4304 (1) unused_thms
  4305      Only finds unused theorems in the current theory.
  4306 
  4307 (2) unused_thms thy_1 ... thy_n -
  4308      Finds unused theorems in the current theory and all of its ancestors,
  4309      excluding the theories thy_1 ... thy_n and all of their ancestors.
  4310 
  4311 (3) unused_thms thy_1 ... thy_n - thy'_1 ... thy'_m
  4312      Finds unused theorems in the theories thy'_1 ... thy'_m and all of
  4313      their ancestors, excluding the theories thy_1 ... thy_n and all of
  4314      their ancestors.
  4315 
  4316 In order to increase the readability of the list produced by
  4317 unused_thms, theorems that have been created by a particular instance
  4318 of a theory command such as 'inductive' or 'function' are considered
  4319 to belong to the same "group", meaning that if at least one theorem in
  4320 this group is used, the other theorems in the same group are no longer
  4321 reported as unused.  Moreover, if all theorems in the group are
  4322 unused, only one theorem in the group is displayed.
  4323 
  4324 Note that proof objects have to be switched on in order for
  4325 unused_thms to work properly (i.e. !proofs must be >= 1, which is
  4326 usually the case when using Proof General with the default settings).
  4327 
  4328 * Authentic naming of facts disallows ad-hoc overwriting of previous
  4329 theorems within the same name space.  INCOMPATIBILITY, need to remove
  4330 duplicate fact bindings, or even accidental fact duplications.  Note
  4331 that tools may maintain dynamically scoped facts systematically, using
  4332 PureThy.add_thms_dynamic.
  4333 
  4334 * Command 'hide' now allows to hide from "fact" name space as well.
  4335 
  4336 * Eliminated destructive theorem database, simpset, claset, and
  4337 clasimpset.  Potential INCOMPATIBILITY, really need to observe linear
  4338 update of theories within ML code.
  4339 
  4340 * Eliminated theory ProtoPure and CPure, leaving just one Pure theory.
  4341 INCOMPATIBILITY, object-logics depending on former Pure require
  4342 additional setup PureThy.old_appl_syntax_setup; object-logics
  4343 depending on former CPure need to refer to Pure.
  4344 
  4345 * Commands 'use' and 'ML' are now purely functional, operating on
  4346 theory/local_theory.  Removed former 'ML_setup' (on theory), use 'ML'
  4347 instead.  Added 'ML_val' as mere diagnostic replacement for 'ML'.
  4348 INCOMPATIBILITY.
  4349 
  4350 * Command 'setup': discontinued implicit version with ML reference.
  4351 
  4352 * Instantiation target allows for simultaneous specification of class
  4353 instance operations together with an instantiation proof.
  4354 Type-checking phase allows to refer to class operations uniformly.
  4355 See src/HOL/Complex/Complex.thy for an Isar example and
  4356 src/HOL/Library/Eval.thy for an ML example.
  4357 
  4358 * Indexing of literal facts: be more serious about including only
  4359 facts from the visible specification/proof context, but not the
  4360 background context (locale etc.).  Affects `prop` notation and method
  4361 "fact".  INCOMPATIBILITY: need to name facts explicitly in rare
  4362 situations.
  4363 
  4364 * Method "cases", "induct", "coinduct": removed obsolete/undocumented
  4365 "(open)" option, which used to expose internal bound variables to the
  4366 proof text.
  4367 
  4368 * Isar statements: removed obsolete case "rule_context".
  4369 INCOMPATIBILITY, better use explicit fixes/assumes.
  4370 
  4371 * Locale proofs: default proof step now includes 'unfold_locales';
  4372 hence 'proof' without argument may be used to unfold locale
  4373 predicates.
  4374 
  4375 
  4376 *** Document preparation ***
  4377 
  4378 * Simplified pdfsetup.sty: color/hyperref is used unconditionally for
  4379 both pdf and dvi (hyperlinks usually work in xdvi as well); removed
  4380 obsolete thumbpdf setup (contemporary PDF viewers do this on the
  4381 spot); renamed link color from "darkblue" to "linkcolor" (default
  4382 value unchanged, can be redefined via \definecolor); no longer sets
  4383 "a4paper" option (unnecessary or even intrusive).
  4384 
  4385 * Antiquotation @{lemma A method} proves proposition A by the given
  4386 method (either a method name or a method name plus (optional) method
  4387 arguments in parentheses) and prints A just like @{prop A}.
  4388 
  4389 
  4390 *** HOL ***
  4391 
  4392 * New primrec package.  Specification syntax conforms in style to
  4393 definition/function/....  No separate induction rule is provided.  The
  4394 "primrec" command distinguishes old-style and new-style specifications
  4395 by syntax.  The former primrec package is now named OldPrimrecPackage.
  4396 When adjusting theories, beware: constants stemming from new-style
  4397 primrec specifications have authentic syntax.
  4398 
  4399 * Metis prover is now an order of magnitude faster, and also works
  4400 with multithreading.
  4401 
  4402 * Metis: the maximum number of clauses that can be produced from a
  4403 theorem is now given by the attribute max_clauses.  Theorems that
  4404 exceed this number are ignored, with a warning printed.
  4405 
  4406 * Sledgehammer no longer produces structured proofs by default. To
  4407 enable, declare [[sledgehammer_full = true]].  Attributes
  4408 reconstruction_modulus, reconstruction_sorts renamed
  4409 sledgehammer_modulus, sledgehammer_sorts.  INCOMPATIBILITY.
  4410 
  4411 * Method "induct_scheme" derives user-specified induction rules
  4412 from well-founded induction and completeness of patterns. This factors
  4413 out some operations that are done internally by the function package
  4414 and makes them available separately.  See
  4415 src/HOL/ex/Induction_Scheme.thy for examples.
  4416 
  4417 * More flexible generation of measure functions for termination
  4418 proofs: Measure functions can be declared by proving a rule of the
  4419 form "is_measure f" and giving it the [measure_function] attribute.
  4420 The "is_measure" predicate is logically meaningless (always true), and
  4421 just guides the heuristic.  To find suitable measure functions, the
  4422 termination prover sets up the goal "is_measure ?f" of the appropriate
  4423 type and generates all solutions by prolog-style backwards proof using
  4424 the declared rules.
  4425 
  4426 This setup also deals with rules like 
  4427 
  4428   "is_measure f ==> is_measure (list_size f)"
  4429 
  4430 which accommodates nested datatypes that recurse through lists.
  4431 Similar rules are predeclared for products and option types.
  4432 
  4433 * Turned the type of sets "'a set" into an abbreviation for "'a => bool"
  4434 
  4435   INCOMPATIBILITIES:
  4436 
  4437   - Definitions of overloaded constants on sets have to be replaced by
  4438     definitions on => and bool.
  4439 
  4440   - Some definitions of overloaded operators on sets can now be proved
  4441     using the definitions of the operators on => and bool.  Therefore,
  4442     the following theorems have been renamed:
  4443 
  4444       subset_def   -> subset_eq
  4445       psubset_def  -> psubset_eq
  4446       set_diff_def -> set_diff_eq
  4447       Compl_def    -> Compl_eq
  4448       Sup_set_def  -> Sup_set_eq
  4449       Inf_set_def  -> Inf_set_eq
  4450       sup_set_def  -> sup_set_eq
  4451       inf_set_def  -> inf_set_eq
  4452 
  4453   - Due to the incompleteness of the HO unification algorithm, some
  4454     rules such as subst may require manual instantiation, if some of
  4455     the unknowns in the rule is a set.
  4456 
  4457   - Higher order unification and forward proofs:
  4458     The proof pattern
  4459 
  4460       have "P (S::'a set)" <...>
  4461       then have "EX S. P S" ..
  4462 
  4463     no longer works (due to the incompleteness of the HO unification
  4464     algorithm) and must be replaced by the pattern
  4465 
  4466       have "EX S. P S"
  4467       proof
  4468         show "P S" <...>
  4469       qed
  4470 
  4471   - Calculational reasoning with subst (or similar rules):
  4472     The proof pattern
  4473 
  4474       have "P (S::'a set)" <...>
  4475       also have "S = T" <...>
  4476       finally have "P T" .
  4477 
  4478     no longer works (for similar reasons as the previous example) and
  4479     must be replaced by something like
  4480 
  4481       have "P (S::'a set)" <...>
  4482       moreover have "S = T" <...>
  4483       ultimately have "P T" by simp
  4484 
  4485   - Tactics or packages written in ML code:
  4486     Code performing pattern matching on types via
  4487 
  4488       Type ("set", [T]) => ...
  4489 
  4490     must be rewritten. Moreover, functions like strip_type or
  4491     binder_types no longer return the right value when applied to a
  4492     type of the form
  4493 
  4494       T1 => ... => Tn => U => bool
  4495 
  4496     rather than
  4497 
  4498       T1 => ... => Tn => U set
  4499 
  4500 * Merged theories Wellfounded_Recursion, Accessible_Part and
  4501 Wellfounded_Relations to theory Wellfounded.
  4502 
  4503 * Explicit class "eq" for executable equality.  INCOMPATIBILITY.
  4504 
  4505 * Class finite no longer treats UNIV as class parameter.  Use class
  4506 enum from theory Library/Enum instead to achieve a similar effect.
  4507 INCOMPATIBILITY.
  4508 
  4509 * Theory List: rule list_induct2 now has explicitly named cases "Nil"
  4510 and "Cons".  INCOMPATIBILITY.
  4511 
  4512 * HOL (and FOL): renamed variables in rules imp_elim and swap.
  4513 Potential INCOMPATIBILITY.
  4514 
  4515 * Theory Product_Type: duplicated lemmas split_Pair_apply and
  4516 injective_fst_snd removed, use split_eta and prod_eqI instead.
  4517 Renamed upd_fst to apfst and upd_snd to apsnd.  INCOMPATIBILITY.
  4518 
  4519 * Theory Nat: removed redundant lemmas that merely duplicate lemmas of
  4520 the same name in theory Orderings:
  4521 
  4522   less_trans
  4523   less_linear
  4524   le_imp_less_or_eq
  4525   le_less_trans
  4526   less_le_trans
  4527   less_not_sym
  4528   less_asym
  4529 
  4530 Renamed less_imp_le to less_imp_le_nat, and less_irrefl to
  4531 less_irrefl_nat.  Potential INCOMPATIBILITY due to more general types
  4532 and different variable names.
  4533 
  4534 * Library/Option_ord.thy: Canonical order on option type.
  4535 
  4536 * Library/RBT.thy: Red-black trees, an efficient implementation of
  4537 finite maps.
  4538 
  4539 * Library/Countable.thy: Type class for countable types.
  4540 
  4541 * Theory Int: The representation of numerals has changed.  The infix
  4542 operator BIT and the bit datatype with constructors B0 and B1 have
  4543 disappeared.  INCOMPATIBILITY, use "Int.Bit0 x" and "Int.Bit1 y" in
  4544 place of "x BIT bit.B0" and "y BIT bit.B1", respectively.  Theorems
  4545 involving BIT, B0, or B1 have been renamed with "Bit0" or "Bit1"
  4546 accordingly.
  4547 
  4548 * Theory Nat: definition of <= and < on natural numbers no longer
  4549 depend on well-founded relations.  INCOMPATIBILITY.  Definitions
  4550 le_def and less_def have disappeared.  Consider lemmas not_less
  4551 [symmetric, where ?'a = nat] and less_eq [symmetric] instead.
  4552 
  4553 * Theory Finite_Set: locales ACf, ACe, ACIf, ACIfSL and ACIfSLlin
  4554 (whose purpose mainly is for various fold_set functionals) have been
  4555 abandoned in favor of the existing algebraic classes
  4556 ab_semigroup_mult, comm_monoid_mult, ab_semigroup_idem_mult,
  4557 lower_semilattice (resp. upper_semilattice) and linorder.
  4558 INCOMPATIBILITY.
  4559 
  4560 * Theory Transitive_Closure: induct and cases rules now declare proper
  4561 case_names ("base" and "step").  INCOMPATIBILITY.
  4562 
  4563 * Theorem Inductive.lfp_ordinal_induct generalized to complete
  4564 lattices.  The form set-specific version is available as
  4565 Inductive.lfp_ordinal_induct_set.
  4566 
  4567 * Renamed theorems "power.simps" to "power_int.simps".
  4568 INCOMPATIBILITY.
  4569 
  4570 * Class semiring_div provides basic abstract properties of semirings
  4571 with division and modulo operations.  Subsumes former class dvd_mod.
  4572 
  4573 * Merged theories IntDef, Numeral and IntArith into unified theory
  4574 Int.  INCOMPATIBILITY.
  4575 
  4576 * Theory Library/Code_Index: type "index" now represents natural
  4577 numbers rather than integers.  INCOMPATIBILITY.
  4578 
  4579 * New class "uminus" with operation "uminus" (split of from class
  4580 "minus" which now only has operation "minus", binary).
  4581 INCOMPATIBILITY.
  4582 
  4583 * Constants "card", "internal_split", "option_map" now with authentic
  4584 syntax.  INCOMPATIBILITY.
  4585 
  4586 * Definitions subset_def, psubset_def, set_diff_def, Compl_def,
  4587 le_bool_def, less_bool_def, le_fun_def, less_fun_def, inf_bool_def,
  4588 sup_bool_def, Inf_bool_def, Sup_bool_def, inf_fun_def, sup_fun_def,
  4589 Inf_fun_def, Sup_fun_def, inf_set_def, sup_set_def, Inf_set_def,
  4590 Sup_set_def, le_def, less_def, option_map_def now with object
  4591 equality.  INCOMPATIBILITY.
  4592 
  4593 * Records. Removed K_record, and replaced it by pure lambda term
  4594 %x. c. The simplifier setup is now more robust against eta expansion.
  4595 INCOMPATIBILITY: in cases explicitly referring to K_record.
  4596 
  4597 * Library/Multiset: {#a, b, c#} abbreviates {#a#} + {#b#} + {#c#}.
  4598 
  4599 * Library/ListVector: new theory of arithmetic vector operations.
  4600 
  4601 * Library/Order_Relation: new theory of various orderings as sets of
  4602 pairs.  Defines preorders, partial orders, linear orders and
  4603 well-orders on sets and on types.
  4604 
  4605 
  4606 *** ZF ***
  4607 
  4608 * Renamed some theories to allow to loading both ZF and HOL in the
  4609 same session:
  4610 
  4611   Datatype  -> Datatype_ZF
  4612   Inductive -> Inductive_ZF
  4613   Int       -> Int_ZF
  4614   IntDiv    -> IntDiv_ZF
  4615   Nat       -> Nat_ZF
  4616   List      -> List_ZF
  4617   Main      -> Main_ZF
  4618 
  4619 INCOMPATIBILITY: ZF theories that import individual theories below
  4620 Main might need to be adapted.  Regular theory Main is still
  4621 available, as trivial extension of Main_ZF.
  4622 
  4623 
  4624 *** ML ***
  4625 
  4626 * ML within Isar: antiquotation @{const name} or @{const
  4627 name(typargs)} produces statically-checked Const term.
  4628 
  4629 * Functor NamedThmsFun: data is available to the user as dynamic fact
  4630 (of the same name).  Removed obsolete print command.
  4631 
  4632 * Removed obsolete "use_legacy_bindings" function.
  4633 
  4634 * The ``print mode'' is now a thread-local value derived from a global
  4635 template (the former print_mode reference), thus access becomes
  4636 non-critical.  The global print_mode reference is for session
  4637 management only; user-code should use print_mode_value,
  4638 print_mode_active, PrintMode.setmp etc.  INCOMPATIBILITY.
  4639 
  4640 * Functions system/system_out provide a robust way to invoke external
  4641 shell commands, with propagation of interrupts (requires Poly/ML
  4642 5.2.1).  Do not use OS.Process.system etc. from the basis library!
  4643 
  4644 
  4645 *** System ***
  4646 
  4647 * Default settings: PROOFGENERAL_OPTIONS no longer impose xemacs ---
  4648 in accordance with Proof General 3.7, which prefers GNU emacs.
  4649 
  4650 * isatool tty runs Isabelle process with plain tty interaction;
  4651 optional line editor may be specified via ISABELLE_LINE_EDITOR
  4652 setting, the default settings attempt to locate "ledit" and "rlwrap".
  4653 
  4654 * isatool browser now works with Cygwin as well, using general
  4655 "javapath" function defined in Isabelle process environment.
  4656 
  4657 * YXML notation provides a simple and efficient alternative to
  4658 standard XML transfer syntax.  See src/Pure/General/yxml.ML and
  4659 isatool yxml as described in the Isabelle system manual.
  4660 
  4661 * JVM class isabelle.IsabelleProcess (located in Isabelle/lib/classes)
  4662 provides general wrapper for managing an Isabelle process in a robust
  4663 fashion, with ``cooked'' output from stdin/stderr.
  4664 
  4665 * Rudimentary Isabelle plugin for jEdit (see Isabelle/lib/jedit),
  4666 based on Isabelle/JVM process wrapper (see Isabelle/lib/classes).
  4667 
  4668 * Removed obsolete THIS_IS_ISABELLE_BUILD feature.  NB: the documented
  4669 way of changing the user's settings is via
  4670 ISABELLE_HOME_USER/etc/settings, which is a fully featured bash
  4671 script.
  4672 
  4673 * Multithreading.max_threads := 0 refers to the number of actual CPU
  4674 cores of the underlying machine, which is a good starting point for
  4675 optimal performance tuning.  The corresponding usedir option -M allows
  4676 "max" as an alias for "0".  WARNING: does not work on certain versions
  4677 of Mac OS (with Poly/ML 5.1).
  4678 
  4679 * isabelle-process: non-ML sessions are run with "nice", to reduce the
  4680 adverse effect of Isabelle flooding interactive front-ends (notably
  4681 ProofGeneral / XEmacs).
  4682 
  4683 
  4684 
  4685 New in Isabelle2007 (November 2007)
  4686 -----------------------------------
  4687 
  4688 *** General ***
  4689 
  4690 * More uniform information about legacy features, notably a
  4691 warning/error of "Legacy feature: ...", depending on the state of the
  4692 tolerate_legacy_features flag (default true). FUTURE INCOMPATIBILITY:
  4693 legacy features will disappear eventually.
  4694 
  4695 * Theory syntax: the header format ``theory A = B + C:'' has been
  4696 discontinued in favour of ``theory A imports B C begin''.  Use isatool
  4697 fixheaders to convert existing theory files.  INCOMPATIBILITY.
  4698 
  4699 * Theory syntax: the old non-Isar theory file format has been
  4700 discontinued altogether.  Note that ML proof scripts may still be used
  4701 with Isar theories; migration is usually quite simple with the ML
  4702 function use_legacy_bindings.  INCOMPATIBILITY.
  4703 
  4704 * Theory syntax: some popular names (e.g. 'class', 'declaration',
  4705 'fun', 'help', 'if') are now keywords.  INCOMPATIBILITY, use double
  4706 quotes.
  4707 
  4708 * Theory loader: be more serious about observing the static theory
  4709 header specifications (including optional directories), but not the
  4710 accidental file locations of previously successful loads.  The strict
  4711 update policy of former update_thy is now already performed by
  4712 use_thy, so the former has been removed; use_thys updates several
  4713 theories simultaneously, just as 'imports' within a theory header
  4714 specification, but without merging the results.  Potential
  4715 INCOMPATIBILITY: may need to refine theory headers and commands
  4716 ROOT.ML which depend on load order.
  4717 
  4718 * Theory loader: optional support for content-based file
  4719 identification, instead of the traditional scheme of full physical
  4720 path plus date stamp; configured by the ISABELLE_FILE_IDENT setting
  4721 (cf. the system manual).  The new scheme allows to work with
  4722 non-finished theories in persistent session images, such that source
  4723 files may be moved later on without requiring reloads.
  4724 
  4725 * Theory loader: old-style ML proof scripts being *attached* to a thy
  4726 file (with the same base name as the theory) are considered a legacy
  4727 feature, which will disappear eventually. Even now, the theory loader
  4728 no longer maintains dependencies on such files.
  4729 
  4730 * Syntax: the scope for resolving ambiguities via type-inference is
  4731 now limited to individual terms, instead of whole simultaneous
  4732 specifications as before. This greatly reduces the complexity of the
  4733 syntax module and improves flexibility by separating parsing and
  4734 type-checking. INCOMPATIBILITY: additional type-constraints (explicit
  4735 'fixes' etc.) are required in rare situations.
  4736 
  4737 * Syntax: constants introduced by new-style packages ('definition',
  4738 'abbreviation' etc.) are passed through the syntax module in
  4739 ``authentic mode''. This means that associated mixfix annotations
  4740 really stick to such constants, independently of potential name space
  4741 ambiguities introduced later on. INCOMPATIBILITY: constants in parse
  4742 trees are represented slightly differently, may need to adapt syntax
  4743 translations accordingly. Use CONST marker in 'translations' and
  4744 @{const_syntax} antiquotation in 'parse_translation' etc.
  4745 
  4746 * Legacy goal package: reduced interface to the bare minimum required
  4747 to keep existing proof scripts running.  Most other user-level
  4748 functions are now part of the OldGoals structure, which is *not* open
  4749 by default (consider isatool expandshort before open OldGoals).
  4750 Removed top_sg, prin, printyp, pprint_term/typ altogether, because
  4751 these tend to cause confusion about the actual goal (!) context being
  4752 used here, which is not necessarily the same as the_context().
  4753 
  4754 * Command 'find_theorems': supports "*" wild-card in "name:"
  4755 criterion; "with_dups" option.  Certain ProofGeneral versions might
  4756 support a specific search form (see ProofGeneral/CHANGES).
  4757 
  4758 * The ``prems limit'' option (cf. ProofContext.prems_limit) is now -1
  4759 by default, which means that "prems" (and also "fixed variables") are
  4760 suppressed from proof state output.  Note that the ProofGeneral
  4761 settings mechanism allows to change and save options persistently, but
  4762 older versions of Isabelle will fail to start up if a negative prems
  4763 limit is imposed.
  4764 
  4765 * Local theory targets may be specified by non-nested blocks of
  4766 ``context/locale/class ... begin'' followed by ``end''.  The body may
  4767 contain definitions, theorems etc., including any derived mechanism
  4768 that has been implemented on top of these primitives.  This concept
  4769 generalizes the existing ``theorem (in ...)'' towards more versatility
  4770 and scalability.
  4771 
  4772 * Proof General interface: proper undo of final 'end' command;
  4773 discontinued Isabelle/classic mode (ML proof scripts).
  4774 
  4775 
  4776 *** Document preparation ***
  4777 
  4778 * Added antiquotation @{theory name} which prints the given name,
  4779 after checking that it refers to a valid ancestor theory in the
  4780 current context.
  4781 
  4782 * Added antiquotations @{ML_type text} and @{ML_struct text} which
  4783 check the given source text as ML type/structure, printing verbatim.
  4784 
  4785 * Added antiquotation @{abbrev "c args"} which prints the abbreviation
  4786 "c args == rhs" given in the current context.  (Any number of
  4787 arguments may be given on the LHS.)
  4788 
  4789 
  4790 *** Pure ***
  4791 
  4792 * The 'class' package offers a combination of axclass and locale to
  4793 achieve Haskell-like type classes in Isabelle.  Definitions and
  4794 theorems within a class context produce both relative results (with
  4795 implicit parameters according to the locale context), and polymorphic
  4796 constants with qualified polymorphism (according to the class
  4797 context).  Within the body context of a 'class' target, a separate
  4798 syntax layer ("user space type system") takes care of converting
  4799 between global polymorphic consts and internal locale representation.
  4800 See src/HOL/ex/Classpackage.thy for examples (as well as main HOL).
  4801 "isatool doc classes" provides a tutorial.
  4802 
  4803 * Generic code generator framework allows to generate executable
  4804 code for ML and Haskell (including Isabelle classes).  A short usage
  4805 sketch:
  4806 
  4807     internal compilation:
  4808         export_code <list of constants (term syntax)> in SML
  4809     writing SML code to a file:
  4810         export_code <list of constants (term syntax)> in SML <filename>
  4811     writing OCaml code to a file:
  4812         export_code <list of constants (term syntax)> in OCaml <filename>
  4813     writing Haskell code to a bunch of files:
  4814         export_code <list of constants (term syntax)> in Haskell <filename>
  4815 
  4816     evaluating closed propositions to True/False using code generation:
  4817         method ``eval''
  4818 
  4819 Reasonable default setup of framework in HOL.
  4820 
  4821 Theorem attributs for selecting and transforming function equations theorems:
  4822 
  4823     [code fun]:        select a theorem as function equation for a specific constant
  4824     [code fun del]:    deselect a theorem as function equation for a specific constant
  4825     [code inline]:     select an equation theorem for unfolding (inlining) in place
  4826     [code inline del]: deselect an equation theorem for unfolding (inlining) in place
  4827 
  4828 User-defined serializations (target in {SML, OCaml, Haskell}):
  4829 
  4830     code_const <and-list of constants (term syntax)>
  4831       {(target) <and-list of const target syntax>}+
  4832 
  4833     code_type <and-list of type constructors>
  4834       {(target) <and-list of type target syntax>}+
  4835 
  4836     code_instance <and-list of instances>
  4837       {(target)}+
  4838         where instance ::= <type constructor> :: <class>
  4839 
  4840     code_class <and_list of classes>
  4841       {(target) <and-list of class target syntax>}+
  4842         where class target syntax ::= <class name> {where {<classop> == <target syntax>}+}?
  4843 
  4844 code_instance and code_class only are effective to target Haskell.
  4845 
  4846 For example usage see src/HOL/ex/Codegenerator.thy and
  4847 src/HOL/ex/Codegenerator_Pretty.thy.  A separate tutorial on code
  4848 generation from Isabelle/HOL theories is available via "isatool doc
  4849 codegen".
  4850 
  4851 * Code generator: consts in 'consts_code' Isar commands are now
  4852 referred to by usual term syntax (including optional type
  4853 annotations).
  4854 
  4855 * Command 'no_translations' removes translation rules from theory
  4856 syntax.
  4857 
  4858 * Overloaded definitions are now actually checked for acyclic
  4859 dependencies.  The overloading scheme is slightly more general than
  4860 that of Haskell98, although Isabelle does not demand an exact
  4861 correspondence to type class and instance declarations.
  4862 INCOMPATIBILITY, use ``defs (unchecked overloaded)'' to admit more
  4863 exotic versions of overloading -- at the discretion of the user!
  4864 
  4865 Polymorphic constants are represented via type arguments, i.e. the
  4866 instantiation that matches an instance against the most general
  4867 declaration given in the signature.  For example, with the declaration
  4868 c :: 'a => 'a => 'a, an instance c :: nat => nat => nat is represented
  4869 as c(nat).  Overloading is essentially simultaneous structural
  4870 recursion over such type arguments.  Incomplete specification patterns
  4871 impose global constraints on all occurrences, e.g. c('a * 'a) on the
  4872 LHS means that more general c('a * 'b) will be disallowed on any RHS.
  4873 Command 'print_theory' outputs the normalized system of recursive
  4874 equations, see section "definitions".
  4875 
  4876 * Configuration options are maintained within the theory or proof
  4877 context (with name and type bool/int/string), providing a very simple
  4878 interface to a poor-man's version of general context data.  Tools may
  4879 declare options in ML (e.g. using Attrib.config_int) and then refer to
  4880 these values using Config.get etc.  Users may change options via an
  4881 associated attribute of the same name.  This form of context
  4882 declaration works particularly well with commands 'declare' or
  4883 'using', for example ``declare [[foo = 42]]''.  Thus it has become
  4884 very easy to avoid global references, which would not observe Isar
  4885 toplevel undo/redo and fail to work with multithreading.
  4886 
  4887 Various global ML references of Pure and HOL have been turned into
  4888 configuration options:
  4889 
  4890   Unify.search_bound		unify_search_bound
  4891   Unify.trace_bound		unify_trace_bound
  4892   Unify.trace_simp		unify_trace_simp
  4893   Unify.trace_types		unify_trace_types
  4894   Simplifier.simp_depth_limit	simp_depth_limit
  4895   Blast.depth_limit		blast_depth_limit
  4896   DatatypeProp.dtK		datatype_distinctness_limit
  4897   fast_arith_neq_limit  	fast_arith_neq_limit
  4898   fast_arith_split_limit	fast_arith_split_limit
  4899 
  4900 * Named collections of theorems may be easily installed as context
  4901 data using the functor NamedThmsFun (see also
  4902 src/Pure/Tools/named_thms.ML).  The user may add or delete facts via
  4903 attributes; there is also a toplevel print command.  This facility is
  4904 just a common case of general context data, which is the preferred way
  4905 for anything more complex than just a list of facts in canonical
  4906 order.
  4907 
  4908 * Isar: command 'declaration' augments a local theory by generic
  4909 declaration functions written in ML.  This enables arbitrary content
  4910 being added to the context, depending on a morphism that tells the
  4911 difference of the original declaration context wrt. the application
  4912 context encountered later on.
  4913 
  4914 * Isar: proper interfaces for simplification procedures.  Command
  4915 'simproc_setup' declares named simprocs (with match patterns, and body
  4916 text in ML).  Attribute "simproc" adds/deletes simprocs in the current
  4917 context.  ML antiquotation @{simproc name} retrieves named simprocs.
  4918 
  4919 * Isar: an extra pair of brackets around attribute declarations
  4920 abbreviates a theorem reference involving an internal dummy fact,
  4921 which will be ignored later --- only the effect of the attribute on
  4922 the background context will persist.  This form of in-place
  4923 declarations is particularly useful with commands like 'declare' and
  4924 'using', for example ``have A using [[simproc a]] by simp''.
  4925 
  4926 * Isar: method "assumption" (and implicit closing of subproofs) now
  4927 takes simple non-atomic goal assumptions into account: after applying
  4928 an assumption as a rule the resulting subgoals are solved by atomic
  4929 assumption steps.  This is particularly useful to finish 'obtain'
  4930 goals, such as "!!x. (!!x. P x ==> thesis) ==> P x ==> thesis",
  4931 without referring to the original premise "!!x. P x ==> thesis" in the
  4932 Isar proof context.  POTENTIAL INCOMPATIBILITY: method "assumption" is
  4933 more permissive.
  4934 
  4935 * Isar: implicit use of prems from the Isar proof context is
  4936 considered a legacy feature.  Common applications like ``have A .''
  4937 may be replaced by ``have A by fact'' or ``note `A`''.  In general,
  4938 referencing facts explicitly here improves readability and
  4939 maintainability of proof texts.
  4940 
  4941 * Isar: improper proof element 'guess' is like 'obtain', but derives
  4942 the obtained context from the course of reasoning!  For example:
  4943 
  4944   assume "EX x y. A x & B y"   -- "any previous fact"
  4945   then guess x and y by clarify
  4946 
  4947 This technique is potentially adventurous, depending on the facts and
  4948 proof tools being involved here.
  4949 
  4950 * Isar: known facts from the proof context may be specified as literal
  4951 propositions, using ASCII back-quote syntax.  This works wherever
  4952 named facts used to be allowed so far, in proof commands, proof
  4953 methods, attributes etc.  Literal facts are retrieved from the context
  4954 according to unification of type and term parameters.  For example,
  4955 provided that "A" and "A ==> B" and "!!x. P x ==> Q x" are known
  4956 theorems in the current context, then these are valid literal facts:
  4957 `A` and `A ==> B` and `!!x. P x ==> Q x" as well as `P a ==> Q a` etc.
  4958 
  4959 There is also a proof method "fact" which does the same composition
  4960 for explicit goal states, e.g. the following proof texts coincide with
  4961 certain special cases of literal facts:
  4962 
  4963   have "A" by fact                 ==  note `A`
  4964   have "A ==> B" by fact           ==  note `A ==> B`
  4965   have "!!x. P x ==> Q x" by fact  ==  note `!!x. P x ==> Q x`
  4966   have "P a ==> Q a" by fact       ==  note `P a ==> Q a`
  4967 
  4968 * Isar: ":" (colon) is no longer a symbolic identifier character in
  4969 outer syntax.  Thus symbolic identifiers may be used without
  4970 additional white space in declarations like this: ``assume *: A''.
  4971 
  4972 * Isar: 'print_facts' prints all local facts of the current context,
  4973 both named and unnamed ones.
  4974 
  4975 * Isar: 'def' now admits simultaneous definitions, e.g.:
  4976 
  4977   def x == "t" and y == "u"
  4978 
  4979 * Isar: added command 'unfolding', which is structurally similar to
  4980 'using', but affects both the goal state and facts by unfolding given
  4981 rewrite rules.  Thus many occurrences of the 'unfold' method or
  4982 'unfolded' attribute may be replaced by first-class proof text.
  4983 
  4984 * Isar: methods 'unfold' / 'fold', attributes 'unfolded' / 'folded',
  4985 and command 'unfolding' now all support object-level equalities
  4986 (potentially conditional).  The underlying notion of rewrite rule is
  4987 analogous to the 'rule_format' attribute, but *not* that of the
  4988 Simplifier (which is usually more generous).
  4989 
  4990 * Isar: the new attribute [rotated n] (default n = 1) rotates the
  4991 premises of a theorem by n. Useful in conjunction with drule.
  4992 
  4993 * Isar: the goal restriction operator [N] (default N = 1) evaluates a
  4994 method expression within a sandbox consisting of the first N
  4995 sub-goals, which need to exist.  For example, ``simp_all [3]''
  4996 simplifies the first three sub-goals, while (rule foo, simp_all)[]
  4997 simplifies all new goals that emerge from applying rule foo to the
  4998 originally first one.
  4999 
  5000 * Isar: schematic goals are no longer restricted to higher-order
  5001 patterns; e.g. ``lemma "?P(?x)" by (rule TrueI)'' now works as
  5002 expected.
  5003 
  5004 * Isar: the conclusion of a long theorem statement is now either
  5005 'shows' (a simultaneous conjunction, as before), or 'obtains'
  5006 (essentially a disjunction of cases with local parameters and
  5007 assumptions).  The latter allows to express general elimination rules
  5008 adequately; in this notation common elimination rules look like this:
  5009 
  5010   lemma exE:    -- "EX x. P x ==> (!!x. P x ==> thesis) ==> thesis"
  5011     assumes "EX x. P x"
  5012     obtains x where "P x"
  5013 
  5014   lemma conjE:  -- "A & B ==> (A ==> B ==> thesis) ==> thesis"
  5015     assumes "A & B"
  5016     obtains A and B
  5017 
  5018   lemma disjE:  -- "A | B ==> (A ==> thesis) ==> (B ==> thesis) ==> thesis"
  5019     assumes "A | B"
  5020     obtains
  5021       A
  5022     | B
  5023 
  5024 The subsequent classical rules even refer to the formal "thesis"
  5025 explicitly:
  5026 
  5027   lemma classical:     -- "(~ thesis ==> thesis) ==> thesis"
  5028     obtains "~ thesis"
  5029 
  5030   lemma Peirce's_Law:  -- "((thesis ==> something) ==> thesis) ==> thesis"
  5031     obtains "thesis ==> something"
  5032 
  5033 The actual proof of an 'obtains' statement is analogous to that of the
  5034 Isar proof element 'obtain', only that there may be several cases.
  5035 Optional case names may be specified in parentheses; these will be
  5036 available both in the present proof and as annotations in the
  5037 resulting rule, for later use with the 'cases' method (cf. attribute
  5038 case_names).
  5039 
  5040 * Isar: the assumptions of a long theorem statement are available as
  5041 "assms" fact in the proof context.  This is more appropriate than the
  5042 (historical) "prems", which refers to all assumptions of the current
  5043 context, including those from the target locale, proof body etc.
  5044 
  5045 * Isar: 'print_statement' prints theorems from the current theory or
  5046 proof context in long statement form, according to the syntax of a
  5047 top-level lemma.
  5048 
  5049 * Isar: 'obtain' takes an optional case name for the local context
  5050 introduction rule (default "that").
  5051 
  5052 * Isar: removed obsolete 'concl is' patterns.  INCOMPATIBILITY, use
  5053 explicit (is "_ ==> ?foo") in the rare cases where this still happens
  5054 to occur.
  5055 
  5056 * Pure: syntax "CONST name" produces a fully internalized constant
  5057 according to the current context.  This is particularly useful for
  5058 syntax translations that should refer to internal constant
  5059 representations independently of name spaces.
  5060 
  5061 * Pure: syntax constant for foo (binder "FOO ") is called "foo_binder"
  5062 instead of "FOO ". This allows multiple binder declarations to coexist
  5063 in the same context.  INCOMPATIBILITY.
  5064 
  5065 * Isar/locales: 'notation' provides a robust interface to the 'syntax'
  5066 primitive that also works in a locale context (both for constants and
  5067 fixed variables). Type declaration and internal syntactic representation
  5068 of given constants retrieved from the context. Likewise, the
  5069 'no_notation' command allows to remove given syntax annotations from the
  5070 current context.
  5071 
  5072 * Isar/locales: new derived specification elements 'axiomatization',
  5073 'definition', 'abbreviation', which support type-inference, admit
  5074 object-level specifications (equality, equivalence).  See also the
  5075 isar-ref manual.  Examples:
  5076 
  5077   axiomatization
  5078     eq  (infix "===" 50) where
  5079     eq_refl: "x === x" and eq_subst: "x === y ==> P x ==> P y"
  5080 
  5081   definition "f x y = x + y + 1"
  5082   definition g where "g x = f x x"
  5083 
  5084   abbreviation
  5085     neq  (infix "=!=" 50) where
  5086     "x =!= y == ~ (x === y)"
  5087 
  5088 These specifications may be also used in a locale context.  Then the
  5089 constants being introduced depend on certain fixed parameters, and the
  5090 constant name is qualified by the locale base name.  An internal
  5091 abbreviation takes care for convenient input and output, making the
  5092 parameters implicit and using the original short name.  See also
  5093 src/HOL/ex/Abstract_NAT.thy for an example of deriving polymorphic
  5094 entities from a monomorphic theory.
  5095 
  5096 Presently, abbreviations are only available 'in' a target locale, but
  5097 not inherited by general import expressions.  Also note that
  5098 'abbreviation' may be used as a type-safe replacement for 'syntax' +
  5099 'translations' in common applications.  The "no_abbrevs" print mode
  5100 prevents folding of abbreviations in term output.
  5101 
  5102 Concrete syntax is attached to specified constants in internal form,
  5103 independently of name spaces.  The parse tree representation is
  5104 slightly different -- use 'notation' instead of raw 'syntax', and
  5105 'translations' with explicit "CONST" markup to accommodate this.
  5106 
  5107 * Pure/Isar: unified syntax for new-style specification mechanisms
  5108 (e.g.  'definition', 'abbreviation', or 'inductive' in HOL) admits
  5109 full type inference and dummy patterns ("_").  For example:
  5110 
  5111   definition "K x _ = x"
  5112 
  5113   inductive conj for A B
  5114   where "A ==> B ==> conj A B"
  5115 
  5116 * Pure: command 'print_abbrevs' prints all constant abbreviations of
  5117 the current context.  Print mode "no_abbrevs" prevents inversion of
  5118 abbreviations on output.
  5119 
  5120 * Isar/locales: improved parameter handling: use of locales "var" and
  5121 "struct" no longer necessary; - parameter renamings are no longer
  5122 required to be injective.  For example, this allows to define
  5123 endomorphisms as locale endom = homom mult mult h.
  5124 
  5125 * Isar/locales: changed the way locales with predicates are defined.
  5126 Instead of accumulating the specification, the imported expression is
  5127 now an interpretation.  INCOMPATIBILITY: different normal form of
  5128 locale expressions.  In particular, in interpretations of locales with
  5129 predicates, goals repesenting already interpreted fragments are not
  5130 removed automatically.  Use methods `intro_locales' and
  5131 `unfold_locales'; see below.
  5132 
  5133 * Isar/locales: new methods `intro_locales' and `unfold_locales'
  5134 provide backward reasoning on locales predicates.  The methods are
  5135 aware of interpretations and discharge corresponding goals.
  5136 `intro_locales' is less aggressive then `unfold_locales' and does not
  5137 unfold predicates to assumptions.
  5138 
  5139 * Isar/locales: the order in which locale fragments are accumulated
  5140 has changed.  This enables to override declarations from fragments due
  5141 to interpretations -- for example, unwanted simp rules.
  5142 
  5143 * Isar/locales: interpretation in theories and proof contexts has been
  5144 extended.  One may now specify (and prove) equations, which are
  5145 unfolded in interpreted theorems.  This is useful for replacing
  5146 defined concepts (constants depending on locale parameters) by
  5147 concepts already existing in the target context.  Example:
  5148 
  5149   interpretation partial_order ["op <= :: [int, int] => bool"]
  5150     where "partial_order.less (op <=) (x::int) y = (x < y)"
  5151 
  5152 Typically, the constant `partial_order.less' is created by a
  5153 definition specification element in the context of locale
  5154 partial_order.
  5155 
  5156 * Method "induct": improved internal context management to support
  5157 local fixes and defines on-the-fly. Thus explicit meta-level
  5158 connectives !!  and ==> are rarely required anymore in inductive goals
  5159 (using object-logic connectives for this purpose has been long
  5160 obsolete anyway). Common proof patterns are explained in
  5161 src/HOL/Induct/Common_Patterns.thy, see also
  5162 src/HOL/Isar_examples/Puzzle.thy and src/HOL/Lambda for realistic
  5163 examples.
  5164 
  5165 * Method "induct": improved handling of simultaneous goals. Instead of
  5166 introducing object-level conjunction, the statement is now split into
  5167 several conclusions, while the corresponding symbolic cases are nested
  5168 accordingly. INCOMPATIBILITY, proofs need to be structured explicitly,
  5169 see src/HOL/Induct/Common_Patterns.thy, for example.
  5170 
  5171 * Method "induct": mutual induction rules are now specified as a list
  5172 of rule sharing the same induction cases. HOL packages usually provide
  5173 foo_bar.inducts for mutually defined items foo and bar (e.g. inductive
  5174 predicates/sets or datatypes). INCOMPATIBILITY, users need to specify
  5175 mutual induction rules differently, i.e. like this:
  5176 
  5177   (induct rule: foo_bar.inducts)
  5178   (induct set: foo bar)
  5179   (induct pred: foo bar)
  5180   (induct type: foo bar)
  5181 
  5182 The ML function ProjectRule.projections turns old-style rules into the
  5183 new format.
  5184 
  5185 * Method "coinduct": dual of induction, see
  5186 src/HOL/Library/Coinductive_List.thy for various examples.
  5187 
  5188 * Method "cases", "induct", "coinduct": the ``(open)'' option is
  5189 considered a legacy feature.
  5190 
  5191 * Attribute "symmetric" produces result with standardized schematic
  5192 variables (index 0).  Potential INCOMPATIBILITY.
  5193 
  5194 * Simplifier: by default the simplifier trace only shows top level
  5195 rewrites now. That is, trace_simp_depth_limit is set to 1 by
  5196 default. Thus there is less danger of being flooded by the trace. The
  5197 trace indicates where parts have been suppressed.
  5198   
  5199 * Provers/classical: removed obsolete classical version of elim_format
  5200 attribute; classical elim/dest rules are now treated uniformly when
  5201 manipulating the claset.
  5202 
  5203 * Provers/classical: stricter checks to ensure that supplied intro,
  5204 dest and elim rules are well-formed; dest and elim rules must have at
  5205 least one premise.
  5206 
  5207 * Provers/classical: attributes dest/elim/intro take an optional
  5208 weight argument for the rule (just as the Pure versions).  Weights are
  5209 ignored by automated tools, but determine the search order of single
  5210 rule steps.
  5211 
  5212 * Syntax: input syntax now supports dummy variable binding "%_. b",
  5213 where the body does not mention the bound variable.  Note that dummy
  5214 patterns implicitly depend on their context of bounds, which makes
  5215 "{_. _}" match any set comprehension as expected.  Potential
  5216 INCOMPATIBILITY -- parse translations need to cope with syntactic
  5217 constant "_idtdummy" in the binding position.
  5218 
  5219 * Syntax: removed obsolete syntactic constant "_K" and its associated
  5220 parse translation.  INCOMPATIBILITY -- use dummy abstraction instead,
  5221 for example "A -> B" => "Pi A (%_. B)".
  5222 
  5223 * Pure: 'class_deps' command visualizes the subclass relation, using
  5224 the graph browser tool.
  5225 
  5226 * Pure: 'print_theory' now suppresses certain internal declarations by
  5227 default; use '!' option for full details.
  5228 
  5229 
  5230 *** HOL ***
  5231 
  5232 * Method "metis" proves goals by applying the Metis general-purpose
  5233 resolution prover (see also http://gilith.com/software/metis/).
  5234 Examples are in the directory MetisExamples.  WARNING: the
  5235 Isabelle/HOL-Metis integration does not yet work properly with
  5236 multi-threading.
  5237   
  5238 * Command 'sledgehammer' invokes external automatic theorem provers as
  5239 background processes.  It generates calls to the "metis" method if
  5240 successful. These can be pasted into the proof.  Users do not have to
  5241 wait for the automatic provers to return.  WARNING: does not really
  5242 work with multi-threading.
  5243 
  5244 * New "auto_quickcheck" feature tests outermost goal statements for
  5245 potential counter-examples.  Controlled by ML references
  5246 auto_quickcheck (default true) and auto_quickcheck_time_limit (default
  5247 5000 milliseconds).  Fails silently if statements is outside of
  5248 executable fragment, or any other codgenerator problem occurs.
  5249 
  5250 * New constant "undefined" with axiom "undefined x = undefined".
  5251 
  5252 * Added class "HOL.eq", allowing for code generation with polymorphic
  5253 equality.
  5254 
  5255 * Some renaming of class constants due to canonical name prefixing in
  5256 the new 'class' package:
  5257 
  5258     HOL.abs ~> HOL.abs_class.abs
  5259     HOL.divide ~> HOL.divide_class.divide
  5260     0 ~> HOL.zero_class.zero
  5261     1 ~> HOL.one_class.one
  5262     op + ~> HOL.plus_class.plus
  5263     op - ~> HOL.minus_class.minus
  5264     uminus ~> HOL.minus_class.uminus
  5265     op * ~> HOL.times_class.times
  5266     op < ~> HOL.ord_class.less
  5267     op <= > HOL.ord_class.less_eq
  5268     Nat.power ~> Power.power_class.power
  5269     Nat.size ~> Nat.size_class.size
  5270     Numeral.number_of ~> Numeral.number_class.number_of
  5271     FixedPoint.Inf ~> Lattices.complete_lattice_class.Inf
  5272     FixedPoint.Sup ~> Lattices.complete_lattice_class.Sup
  5273     Orderings.min ~> Orderings.ord_class.min
  5274     Orderings.max ~> Orderings.ord_class.max
  5275     Divides.op div ~> Divides.div_class.div
  5276     Divides.op mod ~> Divides.div_class.mod
  5277     Divides.op dvd ~> Divides.div_class.dvd
  5278 
  5279 INCOMPATIBILITY.  Adaptions may be required in the following cases:
  5280 
  5281 a) User-defined constants using any of the names "plus", "minus",
  5282 "times", "less" or "less_eq". The standard syntax translations for
  5283 "+", "-" and "*" may go wrong.  INCOMPATIBILITY: use more specific
  5284 names.
  5285 
  5286 b) Variables named "plus", "minus", "times", "less", "less_eq"
  5287 INCOMPATIBILITY: use more specific names.
  5288 
  5289 c) Permutative equations (e.g. "a + b = b + a")
  5290 Since the change of names also changes the order of terms, permutative
  5291 rewrite rules may get applied in a different order. Experience shows
  5292 that this is rarely the case (only two adaptions in the whole Isabelle
  5293 distribution).  INCOMPATIBILITY: rewrite proofs
  5294 
  5295 d) ML code directly refering to constant names
  5296 This in general only affects hand-written proof tactics, simprocs and
  5297 so on.  INCOMPATIBILITY: grep your sourcecode and replace names.
  5298 Consider using @{const_name} antiquotation.
  5299 
  5300 * New class "default" with associated constant "default".
  5301 
  5302 * Function "sgn" is now overloaded and available on int, real, complex
  5303 (and other numeric types), using class "sgn".  Two possible defs of
  5304 sgn are given as equational assumptions in the classes sgn_if and
  5305 sgn_div_norm; ordered_idom now also inherits from sgn_if.
  5306 INCOMPATIBILITY.
  5307 
  5308 * Locale "partial_order" now unified with class "order" (cf. theory
  5309 Orderings), added parameter "less".  INCOMPATIBILITY.
  5310 
  5311 * Renamings in classes "order" and "linorder": facts "refl", "trans" and
  5312 "cases" to "order_refl", "order_trans" and "linorder_cases", to avoid
  5313 clashes with HOL "refl" and "trans".  INCOMPATIBILITY.
  5314 
  5315 * Classes "order" and "linorder": potential INCOMPATIBILITY due to
  5316 changed order of proof goals in instance proofs.
  5317 
  5318 * The transitivity reasoner for partial and linear orders is set up
  5319 for classes "order" and "linorder".  Instances of the reasoner are available
  5320 in all contexts importing or interpreting the corresponding locales.
  5321 Method "order" invokes the reasoner separately; the reasoner
  5322 is also integrated with the Simplifier as a solver.  Diagnostic
  5323 command 'print_orders' shows the available instances of the reasoner
  5324 in the current context.
  5325 
  5326 * Localized monotonicity predicate in theory "Orderings"; integrated
  5327 lemmas max_of_mono and min_of_mono with this predicate.
  5328 INCOMPATIBILITY.
  5329 
  5330 * Formulation of theorem "dense" changed slightly due to integration
  5331 with new class dense_linear_order.
  5332 
  5333 * Uniform lattice theory development in HOL.
  5334 
  5335     constants "meet" and "join" now named "inf" and "sup"
  5336     constant "Meet" now named "Inf"
  5337 
  5338     classes "meet_semilorder" and "join_semilorder" now named
  5339       "lower_semilattice" and "upper_semilattice"
  5340     class "lorder" now named "lattice"
  5341     class "comp_lat" now named "complete_lattice"
  5342 
  5343     Instantiation of lattice classes allows explicit definitions
  5344     for "inf" and "sup" operations (or "Inf" and "Sup" for complete lattices).
  5345 
  5346   INCOMPATIBILITY.  Theorem renames:
  5347 
  5348     meet_left_le            ~> inf_le1
  5349     meet_right_le           ~> inf_le2
  5350     join_left_le            ~> sup_ge1
  5351     join_right_le           ~> sup_ge2
  5352     meet_join_le            ~> inf_sup_ord
  5353     le_meetI                ~> le_infI
  5354     join_leI                ~> le_supI
  5355     le_meet                 ~> le_inf_iff
  5356     le_join                 ~> ge_sup_conv
  5357     meet_idempotent         ~> inf_idem
  5358     join_idempotent         ~> sup_idem
  5359     meet_comm               ~> inf_commute
  5360     join_comm               ~> sup_commute
  5361     meet_leI1               ~> le_infI1
  5362     meet_leI2               ~> le_infI2
  5363     le_joinI1               ~> le_supI1
  5364     le_joinI2               ~> le_supI2
  5365     meet_assoc              ~> inf_assoc
  5366     join_assoc              ~> sup_assoc
  5367     meet_left_comm          ~> inf_left_commute
  5368     meet_left_idempotent    ~> inf_left_idem
  5369     join_left_comm          ~> sup_left_commute
  5370     join_left_idempotent    ~> sup_left_idem
  5371     meet_aci                ~> inf_aci
  5372     join_aci                ~> sup_aci
  5373     le_def_meet             ~> le_iff_inf
  5374     le_def_join             ~> le_iff_sup
  5375     join_absorp2            ~> sup_absorb2
  5376     join_absorp1            ~> sup_absorb1
  5377     meet_absorp1            ~> inf_absorb1
  5378     meet_absorp2            ~> inf_absorb2
  5379     meet_join_absorp        ~> inf_sup_absorb
  5380     join_meet_absorp        ~> sup_inf_absorb
  5381     distrib_join_le         ~> distrib_sup_le
  5382     distrib_meet_le         ~> distrib_inf_le
  5383 
  5384     add_meet_distrib_left   ~> add_inf_distrib_left
  5385     add_join_distrib_left   ~> add_sup_distrib_left
  5386     is_join_neg_meet        ~> is_join_neg_inf
  5387     is_meet_neg_join        ~> is_meet_neg_sup
  5388     add_meet_distrib_right  ~> add_inf_distrib_right
  5389     add_join_distrib_right  ~> add_sup_distrib_right
  5390     add_meet_join_distribs  ~> add_sup_inf_distribs
  5391     join_eq_neg_meet        ~> sup_eq_neg_inf
  5392     meet_eq_neg_join        ~> inf_eq_neg_sup
  5393     add_eq_meet_join        ~> add_eq_inf_sup
  5394     meet_0_imp_0            ~> inf_0_imp_0
  5395     join_0_imp_0            ~> sup_0_imp_0
  5396     meet_0_eq_0             ~> inf_0_eq_0
  5397     join_0_eq_0             ~> sup_0_eq_0
  5398     neg_meet_eq_join        ~> neg_inf_eq_sup
  5399     neg_join_eq_meet        ~> neg_sup_eq_inf
  5400     join_eq_if              ~> sup_eq_if
  5401 
  5402     mono_meet               ~> mono_inf
  5403     mono_join               ~> mono_sup
  5404     meet_bool_eq            ~> inf_bool_eq
  5405     join_bool_eq            ~> sup_bool_eq
  5406     meet_fun_eq             ~> inf_fun_eq
  5407     join_fun_eq             ~> sup_fun_eq
  5408     meet_set_eq             ~> inf_set_eq
  5409     join_set_eq             ~> sup_set_eq
  5410     meet1_iff               ~> inf1_iff
  5411     meet2_iff               ~> inf2_iff
  5412     meet1I                  ~> inf1I
  5413     meet2I                  ~> inf2I
  5414     meet1D1                 ~> inf1D1
  5415     meet2D1                 ~> inf2D1
  5416     meet1D2                 ~> inf1D2
  5417     meet2D2                 ~> inf2D2
  5418     meet1E                  ~> inf1E
  5419     meet2E                  ~> inf2E
  5420     join1_iff               ~> sup1_iff
  5421     join2_iff               ~> sup2_iff
  5422     join1I1                 ~> sup1I1
  5423     join2I1                 ~> sup2I1
  5424     join1I1                 ~> sup1I1
  5425     join2I2                 ~> sup1I2
  5426     join1CI                 ~> sup1CI
  5427     join2CI                 ~> sup2CI
  5428     join1E                  ~> sup1E
  5429     join2E                  ~> sup2E
  5430 
  5431     is_meet_Meet            ~> is_meet_Inf
  5432     Meet_bool_def           ~> Inf_bool_def
  5433     Meet_fun_def            ~> Inf_fun_def
  5434     Meet_greatest           ~> Inf_greatest
  5435     Meet_lower              ~> Inf_lower
  5436     Meet_set_def            ~> Inf_set_def
  5437 
  5438     Sup_def                 ~> Sup_Inf
  5439     Sup_bool_eq             ~> Sup_bool_def
  5440     Sup_fun_eq              ~> Sup_fun_def
  5441     Sup_set_eq              ~> Sup_set_def
  5442 
  5443     listsp_meetI            ~> listsp_infI
  5444     listsp_meet_eq          ~> listsp_inf_eq
  5445 
  5446     meet_min                ~> inf_min
  5447     join_max                ~> sup_max
  5448 
  5449 * Added syntactic class "size"; overloaded constant "size" now has
  5450 type "'a::size ==> bool"
  5451 
  5452 * Internal reorganisation of `size' of datatypes: size theorems
  5453 "foo.size" are no longer subsumed by "foo.simps" (but are still
  5454 simplification rules by default!); theorems "prod.size" now named
  5455 "*.size".
  5456 
  5457 * Class "div" now inherits from class "times" rather than "type".
  5458 INCOMPATIBILITY.
  5459 
  5460 * HOL/Finite_Set: "name-space" locales Lattice, Distrib_lattice,
  5461 Linorder etc.  have disappeared; operations defined in terms of
  5462 fold_set now are named Inf_fin, Sup_fin.  INCOMPATIBILITY.
  5463 
  5464 * HOL/Nat: neq0_conv no longer declared as iff.  INCOMPATIBILITY.
  5465 
  5466 * HOL-Word: New extensive library and type for generic, fixed size
  5467 machine words, with arithmetic, bit-wise, shifting and rotating
  5468 operations, reflection into int, nat, and bool lists, automation for
  5469 linear arithmetic (by automatic reflection into nat or int), including
  5470 lemmas on overflow and monotonicity.  Instantiated to all appropriate
  5471 arithmetic type classes, supporting automatic simplification of
  5472 numerals on all operations.
  5473 
  5474 * Library/Boolean_Algebra: locales for abstract boolean algebras.
  5475 
  5476 * Library/Numeral_Type: numbers as types, e.g. TYPE(32).
  5477 
  5478 * Code generator library theories:
  5479   - Code_Integer represents HOL integers by big integer literals in target
  5480     languages.
  5481   - Code_Char represents HOL characters by character literals in target
  5482     languages.
  5483   - Code_Char_chr like Code_Char, but also offers treatment of character
  5484     codes; includes Code_Integer.
  5485   - Executable_Set allows to generate code for finite sets using lists.
  5486   - Executable_Rat implements rational numbers as triples (sign, enumerator,
  5487     denominator).
  5488   - Executable_Real implements a subset of real numbers, namly those
  5489     representable by rational numbers.
  5490   - Efficient_Nat implements natural numbers by integers, which in general will
  5491     result in higher efficency; pattern matching with 0/Suc is eliminated;
  5492     includes Code_Integer.
  5493   - Code_Index provides an additional datatype index which is mapped to
  5494     target-language built-in integers.
  5495   - Code_Message provides an additional datatype message_string which is isomorphic to
  5496     strings; messages are mapped to target-language strings.
  5497 
  5498 * New package for inductive predicates
  5499 
  5500   An n-ary predicate p with m parameters z_1, ..., z_m can now be defined via
  5501 
  5502     inductive
  5503       p :: "U_1 => ... => U_m => T_1 => ... => T_n => bool"
  5504       for z_1 :: U_1 and ... and z_n :: U_m
  5505     where
  5506       rule_1: "... ==> p z_1 ... z_m t_1_1 ... t_1_n"
  5507     | ...
  5508 
  5509   with full support for type-inference, rather than
  5510 
  5511     consts s :: "U_1 => ... => U_m => (T_1 * ... * T_n) set"
  5512 
  5513     abbreviation p :: "U_1 => ... => U_m => T_1 => ... => T_n => bool"
  5514     where "p z_1 ... z_m x_1 ... x_n == (x_1, ..., x_n) : s z_1 ... z_m"
  5515 
  5516     inductive "s z_1 ... z_m"
  5517     intros
  5518       rule_1: "... ==> (t_1_1, ..., t_1_n) : s z_1 ... z_m"
  5519       ...
  5520 
  5521   For backward compatibility, there is a wrapper allowing inductive
  5522   sets to be defined with the new package via
  5523 
  5524     inductive_set
  5525       s :: "U_1 => ... => U_m => (T_1 * ... * T_n) set"
  5526       for z_1 :: U_1 and ... and z_n :: U_m
  5527     where
  5528       rule_1: "... ==> (t_1_1, ..., t_1_n) : s z_1 ... z_m"
  5529     | ...
  5530 
  5531   or
  5532 
  5533     inductive_set
  5534       s :: "U_1 => ... => U_m => (T_1 * ... * T_n) set"
  5535       and p :: "U_1 => ... => U_m => T_1 => ... => T_n => bool"
  5536       for z_1 :: U_1 and ... and z_n :: U_m
  5537     where
  5538       "p z_1 ... z_m x_1 ... x_n == (x_1, ..., x_n) : s z_1 ... z_m"
  5539     | rule_1: "... ==> p z_1 ... z_m t_1_1 ... t_1_n"
  5540     | ...
  5541 
  5542   if the additional syntax "p ..." is required.
  5543 
  5544   Numerous examples can be found in the subdirectories src/HOL/Auth,
  5545   src/HOL/Bali, src/HOL/Induct, and src/HOL/MicroJava.
  5546 
  5547   INCOMPATIBILITIES:
  5548 
  5549   - Since declaration and definition of inductive sets or predicates
  5550     is no longer separated, abbreviations involving the newly
  5551     introduced sets or predicates must be specified together with the
  5552     introduction rules after the 'where' keyword (see above), rather
  5553     than before the actual inductive definition.
  5554 
  5555   - The variables in induction and elimination rules are now
  5556     quantified in the order of their occurrence in the introduction
  5557     rules, rather than in alphabetical order. Since this may break
  5558     some proofs, these proofs either have to be repaired, e.g. by
  5559     reordering the variables a_i_1 ... a_i_{k_i} in Isar 'case'
  5560     statements of the form
  5561 
  5562       case (rule_i a_i_1 ... a_i_{k_i})
  5563 
  5564     or the old order of quantification has to be restored by explicitly adding
  5565     meta-level quantifiers in the introduction rules, i.e.
  5566 
  5567       | rule_i: "!!a_i_1 ... a_i_{k_i}. ... ==> p z_1 ... z_m t_i_1 ... t_i_n"
  5568 
  5569   - The format of the elimination rules is now
  5570 
  5571       p z_1 ... z_m x_1 ... x_n ==>
  5572         (!!a_1_1 ... a_1_{k_1}. x_1 = t_1_1 ==> ... ==> x_n = t_1_n ==> ... ==> P)
  5573         ==> ... ==> P
  5574 
  5575     for predicates and
  5576 
  5577       (x_1, ..., x_n) : s z_1 ... z_m ==>
  5578         (!!a_1_1 ... a_1_{k_1}. x_1 = t_1_1 ==> ... ==> x_n = t_1_n ==> ... ==> P)
  5579         ==> ... ==> P
  5580 
  5581     for sets rather than
  5582 
  5583       x : s z_1 ... z_m ==>
  5584         (!!a_1_1 ... a_1_{k_1}. x = (t_1_1, ..., t_1_n) ==> ... ==> P)
  5585         ==> ... ==> P
  5586 
  5587     This may require terms in goals to be expanded to n-tuples
  5588     (e.g. using case_tac or simplification with the split_paired_all
  5589     rule) before the above elimination rule is applicable.
  5590 
  5591   - The elimination or case analysis rules for (mutually) inductive
  5592     sets or predicates are now called "p_1.cases" ... "p_k.cases". The
  5593     list of rules "p_1_..._p_k.elims" is no longer available.
  5594 
  5595 * New package "function"/"fun" for general recursive functions,
  5596 supporting mutual and nested recursion, definitions in local contexts,
  5597 more general pattern matching and partiality. See HOL/ex/Fundefs.thy
  5598 for small examples, and the separate tutorial on the function
  5599 package. The old recdef "package" is still available as before, but
  5600 users are encouraged to use the new package.
  5601 
  5602 * Method "lexicographic_order" automatically synthesizes termination
  5603 relations as lexicographic combinations of size measures. 
  5604 
  5605 * Case-expressions allow arbitrary constructor-patterns (including
  5606 "_") and take their order into account, like in functional
  5607 programming.  Internally, this is translated into nested
  5608 case-expressions; missing cases are added and mapped to the predefined
  5609 constant "undefined". In complicated cases printing may no longer show
  5610 the original input but the internal form. Lambda-abstractions allow
  5611 the same form of pattern matching: "% pat1 => e1 | ..." is an
  5612 abbreviation for "%x. case x of pat1 => e1 | ..." where x is a new
  5613 variable.
  5614 
  5615 * IntDef: The constant "int :: nat => int" has been removed; now "int"
  5616 is an abbreviation for "of_nat :: nat => int". The simplification
  5617 rules for "of_nat" have been changed to work like "int" did
  5618 previously.  Potential INCOMPATIBILITY:
  5619   - "of_nat (Suc m)" simplifies to "1 + of_nat m" instead of "of_nat m + 1"
  5620   - of_nat_diff and of_nat_mult are no longer default simp rules
  5621 
  5622 * Method "algebra" solves polynomial equations over (semi)rings using
  5623 Groebner bases. The (semi)ring structure is defined by locales and the
  5624 tool setup depends on that generic context. Installing the method for
  5625 a specific type involves instantiating the locale and possibly adding
  5626 declarations for computation on the coefficients.  The method is
  5627 already instantiated for natural numbers and for the axiomatic class
  5628 of idoms with numerals.  See also the paper by Chaieb and Wenzel at
  5629 CALCULEMUS 2007 for the general principles underlying this
  5630 architecture of context-aware proof-tools.
  5631 
  5632 * Method "ferrack" implements quantifier elimination over
  5633 special-purpose dense linear orders using locales (analogous to
  5634 "algebra"). The method is already installed for class
  5635 {ordered_field,recpower,number_ring} which subsumes real, hyperreal,
  5636 rat, etc.
  5637 
  5638 * Former constant "List.op @" now named "List.append".  Use ML
  5639 antiquotations @{const_name List.append} or @{term " ... @ ... "} to
  5640 circumvent possible incompatibilities when working on ML level.
  5641 
  5642 * primrec: missing cases mapped to "undefined" instead of "arbitrary".
  5643 
  5644 * New function listsum :: 'a list => 'a for arbitrary monoids.
  5645 Special syntax: "SUM x <- xs. f x" (and latex variants)
  5646 
  5647 * New syntax for Haskell-like list comprehension (input only), eg.
  5648 [(x,y). x <- xs, y <- ys, x ~= y], see also src/HOL/List.thy.
  5649 
  5650 * The special syntax for function "filter" has changed from [x :
  5651 xs. P] to [x <- xs. P] to avoid an ambiguity caused by list
  5652 comprehension syntax, and for uniformity.  INCOMPATIBILITY.
  5653 
  5654 * [a..b] is now defined for arbitrary linear orders.  It used to be
  5655 defined on nat only, as an abbreviation for [a..<Suc b]
  5656 INCOMPATIBILITY.
  5657 
  5658 * Renamed lemma "set_take_whileD"  to "set_takeWhileD".
  5659 
  5660 * New functions "sorted" and "sort" in src/HOL/List.thy.
  5661 
  5662 * New lemma collection field_simps (an extension of ring_simps) for
  5663 manipulating (in)equations involving division. Multiplies with all
  5664 denominators that can be proved to be non-zero (in equations) or
  5665 positive/negative (in inequations).
  5666 
  5667 * Lemma collections ring_eq_simps, group_eq_simps and ring_distrib
  5668 have been improved and renamed to ring_simps, group_simps and
  5669 ring_distribs.  Removed lemmas field_xyz in theory Ring_and_Field
  5670 because they were subsumed by lemmas xyz.  INCOMPATIBILITY.
  5671 
  5672 * Theory Library/Commutative_Ring: switched from recdef to function
  5673 package; constants add, mul, pow now curried.  Infix syntax for
  5674 algebraic operations.
  5675 
  5676 * Dropped redundant lemma def_imp_eq in favor of meta_eq_to_obj_eq.
  5677 INCOMPATIBILITY.
  5678 
  5679 * Dropped redundant lemma if_def2 in favor of if_bool_eq_conj.
  5680 INCOMPATIBILITY.
  5681 
  5682 * HOL/records: generalised field-update to take a function on the
  5683 field rather than the new value: r(|A := x|) is translated to A_update
  5684 (K x) r The K-combinator that is internally used is called K_record.
  5685 INCOMPATIBILITY: Usage of the plain update functions has to be
  5686 adapted.
  5687  
  5688 * Class "semiring_0" now contains annihilation axioms x * 0 = 0 and 0
  5689 * x = 0, which are required for a semiring.  Richer structures do not
  5690 inherit from semiring_0 anymore, because this property is a theorem
  5691 there, not an axiom.  INCOMPATIBILITY: In instances of semiring_0,
  5692 there is more to prove, but this is mostly trivial.
  5693 
  5694 * Class "recpower" is generalized to arbitrary monoids, not just
  5695 commutative semirings.  INCOMPATIBILITY: may need to incorporate
  5696 commutativity or semiring properties additionally.
  5697 
  5698 * Constant "List.list_all2" in List.thy now uses authentic syntax.
  5699 INCOMPATIBILITY: translations containing list_all2 may go wrong,
  5700 better use 'abbreviation'.
  5701 
  5702 * Renamed constant "List.op mem" to "List.member".  INCOMPATIBILITY.
  5703 
  5704 * Numeral syntax: type 'bin' which was a mere type copy of 'int' has
  5705 been abandoned in favour of plain 'int'.  INCOMPATIBILITY --
  5706 significant changes for setting up numeral syntax for types:
  5707   - New constants Numeral.pred and Numeral.succ instead
  5708       of former Numeral.bin_pred and Numeral.bin_succ.
  5709   - Use integer operations instead of bin_add, bin_mult and so on.
  5710   - Numeral simplification theorems named Numeral.numeral_simps instead of Bin_simps.
  5711   - ML structure Bin_Simprocs now named Int_Numeral_Base_Simprocs.
  5712 
  5713 See src/HOL/Integ/IntArith.thy for an example setup.
  5714 
  5715 * Command 'normal_form' computes the normal form of a term that may
  5716 contain free variables.  For example ``normal_form "rev [a, b, c]"''
  5717 produces ``[b, c, a]'' (without proof).  This command is suitable for
  5718 heavy-duty computations because the functions are compiled to ML
  5719 first.  Correspondingly, a method "normalization" is provided.  See
  5720 further src/HOL/ex/NormalForm.thy and src/Tools/nbe.ML.
  5721 
  5722 * Alternative iff syntax "A <-> B" for equality on bool (with priority
  5723 25 like -->); output depends on the "iff" print_mode, the default is
  5724 "A = B" (with priority 50).
  5725 
  5726 * Relations less (<) and less_eq (<=) are also available on type bool.
  5727 Modified syntax to disallow nesting without explicit parentheses,
  5728 e.g. "(x < y) < z" or "x < (y < z)", but NOT "x < y < z".  Potential
  5729 INCOMPATIBILITY.
  5730 
  5731 * "LEAST x:A. P" expands to "LEAST x. x:A & P" (input only).
  5732 
  5733 * Relation composition operator "op O" now has precedence 75 and binds
  5734 stronger than union and intersection. INCOMPATIBILITY.
  5735 
  5736 * The old set interval syntax "{m..n(}" (and relatives) has been
  5737 removed.  Use "{m..<n}" (and relatives) instead.
  5738 
  5739 * In the context of the assumption "~(s = t)" the Simplifier rewrites
  5740 "t = s" to False (by simproc "neq").  INCOMPATIBILITY, consider using
  5741 ``declare [[simproc del: neq]]''.
  5742 
  5743 * Simplifier: "m dvd n" where m and n are numbers is evaluated to
  5744 True/False.
  5745 
  5746 * Theorem Cons_eq_map_conv no longer declared as "simp".
  5747 
  5748 * Theorem setsum_mult renamed to setsum_right_distrib.
  5749 
  5750 * Prefer ex1I over ex_ex1I in single-step reasoning, e.g. by the
  5751 ``rule'' method.
  5752 
  5753 * Reimplemented methods "sat" and "satx", with several improvements:
  5754 goals no longer need to be stated as "<prems> ==> False", equivalences
  5755 (i.e. "=" on type bool) are handled, variable names of the form
  5756 "lit_<n>" are no longer reserved, significant speedup.
  5757 
  5758 * Methods "sat" and "satx" can now replay MiniSat proof traces.
  5759 zChaff is still supported as well.
  5760 
  5761 * 'inductive' and 'datatype': provide projections of mutual rules,
  5762 bundled as foo_bar.inducts;
  5763 
  5764 * Library: moved theories Parity, GCD, Binomial, Infinite_Set to
  5765 Library.
  5766 
  5767 * Library: moved theory Accessible_Part to main HOL.
  5768 
  5769 * Library: added theory Coinductive_List of potentially infinite lists
  5770 as greatest fixed-point.
  5771 
  5772 * Library: added theory AssocList which implements (finite) maps as
  5773 association lists.
  5774 
  5775 * Method "evaluation" solves goals (i.e. a boolean expression)
  5776 efficiently by compiling it to ML.  The goal is "proved" (via an
  5777 oracle) if it evaluates to True.
  5778 
  5779 * Linear arithmetic now splits certain operators (e.g. min, max, abs)
  5780 also when invoked by the simplifier.  This results in the Simplifier
  5781 being more powerful on arithmetic goals.  INCOMPATIBILITY.
  5782 Configuration option fast_arith_split_limit=0 recovers the old
  5783 behavior.
  5784 
  5785 * Support for hex (0x20) and binary (0b1001) numerals.
  5786 
  5787 * New method: reify eqs (t), where eqs are equations for an
  5788 interpretation I :: 'a list => 'b => 'c and t::'c is an optional
  5789 parameter, computes a term s::'b and a list xs::'a list and proves the
  5790 theorem I xs s = t. This is also known as reification or quoting. The
  5791 resulting theorem is applied to the subgoal to substitute t with I xs
  5792 s.  If t is omitted, the subgoal itself is reified.
  5793 
  5794 * New method: reflection corr_thm eqs (t). The parameters eqs and (t)
  5795 are as explained above. corr_thm is a theorem for I vs (f t) = I vs t,
  5796 where f is supposed to be a computable function (in the sense of code
  5797 generattion). The method uses reify to compute s and xs as above then
  5798 applies corr_thm and uses normalization by evaluation to "prove" f s =
  5799 r and finally gets the theorem t = r, which is again applied to the
  5800 subgoal. An Example is available in src/HOL/ex/ReflectionEx.thy.
  5801 
  5802 * Reflection: Automatic reification now handels binding, an example is
  5803 available in src/HOL/ex/ReflectionEx.thy
  5804 
  5805 * HOL-Statespace: ``State Spaces: The Locale Way'' introduces a
  5806 command 'statespace' that is similar to 'record', but introduces an
  5807 abstract specification based on the locale infrastructure instead of
  5808 HOL types.  This leads to extra flexibility in composing state spaces,
  5809 in particular multiple inheritance and renaming of components.
  5810 
  5811 
  5812 *** HOL-Complex ***
  5813 
  5814 * Hyperreal: Functions root and sqrt are now defined on negative real
  5815 inputs so that root n (- x) = - root n x and sqrt (- x) = - sqrt x.
  5816 Nonnegativity side conditions have been removed from many lemmas, so
  5817 that more subgoals may now be solved by simplification; potential
  5818 INCOMPATIBILITY.
  5819 
  5820 * Real: new type classes formalize real normed vector spaces and
  5821 algebras, using new overloaded constants scaleR :: real => 'a => 'a
  5822 and norm :: 'a => real.
  5823 
  5824 * Real: constant of_real :: real => 'a::real_algebra_1 injects from
  5825 reals into other types. The overloaded constant Reals :: 'a set is now
  5826 defined as range of_real; potential INCOMPATIBILITY.
  5827 
  5828 * Real: proper support for ML code generation, including 'quickcheck'.
  5829 Reals are implemented as arbitrary precision rationals.
  5830 
  5831 * Hyperreal: Several constants that previously worked only for the
  5832 reals have been generalized, so they now work over arbitrary vector
  5833 spaces. Type annotations may need to be added in some cases; potential
  5834 INCOMPATIBILITY.
  5835 
  5836   Infinitesimal  :: ('a::real_normed_vector) star set
  5837   HFinite        :: ('a::real_normed_vector) star set
  5838   HInfinite      :: ('a::real_normed_vector) star set
  5839   approx         :: ('a::real_normed_vector) star => 'a star => bool
  5840   monad          :: ('a::real_normed_vector) star => 'a star set
  5841   galaxy         :: ('a::real_normed_vector) star => 'a star set
  5842   (NS)LIMSEQ     :: [nat => 'a::real_normed_vector, 'a] => bool
  5843   (NS)convergent :: (nat => 'a::real_normed_vector) => bool
  5844   (NS)Bseq       :: (nat => 'a::real_normed_vector) => bool
  5845   (NS)Cauchy     :: (nat => 'a::real_normed_vector) => bool
  5846   (NS)LIM        :: ['a::real_normed_vector => 'b::real_normed_vector, 'a, 'b] => bool
  5847   is(NS)Cont     :: ['a::real_normed_vector => 'b::real_normed_vector, 'a] => bool
  5848   deriv          :: ['a::real_normed_field => 'a, 'a, 'a] => bool
  5849   sgn            :: 'a::real_normed_vector => 'a
  5850   exp            :: 'a::{recpower,real_normed_field,banach} => 'a
  5851 
  5852 * Complex: Some complex-specific constants are now abbreviations for
  5853 overloaded ones: complex_of_real = of_real, cmod = norm, hcmod =
  5854 hnorm.  Other constants have been entirely removed in favor of the
  5855 polymorphic versions (INCOMPATIBILITY):
  5856 
  5857   approx        <-- capprox
  5858   HFinite       <-- CFinite
  5859   HInfinite     <-- CInfinite
  5860   Infinitesimal <-- CInfinitesimal
  5861   monad         <-- cmonad
  5862   galaxy        <-- cgalaxy
  5863   (NS)LIM       <-- (NS)CLIM, (NS)CRLIM
  5864   is(NS)Cont    <-- is(NS)Contc, is(NS)contCR
  5865   (ns)deriv     <-- (ns)cderiv
  5866 
  5867 
  5868 *** HOL-Algebra ***
  5869 
  5870 * Formalisation of ideals and the quotient construction over rings.
  5871 
  5872 * Order and lattice theory no longer based on records.
  5873 INCOMPATIBILITY.
  5874 
  5875 * Renamed lemmas least_carrier -> least_closed and greatest_carrier ->
  5876 greatest_closed.  INCOMPATIBILITY.
  5877 
  5878 * Method algebra is now set up via an attribute.  For examples see
  5879 Ring.thy.  INCOMPATIBILITY: the method is now weaker on combinations
  5880 of algebraic structures.
  5881 
  5882 * Renamed theory CRing to Ring.
  5883 
  5884 
  5885 *** HOL-Nominal ***
  5886 
  5887 * Substantial, yet incomplete support for nominal datatypes (binding
  5888 structures) based on HOL-Nominal logic.  See src/HOL/Nominal and
  5889 src/HOL/Nominal/Examples.  Prospective users should consult
  5890 http://isabelle.in.tum.de/nominal/
  5891 
  5892 
  5893 *** ML ***
  5894 
  5895 * ML basics: just one true type int, which coincides with IntInf.int
  5896 (even on SML/NJ).
  5897 
  5898 * ML within Isar: antiquotations allow to embed statically-checked
  5899 formal entities in the source, referring to the context available at
  5900 compile-time.  For example:
  5901 
  5902 ML {* @{sort "{zero,one}"} *}
  5903 ML {* @{typ "'a => 'b"} *}
  5904 ML {* @{term "%x. x"} *}
  5905 ML {* @{prop "x == y"} *}
  5906 ML {* @{ctyp "'a => 'b"} *}
  5907 ML {* @{cterm "%x. x"} *}
  5908 ML {* @{cprop "x == y"} *}
  5909 ML {* @{thm asm_rl} *}
  5910 ML {* @{thms asm_rl} *}
  5911 ML {* @{type_name c} *}
  5912 ML {* @{type_syntax c} *}
  5913 ML {* @{const_name c} *}
  5914 ML {* @{const_syntax c} *}
  5915 ML {* @{context} *}
  5916 ML {* @{theory} *}
  5917 ML {* @{theory Pure} *}
  5918 ML {* @{theory_ref} *}
  5919 ML {* @{theory_ref Pure} *}
  5920 ML {* @{simpset} *}
  5921 ML {* @{claset} *}
  5922 ML {* @{clasimpset} *}
  5923 
  5924 The same works for sources being ``used'' within an Isar context.
  5925 
  5926 * ML in Isar: improved error reporting; extra verbosity with
  5927 ML_Context.trace enabled.
  5928 
  5929 * Pure/General/table.ML: the join operations now works via exceptions
  5930 DUP/SAME instead of type option. This is simpler in simple cases, and
  5931 admits slightly more efficient complex applications.
  5932 
  5933 * Pure: 'advanced' translation functions (parse_translation etc.) now
  5934 use Context.generic instead of just theory.
  5935 
  5936 * Pure: datatype Context.generic joins theory/Proof.context and
  5937 provides some facilities for code that works in either kind of
  5938 context, notably GenericDataFun for uniform theory and proof data.
  5939 
  5940 * Pure: simplified internal attribute type, which is now always
  5941 Context.generic * thm -> Context.generic * thm. Global (theory) vs.
  5942 local (Proof.context) attributes have been discontinued, while
  5943 minimizing code duplication. Thm.rule_attribute and
  5944 Thm.declaration_attribute build canonical attributes; see also structure
  5945 Context for further operations on Context.generic, notably
  5946 GenericDataFun. INCOMPATIBILITY, need to adapt attribute type
  5947 declarations and definitions.
  5948 
  5949 * Context data interfaces (Theory/Proof/GenericDataFun): removed
  5950 name/print, uninitialized data defaults to ad-hoc copy of empty value,
  5951 init only required for impure data. INCOMPATIBILITY: empty really need
  5952 to be empty (no dependencies on theory content!)
  5953 
  5954 * Pure/kernel: consts certification ignores sort constraints given in
  5955 signature declarations. (This information is not relevant to the
  5956 logic, but only for type inference.) SIGNIFICANT INTERNAL CHANGE,
  5957 potential INCOMPATIBILITY.
  5958 
  5959 * Pure: axiomatic type classes are now purely definitional, with
  5960 explicit proofs of class axioms and super class relations performed
  5961 internally. See Pure/axclass.ML for the main internal interfaces --
  5962 notably AxClass.define_class supercedes AxClass.add_axclass, and
  5963 AxClass.axiomatize_class/classrel/arity supersede
  5964 Sign.add_classes/classrel/arities.
  5965 
  5966 * Pure/Isar: Args/Attrib parsers operate on Context.generic --
  5967 global/local versions on theory vs. Proof.context have been
  5968 discontinued; Attrib.syntax and Method.syntax have been adapted
  5969 accordingly.  INCOMPATIBILITY, need to adapt parser expressions for
  5970 attributes, methods, etc.
  5971 
  5972 * Pure: several functions of signature "... -> theory -> theory * ..."
  5973 have been reoriented to "... -> theory -> ... * theory" in order to
  5974 allow natural usage in combination with the ||>, ||>>, |-> and
  5975 fold_map combinators.
  5976 
  5977 * Pure: official theorem names (closed derivations) and additional
  5978 comments (tags) are now strictly separate.  Name hints -- which are
  5979 maintained as tags -- may be attached any time without affecting the
  5980 derivation.
  5981 
  5982 * Pure: primitive rule lift_rule now takes goal cterm instead of an
  5983 actual goal state (thm).  Use Thm.lift_rule (Thm.cprem_of st i) to
  5984 achieve the old behaviour.
  5985 
  5986 * Pure: the "Goal" constant is now called "prop", supporting a
  5987 slightly more general idea of ``protecting'' meta-level rule
  5988 statements.
  5989 
  5990 * Pure: Logic.(un)varify only works in a global context, which is now
  5991 enforced instead of silently assumed.  INCOMPATIBILITY, may use
  5992 Logic.legacy_(un)varify as temporary workaround.
  5993 
  5994 * Pure: structure Name provides scalable operations for generating
  5995 internal variable names, notably Name.variants etc.  This replaces
  5996 some popular functions from term.ML:
  5997 
  5998   Term.variant		->  Name.variant
  5999   Term.variantlist	->  Name.variant_list
  6000   Term.invent_names	->  Name.invent_list
  6001 
  6002 Note that low-level renaming rarely occurs in new code -- operations
  6003 from structure Variable are used instead (see below).
  6004 
  6005 * Pure: structure Variable provides fundamental operations for proper
  6006 treatment of fixed/schematic variables in a context.  For example,
  6007 Variable.import introduces fixes for schematics of given facts and
  6008 Variable.export reverses the effect (up to renaming) -- this replaces
  6009 various freeze_thaw operations.
  6010 
  6011 * Pure: structure Goal provides simple interfaces for
  6012 init/conclude/finish and tactical prove operations (replacing former
  6013 Tactic.prove).  Goal.prove is the canonical way to prove results
  6014 within a given context; Goal.prove_global is a degraded version for
  6015 theory level goals, including a global Drule.standard.  Note that
  6016 OldGoals.prove_goalw_cterm has long been obsolete, since it is
  6017 ill-behaved in a local proof context (e.g. with local fixes/assumes or
  6018 in a locale context).
  6019 
  6020 * Pure/Syntax: generic interfaces for parsing (Syntax.parse_term etc.)
  6021 and type checking (Syntax.check_term etc.), with common combinations
  6022 (Syntax.read_term etc.). These supersede former Sign.read_term etc.
  6023 which are considered legacy and await removal.
  6024 
  6025 * Pure/Syntax: generic interfaces for type unchecking
  6026 (Syntax.uncheck_terms etc.) and unparsing (Syntax.unparse_term etc.),
  6027 with common combinations (Syntax.pretty_term, Syntax.string_of_term
  6028 etc.).  Former Sign.pretty_term, Sign.string_of_term etc. are still
  6029 available for convenience, but refer to the very same operations using
  6030 a mere theory instead of a full context.
  6031 
  6032 * Isar: simplified treatment of user-level errors, using exception
  6033 ERROR of string uniformly.  Function error now merely raises ERROR,
  6034 without any side effect on output channels.  The Isar toplevel takes
  6035 care of proper display of ERROR exceptions.  ML code may use plain
  6036 handle/can/try; cat_error may be used to concatenate errors like this:
  6037 
  6038   ... handle ERROR msg => cat_error msg "..."
  6039 
  6040 Toplevel ML code (run directly or through the Isar toplevel) may be
  6041 embedded into the Isar toplevel with exception display/debug like
  6042 this:
  6043 
  6044   Isar.toplevel (fn () => ...)
  6045 
  6046 INCOMPATIBILITY, removed special transform_error facilities, removed
  6047 obsolete variants of user-level exceptions (ERROR_MESSAGE,
  6048 Context.PROOF, ProofContext.CONTEXT, Proof.STATE, ProofHistory.FAIL)
  6049 -- use plain ERROR instead.
  6050 
  6051 * Isar: theory setup now has type (theory -> theory), instead of a
  6052 list.  INCOMPATIBILITY, may use #> to compose setup functions.
  6053 
  6054 * Isar: ML toplevel pretty printer for type Proof.context, subject to
  6055 ProofContext.debug/verbose flags.
  6056 
  6057 * Isar: Toplevel.theory_to_proof admits transactions that modify the
  6058 theory before entering a proof state.  Transactions now always see a
  6059 quasi-functional intermediate checkpoint, both in interactive and
  6060 batch mode.
  6061 
  6062 * Isar: simplified interfaces for outer syntax.  Renamed
  6063 OuterSyntax.add_keywords to OuterSyntax.keywords.  Removed
  6064 OuterSyntax.add_parsers -- this functionality is now included in
  6065 OuterSyntax.command etc.  INCOMPATIBILITY.
  6066 
  6067 * Simplifier: the simpset of a running simplification process now
  6068 contains a proof context (cf. Simplifier.the_context), which is the
  6069 very context that the initial simpset has been retrieved from (by
  6070 simpset_of/local_simpset_of).  Consequently, all plug-in components
  6071 (solver, looper etc.) may depend on arbitrary proof data.
  6072 
  6073 * Simplifier.inherit_context inherits the proof context (plus the
  6074 local bounds) of the current simplification process; any simproc
  6075 etc. that calls the Simplifier recursively should do this!  Removed
  6076 former Simplifier.inherit_bounds, which is already included here --
  6077 INCOMPATIBILITY.  Tools based on low-level rewriting may even have to
  6078 specify an explicit context using Simplifier.context/theory_context.
  6079 
  6080 * Simplifier/Classical Reasoner: more abstract interfaces
  6081 change_simpset/claset for modifying the simpset/claset reference of a
  6082 theory; raw versions simpset/claset_ref etc. have been discontinued --
  6083 INCOMPATIBILITY.
  6084 
  6085 * Provers: more generic wrt. syntax of object-logics, avoid hardwired
  6086 "Trueprop" etc.
  6087 
  6088 
  6089 *** System ***
  6090 
  6091 * settings: the default heap location within ISABELLE_HOME_USER now
  6092 includes ISABELLE_IDENTIFIER.  This simplifies use of multiple
  6093 Isabelle installations.
  6094 
  6095 * isabelle-process: option -S (secure mode) disables some critical
  6096 operations, notably runtime compilation and evaluation of ML source
  6097 code.
  6098 
  6099 * Basic Isabelle mode for jEdit, see Isabelle/lib/jedit/.
  6100 
  6101 * Support for parallel execution, using native multicore support of
  6102 Poly/ML 5.1.  The theory loader exploits parallelism when processing
  6103 independent theories, according to the given theory header
  6104 specifications. The maximum number of worker threads is specified via
  6105 usedir option -M or the "max-threads" setting in Proof General. A
  6106 speedup factor of 1.5--3.5 can be expected on a 4-core machine, and up
  6107 to 6 on a 8-core machine.  User-code needs to observe certain
  6108 guidelines for thread-safe programming, see appendix A in the Isar
  6109 Implementation manual.
  6110 
  6111 
  6112 
  6113 New in Isabelle2005 (October 2005)
  6114 ----------------------------------
  6115 
  6116 *** General ***
  6117 
  6118 * Theory headers: the new header syntax for Isar theories is
  6119 
  6120   theory <name>
  6121   imports <theory1> ... <theoryN>
  6122   uses <file1> ... <fileM>
  6123   begin
  6124 
  6125 where the 'uses' part is optional.  The previous syntax
  6126 
  6127   theory <name> = <theory1> + ... + <theoryN>:
  6128 
  6129 will disappear in the next release.  Use isatool fixheaders to convert
  6130 existing theory files.  Note that there is no change in ancient
  6131 non-Isar theories now, but these will disappear soon.
  6132 
  6133 * Theory loader: parent theories can now also be referred to via
  6134 relative and absolute paths.
  6135 
  6136 * Command 'find_theorems' searches for a list of criteria instead of a
  6137 list of constants. Known criteria are: intro, elim, dest, name:string,
  6138 simp:term, and any term. Criteria can be preceded by '-' to select
  6139 theorems that do not match. Intro, elim, dest select theorems that
  6140 match the current goal, name:s selects theorems whose fully qualified
  6141 name contain s, and simp:term selects all simplification rules whose
  6142 lhs match term.  Any other term is interpreted as pattern and selects
  6143 all theorems matching the pattern. Available in ProofGeneral under
  6144 'ProofGeneral -> Find Theorems' or C-c C-f.  Example:
  6145 
  6146   C-c C-f (100) "(_::nat) + _ + _" intro -name: "HOL."
  6147 
  6148 prints the last 100 theorems matching the pattern "(_::nat) + _ + _",
  6149 matching the current goal as introduction rule and not having "HOL."
  6150 in their name (i.e. not being defined in theory HOL).
  6151 
  6152 * Command 'thms_containing' has been discontinued in favour of
  6153 'find_theorems'; INCOMPATIBILITY.
  6154 
  6155 * Communication with Proof General is now 8bit clean, which means that
  6156 Unicode text in UTF-8 encoding may be used within theory texts (both
  6157 formal and informal parts).  Cf. option -U of the Isabelle Proof
  6158 General interface.  Here are some simple examples (cf. src/HOL/ex):
  6159 
  6160   http://isabelle.in.tum.de/library/HOL/ex/Hebrew.html
  6161   http://isabelle.in.tum.de/library/HOL/ex/Chinese.html
  6162 
  6163 * Improved efficiency of the Simplifier and, to a lesser degree, the
  6164 Classical Reasoner.  Typical big applications run around 2 times
  6165 faster.
  6166 
  6167 
  6168 *** Document preparation ***
  6169 
  6170 * Commands 'display_drafts' and 'print_drafts' perform simple output
  6171 of raw sources.  Only those symbols that do not require additional
  6172 LaTeX packages (depending on comments in isabellesym.sty) are
  6173 displayed properly, everything else is left verbatim.  isatool display
  6174 and isatool print are used as front ends (these are subject to the
  6175 DVI/PDF_VIEWER and PRINT_COMMAND settings, respectively).
  6176 
  6177 * Command tags control specific markup of certain regions of text,
  6178 notably folding and hiding.  Predefined tags include "theory" (for
  6179 theory begin and end), "proof" for proof commands, and "ML" for
  6180 commands involving ML code; the additional tags "visible" and
  6181 "invisible" are unused by default.  Users may give explicit tag
  6182 specifications in the text, e.g. ''by %invisible (auto)''.  The
  6183 interpretation of tags is determined by the LaTeX job during document
  6184 preparation: see option -V of isatool usedir, or options -n and -t of
  6185 isatool document, or even the LaTeX macros \isakeeptag, \isafoldtag,
  6186 \isadroptag.
  6187 
  6188 Several document versions may be produced at the same time via isatool
  6189 usedir (the generated index.html will link all of them).  Typical
  6190 specifications include ''-V document=theory,proof,ML'' to present
  6191 theory/proof/ML parts faithfully, ''-V outline=/proof,/ML'' to fold
  6192 proof and ML commands, and ''-V mutilated=-theory,-proof,-ML'' to omit
  6193 these parts without any formal replacement text.  The Isabelle site
  6194 default settings produce ''document'' and ''outline'' versions as
  6195 specified above.
  6196 
  6197 * Several new antiquotations:
  6198 
  6199   @{term_type term} prints a term with its type annotated;
  6200 
  6201   @{typeof term} prints the type of a term;
  6202 
  6203   @{const const} is the same as @{term const}, but checks that the
  6204   argument is a known logical constant;
  6205 
  6206   @{term_style style term} and @{thm_style style thm} print a term or
  6207   theorem applying a "style" to it
  6208 
  6209   @{ML text}
  6210 
  6211 Predefined styles are 'lhs' and 'rhs' printing the lhs/rhs of
  6212 definitions, equations, inequations etc., 'concl' printing only the
  6213 conclusion of a meta-logical statement theorem, and 'prem1' .. 'prem19'
  6214 to print the specified premise.  TermStyle.add_style provides an ML
  6215 interface for introducing further styles.  See also the "LaTeX Sugar"
  6216 document practical applications.  The ML antiquotation prints
  6217 type-checked ML expressions verbatim.
  6218 
  6219 * Markup commands 'chapter', 'section', 'subsection', 'subsubsection',
  6220 and 'text' support optional locale specification '(in loc)', which
  6221 specifies the default context for interpreting antiquotations.  For
  6222 example: 'text (in lattice) {* @{thm inf_assoc}*}'.
  6223 
  6224 * Option 'locale=NAME' of antiquotations specifies an alternative
  6225 context interpreting the subsequent argument.  For example: @{thm
  6226 [locale=lattice] inf_assoc}.
  6227 
  6228 * Proper output of proof terms (@{prf ...} and @{full_prf ...}) within
  6229 a proof context.
  6230 
  6231 * Proper output of antiquotations for theory commands involving a
  6232 proof context (such as 'locale' or 'theorem (in loc) ...').
  6233 
  6234 * Delimiters of outer tokens (string etc.) now produce separate LaTeX
  6235 macros (\isachardoublequoteopen, isachardoublequoteclose etc.).
  6236