NEWS
author bulwahn
Thu Jun 21 13:51:44 2012 +0200 (2012-06-21)
changeset 48111 33414f2e82ab
parent 48094 c3d4f4d9e54c
child 48120 9fe0e71052a0
permissions -rw-r--r--
NEWS and CONTRIBUTORS
     1 Isabelle NEWS -- history user-relevant changes
     2 ==============================================
     3 
     4 New in this Isabelle version
     5 ----------------------------
     6 
     7 *** General ***
     8 
     9 * Discontinued obsolete method fastsimp / tactic fast_simp_tac, which
    10 is called fastforce / fast_force_tac already since Isabelle2011-1.
    11 
    12 
    13 *** System ***
    14 
    15 * Discontinued support for Poly/ML 5.2.1, which was the last version
    16 without exception positions and advanced ML compiler/toplevel
    17 configuration.
    18 
    19 
    20 *** HOL ***
    21 
    22 * Simproc for rewriting set comprehensions into pointfree expressions
    23 
    24 * Quickcheck:
    25 
    26   - added an optimisation for equality premises.
    27     It is switched on by default, and can be switched off by setting
    28     the configuration quickcheck_optimise_equality to false.    
    29 
    30 * The SMT solver Z3 has now by default a restricted set of directly
    31 supported features. For the full set of features (div/mod, nonlinear
    32 arithmetic, datatypes/records) with potential proof reconstruction
    33 failures, enable the configuration option "z3_with_extensions".
    34 Minor INCOMPATIBILITY.
    35 
    36 * Sledgehammer:
    37 
    38   - Rationalized type encodings ("type_enc" option).
    39 
    40 New in Isabelle2012 (May 2012)
    41 ------------------------------
    42 
    43 *** General ***
    44 
    45 * Prover IDE (PIDE) improvements:
    46 
    47   - more robust Sledgehammer integration (as before the sledgehammer
    48     command-line needs to be typed into the source buffer)
    49   - markup for bound variables
    50   - markup for types of term variables (displayed as tooltips)
    51   - support for user-defined Isar commands within the running session
    52   - improved support for Unicode outside original 16bit range
    53     e.g. glyph for \<A> (thanks to jEdit 4.5.1)
    54 
    55 * Forward declaration of outer syntax keywords within the theory
    56 header -- minor INCOMPATIBILITY for user-defined commands.  Allow new
    57 commands to be used in the same theory where defined.
    58 
    59 * Auxiliary contexts indicate block structure for specifications with
    60 additional parameters and assumptions.  Such unnamed contexts may be
    61 nested within other targets, like 'theory', 'locale', 'class',
    62 'instantiation' etc.  Results from the local context are generalized
    63 accordingly and applied to the enclosing target context.  Example:
    64 
    65   context
    66     fixes x y z :: 'a
    67     assumes xy: "x = y" and yz: "y = z"
    68   begin
    69 
    70   lemma my_trans: "x = z" using xy yz by simp
    71 
    72   end
    73 
    74   thm my_trans
    75 
    76 The most basic application is to factor-out context elements of
    77 several fixes/assumes/shows theorem statements, e.g. see
    78 ~~/src/HOL/Isar_Examples/Group_Context.thy
    79 
    80 Any other local theory specification element works within the "context
    81 ... begin ... end" block as well.
    82 
    83 * Bundled declarations associate attributed fact expressions with a
    84 given name in the context.  These may be later included in other
    85 contexts.  This allows to manage context extensions casually, without
    86 the logical dependencies of locales and locale interpretation.  See
    87 commands 'bundle', 'include', 'including' etc. in the isar-ref manual.
    88 
    89 * Commands 'lemmas' and 'theorems' allow local variables using 'for'
    90 declaration, and results are standardized before being stored.  Thus
    91 old-style "standard" after instantiation or composition of facts
    92 becomes obsolete.  Minor INCOMPATIBILITY, due to potential change of
    93 indices of schematic variables.
    94 
    95 * Rule attributes in local theory declarations (e.g. locale or class)
    96 are now statically evaluated: the resulting theorem is stored instead
    97 of the original expression.  INCOMPATIBILITY in rare situations, where
    98 the historic accident of dynamic re-evaluation in interpretations
    99 etc. was exploited.
   100 
   101 * New tutorial "Programming and Proving in Isabelle/HOL"
   102 ("prog-prove").  It completely supersedes "A Tutorial Introduction to
   103 Structured Isar Proofs" ("isar-overview"), which has been removed.  It
   104 also supersedes "Isabelle/HOL, A Proof Assistant for Higher-Order
   105 Logic" as the recommended beginners tutorial, but does not cover all
   106 of the material of that old tutorial.
   107 
   108 * Updated and extended reference manuals: "isar-ref",
   109 "implementation", "system"; reduced remaining material in old "ref"
   110 manual.
   111 
   112 
   113 *** Pure ***
   114 
   115 * Command 'definition' no longer exports the foundational "raw_def"
   116 into the user context.  Minor INCOMPATIBILITY, may use the regular
   117 "def" result with attribute "abs_def" to imitate the old version.
   118 
   119 * Attribute "abs_def" turns an equation of the form "f x y == t" into
   120 "f == %x y. t", which ensures that "simp" or "unfold" steps always
   121 expand it.  This also works for object-logic equality.  (Formerly
   122 undocumented feature.)
   123 
   124 * Sort constraints are now propagated in simultaneous statements, just
   125 like type constraints.  INCOMPATIBILITY in rare situations, where
   126 distinct sorts used to be assigned accidentally.  For example:
   127 
   128   lemma "P (x::'a::foo)" and "Q (y::'a::bar)"  -- "now illegal"
   129 
   130   lemma "P (x::'a)" and "Q (y::'a::bar)"
   131     -- "now uniform 'a::bar instead of default sort for first occurrence (!)"
   132 
   133 * Rule composition via attribute "OF" (or ML functions OF/MRS) is more
   134 tolerant against multiple unifiers, as long as the final result is
   135 unique.  (As before, rules are composed in canonical right-to-left
   136 order to accommodate newly introduced premises.)
   137 
   138 * Renamed some inner syntax categories:
   139 
   140     num ~> num_token
   141     xnum ~> xnum_token
   142     xstr ~> str_token
   143 
   144 Minor INCOMPATIBILITY.  Note that in practice "num_const" or
   145 "num_position" etc. are mainly used instead (which also include
   146 position information via constraints).
   147 
   148 * Simplified configuration options for syntax ambiguity: see
   149 "syntax_ambiguity_warning" and "syntax_ambiguity_limit" in isar-ref
   150 manual.  Minor INCOMPATIBILITY.
   151 
   152 * Discontinued configuration option "syntax_positions": atomic terms
   153 in parse trees are always annotated by position constraints.
   154 
   155 * Old code generator for SML and its commands 'code_module',
   156 'code_library', 'consts_code', 'types_code' have been discontinued.
   157 Use commands of the generic code generator instead.  INCOMPATIBILITY.
   158 
   159 * Redundant attribute "code_inline" has been discontinued. Use
   160 "code_unfold" instead.  INCOMPATIBILITY.
   161 
   162 * Dropped attribute "code_unfold_post" in favor of the its dual
   163 "code_abbrev", which yields a common pattern in definitions like
   164 
   165   definition [code_abbrev]: "f = t"
   166 
   167 INCOMPATIBILITY.
   168 
   169 * Obsolete 'types' command has been discontinued.  Use 'type_synonym'
   170 instead.  INCOMPATIBILITY.
   171 
   172 * Discontinued old "prems" fact, which used to refer to the accidental
   173 collection of foundational premises in the context (already marked as
   174 legacy since Isabelle2011).
   175 
   176 
   177 *** HOL ***
   178 
   179 * Type 'a set is now a proper type constructor (just as before
   180 Isabelle2008).  Definitions mem_def and Collect_def have disappeared.
   181 Non-trivial INCOMPATIBILITY.  For developments keeping predicates and
   182 sets separate, it is often sufficient to rephrase some set S that has
   183 been accidentally used as predicates by "%x. x : S", and some
   184 predicate P that has been accidentally used as set by "{x. P x}".
   185 Corresponding proofs in a first step should be pruned from any
   186 tinkering with former theorems mem_def and Collect_def as far as
   187 possible.
   188 
   189 For developments which deliberately mix predicates and sets, a
   190 planning step is necessary to determine what should become a predicate
   191 and what a set.  It can be helpful to carry out that step in
   192 Isabelle2011-1 before jumping right into the current release.
   193 
   194 * Code generation by default implements sets as container type rather
   195 than predicates.  INCOMPATIBILITY.
   196 
   197 * New type synonym 'a rel = ('a * 'a) set
   198 
   199 * The representation of numerals has changed.  Datatype "num"
   200 represents strictly positive binary numerals, along with functions
   201 "numeral :: num => 'a" and "neg_numeral :: num => 'a" to represent
   202 positive and negated numeric literals, respectively.  See also
   203 definitions in ~~/src/HOL/Num.thy.  Potential INCOMPATIBILITY, some
   204 user theories may require adaptations as follows:
   205 
   206   - Theorems with number_ring or number_semiring constraints: These
   207     classes are gone; use comm_ring_1 or comm_semiring_1 instead.
   208 
   209   - Theories defining numeric types: Remove number, number_semiring,
   210     and number_ring instances. Defer all theorems about numerals until
   211     after classes one and semigroup_add have been instantiated.
   212 
   213   - Numeral-only simp rules: Replace each rule having a "number_of v"
   214     pattern with two copies, one for numeral and one for neg_numeral.
   215 
   216   - Theorems about subclasses of semiring_1 or ring_1: These classes
   217     automatically support numerals now, so more simp rules and
   218     simprocs may now apply within the proof.
   219 
   220   - Definitions and theorems using old constructors Pls/Min/Bit0/Bit1:
   221     Redefine using other integer operations.
   222 
   223 * Transfer: New package intended to generalize the existing
   224 "descending" method and related theorem attributes from the Quotient
   225 package.  (Not all functionality is implemented yet, but future
   226 development will focus on Transfer as an eventual replacement for the
   227 corresponding parts of the Quotient package.)
   228 
   229   - transfer_rule attribute: Maintains a collection of transfer rules,
   230     which relate constants at two different types. Transfer rules may
   231     relate different type instances of the same polymorphic constant,
   232     or they may relate an operation on a raw type to a corresponding
   233     operation on an abstract type (quotient or subtype). For example:
   234 
   235     ((A ===> B) ===> list_all2 A ===> list_all2 B) map map
   236     (cr_int ===> cr_int ===> cr_int) (%(x,y) (u,v). (x+u, y+v)) plus_int
   237 
   238   - transfer method: Replaces a subgoal on abstract types with an
   239     equivalent subgoal on the corresponding raw types. Constants are
   240     replaced with corresponding ones according to the transfer rules.
   241     Goals are generalized over all free variables by default; this is
   242     necessary for variables whose types change, but can be overridden
   243     for specific variables with e.g. "transfer fixing: x y z".  The
   244     variant transfer' method allows replacing a subgoal with one that
   245     is logically stronger (rather than equivalent).
   246 
   247   - relator_eq attribute: Collects identity laws for relators of
   248     various type constructors, e.g. "list_all2 (op =) = (op =)".  The
   249     transfer method uses these lemmas to infer transfer rules for
   250     non-polymorphic constants on the fly.
   251 
   252   - transfer_prover method: Assists with proving a transfer rule for a
   253     new constant, provided the constant is defined in terms of other
   254     constants that already have transfer rules. It should be applied
   255     after unfolding the constant definitions.
   256 
   257   - HOL/ex/Transfer_Int_Nat.thy: Example theory demonstrating transfer
   258     from type nat to type int.
   259 
   260 * Lifting: New package intended to generalize the quotient_definition
   261 facility of the Quotient package; designed to work with Transfer.
   262 
   263   - lift_definition command: Defines operations on an abstract type in
   264     terms of a corresponding operation on a representation
   265     type.  Example syntax:
   266 
   267     lift_definition dlist_insert :: "'a => 'a dlist => 'a dlist"
   268       is List.insert
   269 
   270     Users must discharge a respectfulness proof obligation when each
   271     constant is defined. (For a type copy, i.e. a typedef with UNIV,
   272     the proof is discharged automatically.) The obligation is
   273     presented in a user-friendly, readable form; a respectfulness
   274     theorem in the standard format and a transfer rule are generated
   275     by the package.
   276 
   277   - Integration with code_abstype: For typedefs (e.g. subtypes
   278     corresponding to a datatype invariant, such as dlist),
   279     lift_definition generates a code certificate theorem and sets up
   280     code generation for each constant.
   281 
   282   - setup_lifting command: Sets up the Lifting package to work with a
   283     user-defined type. The user must provide either a quotient theorem
   284     or a type_definition theorem.  The package configures transfer
   285     rules for equality and quantifiers on the type, and sets up the
   286     lift_definition command to work with the type.
   287 
   288   - Usage examples: See Quotient_Examples/Lift_DList.thy,
   289     Quotient_Examples/Lift_RBT.thy, Quotient_Examples/Lift_FSet.thy,
   290     Word/Word.thy and Library/Float.thy.
   291 
   292 * Quotient package:
   293 
   294   - The 'quotient_type' command now supports a 'morphisms' option with
   295     rep and abs functions, similar to typedef.
   296 
   297   - 'quotient_type' sets up new types to work with the Lifting and
   298     Transfer packages, as with 'setup_lifting'.
   299 
   300   - The 'quotient_definition' command now requires the user to prove a
   301     respectfulness property at the point where the constant is
   302     defined, similar to lift_definition; INCOMPATIBILITY.
   303 
   304   - Renamed predicate 'Quotient' to 'Quotient3', and renamed theorems
   305     accordingly, INCOMPATIBILITY.
   306 
   307 * New diagnostic command 'find_unused_assms' to find potentially
   308 superfluous assumptions in theorems using Quickcheck.
   309 
   310 * Quickcheck:
   311 
   312   - Quickcheck returns variable assignments as counterexamples, which
   313     allows to reveal the underspecification of functions under test.
   314     For example, refuting "hd xs = x", it presents the variable
   315     assignment xs = [] and x = a1 as a counterexample, assuming that
   316     any property is false whenever "hd []" occurs in it.
   317 
   318     These counterexample are marked as potentially spurious, as
   319     Quickcheck also returns "xs = []" as a counterexample to the
   320     obvious theorem "hd xs = hd xs".
   321 
   322     After finding a potentially spurious counterexample, Quickcheck
   323     continues searching for genuine ones.
   324 
   325     By default, Quickcheck shows potentially spurious and genuine
   326     counterexamples. The option "genuine_only" sets quickcheck to only
   327     show genuine counterexamples.
   328 
   329   - The command 'quickcheck_generator' creates random and exhaustive
   330     value generators for a given type and operations.
   331 
   332     It generates values by using the operations as if they were
   333     constructors of that type.
   334 
   335   - Support for multisets.
   336 
   337   - Added "use_subtype" options.
   338 
   339   - Added "quickcheck_locale" configuration to specify how to process
   340     conjectures in a locale context.
   341 
   342 * Nitpick: Fixed infinite loop caused by the 'peephole_optim' option
   343 and affecting 'rat' and 'real'.
   344 
   345 * Sledgehammer:
   346   - Integrated more tightly with SPASS, as described in the ITP 2012
   347     paper "More SPASS with Isabelle".
   348   - Made it try "smt" as a fallback if "metis" fails or times out.
   349   - Added support for the following provers: Alt-Ergo (via Why3 and
   350     TFF1), iProver, iProver-Eq.
   351   - Replaced remote E-SInE with remote Satallax in the default setup.
   352   - Sped up the minimizer.
   353   - Added "lam_trans", "uncurry_aliases", and "minimize" options.
   354   - Renamed "slicing" ("no_slicing") option to "slice" ("dont_slice").
   355   - Renamed "sound" option to "strict".
   356 
   357 * Metis: Added possibility to specify lambda translations scheme as a
   358 parenthesized argument (e.g., "by (metis (lifting) ...)").
   359 
   360 * SMT: Renamed "smt_fixed" option to "smt_read_only_certificates".
   361 
   362 * Command 'try0': Renamed from 'try_methods'. INCOMPATIBILITY.
   363 
   364 * New "case_product" attribute to generate a case rule doing multiple
   365 case distinctions at the same time.  E.g.
   366 
   367   list.exhaust [case_product nat.exhaust]
   368 
   369 produces a rule which can be used to perform case distinction on both
   370 a list and a nat.
   371 
   372 * New "eventually_elim" method as a generalized variant of the
   373 eventually_elim* rules.  Supports structured proofs.
   374 
   375 * Typedef with implicit set definition is considered legacy.  Use
   376 "typedef (open)" form instead, which will eventually become the
   377 default.
   378 
   379 * Record: code generation can be switched off manually with
   380 
   381   declare [[record_coden = false]]  -- "default true"
   382 
   383 * Datatype: type parameters allow explicit sort constraints.
   384 
   385 * Concrete syntax for case expressions includes constraints for source
   386 positions, and thus produces Prover IDE markup for its bindings.
   387 INCOMPATIBILITY for old-style syntax translations that augment the
   388 pattern notation; e.g. see src/HOL/HOLCF/One.thy for translations of
   389 one_case.
   390 
   391 * Clarified attribute "mono_set": pure declaration without modifying
   392 the result of the fact expression.
   393 
   394 * More default pred/set conversions on a couple of relation operations
   395 and predicates.  Added powers of predicate relations.  Consolidation
   396 of some relation theorems:
   397 
   398   converse_def ~> converse_unfold
   399   rel_comp_def ~> relcomp_unfold
   400   symp_def ~> (modified, use symp_def and sym_def instead)
   401   transp_def ~> transp_trans
   402   Domain_def ~> Domain_unfold
   403   Range_def ~> Domain_converse [symmetric]
   404 
   405 Generalized theorems INF_INT_eq, INF_INT_eq2, SUP_UN_eq, SUP_UN_eq2.
   406 
   407 See theory "Relation" for examples for making use of pred/set
   408 conversions by means of attributes "to_set" and "to_pred".
   409 
   410 INCOMPATIBILITY.
   411 
   412 * Renamed facts about the power operation on relations, i.e., relpow
   413 to match the constant's name:
   414 
   415   rel_pow_1 ~> relpow_1
   416   rel_pow_0_I ~> relpow_0_I
   417   rel_pow_Suc_I ~> relpow_Suc_I
   418   rel_pow_Suc_I2 ~> relpow_Suc_I2
   419   rel_pow_0_E ~> relpow_0_E
   420   rel_pow_Suc_E ~> relpow_Suc_E
   421   rel_pow_E ~> relpow_E
   422   rel_pow_Suc_D2 ~> relpow_Suc_D2
   423   rel_pow_Suc_E2 ~> relpow_Suc_E2
   424   rel_pow_Suc_D2' ~> relpow_Suc_D2'
   425   rel_pow_E2 ~> relpow_E2
   426   rel_pow_add ~> relpow_add
   427   rel_pow_commute ~> relpow
   428   rel_pow_empty ~> relpow_empty:
   429   rtrancl_imp_UN_rel_pow ~> rtrancl_imp_UN_relpow
   430   rel_pow_imp_rtrancl ~> relpow_imp_rtrancl
   431   rtrancl_is_UN_rel_pow ~> rtrancl_is_UN_relpow
   432   rtrancl_imp_rel_pow ~> rtrancl_imp_relpow
   433   rel_pow_fun_conv ~> relpow_fun_conv
   434   rel_pow_finite_bounded1 ~> relpow_finite_bounded1
   435   rel_pow_finite_bounded ~> relpow_finite_bounded
   436   rtrancl_finite_eq_rel_pow ~> rtrancl_finite_eq_relpow
   437   trancl_finite_eq_rel_pow ~> trancl_finite_eq_relpow
   438   single_valued_rel_pow ~> single_valued_relpow
   439 
   440 INCOMPATIBILITY.
   441 
   442 * Theory Relation: Consolidated constant name for relation composition
   443 and corresponding theorem names:
   444 
   445   - Renamed constant rel_comp to relcomp.
   446 
   447   - Dropped abbreviation pred_comp. Use relcompp instead.
   448 
   449   - Renamed theorems:
   450 
   451     rel_compI ~> relcompI
   452     rel_compEpair ~> relcompEpair
   453     rel_compE ~> relcompE
   454     pred_comp_rel_comp_eq ~> relcompp_relcomp_eq
   455     rel_comp_empty1 ~> relcomp_empty1
   456     rel_comp_mono ~> relcomp_mono
   457     rel_comp_subset_Sigma ~> relcomp_subset_Sigma
   458     rel_comp_distrib ~> relcomp_distrib
   459     rel_comp_distrib2 ~> relcomp_distrib2
   460     rel_comp_UNION_distrib ~> relcomp_UNION_distrib
   461     rel_comp_UNION_distrib2 ~> relcomp_UNION_distrib2
   462     single_valued_rel_comp ~> single_valued_relcomp
   463     rel_comp_def ~> relcomp_unfold
   464     converse_rel_comp ~> converse_relcomp
   465     pred_compI ~> relcomppI
   466     pred_compE ~> relcomppE
   467     pred_comp_bot1 ~> relcompp_bot1
   468     pred_comp_bot2 ~> relcompp_bot2
   469     transp_pred_comp_less_eq ~> transp_relcompp_less_eq
   470     pred_comp_mono ~> relcompp_mono
   471     pred_comp_distrib ~> relcompp_distrib
   472     pred_comp_distrib2 ~> relcompp_distrib2
   473     converse_pred_comp ~> converse_relcompp
   474 
   475     finite_rel_comp ~> finite_relcomp
   476 
   477     set_rel_comp ~> set_relcomp
   478 
   479 INCOMPATIBILITY.
   480 
   481 * Theory Divides: Discontinued redundant theorems about div and mod.
   482 INCOMPATIBILITY, use the corresponding generic theorems instead.
   483 
   484   DIVISION_BY_ZERO ~> div_by_0, mod_by_0
   485   zdiv_self ~> div_self
   486   zmod_self ~> mod_self
   487   zdiv_zero ~> div_0
   488   zmod_zero ~> mod_0
   489   zdiv_zmod_equality ~> div_mod_equality2
   490   zdiv_zmod_equality2 ~> div_mod_equality
   491   zmod_zdiv_trivial ~> mod_div_trivial
   492   zdiv_zminus_zminus ~> div_minus_minus
   493   zmod_zminus_zminus ~> mod_minus_minus
   494   zdiv_zminus2 ~> div_minus_right
   495   zmod_zminus2 ~> mod_minus_right
   496   zdiv_minus1_right ~> div_minus1_right
   497   zmod_minus1_right ~> mod_minus1_right
   498   zdvd_mult_div_cancel ~> dvd_mult_div_cancel
   499   zmod_zmult1_eq ~> mod_mult_right_eq
   500   zpower_zmod ~> power_mod
   501   zdvd_zmod ~> dvd_mod
   502   zdvd_zmod_imp_zdvd ~> dvd_mod_imp_dvd
   503   mod_mult_distrib ~> mult_mod_left
   504   mod_mult_distrib2 ~> mult_mod_right
   505 
   506 * Removed redundant theorems nat_mult_2 and nat_mult_2_right; use
   507 generic mult_2 and mult_2_right instead. INCOMPATIBILITY.
   508 
   509 * Finite_Set.fold now qualified.  INCOMPATIBILITY.
   510 
   511 * Consolidated theorem names concerning fold combinators:
   512 
   513   inf_INFI_fold_inf ~> inf_INF_fold_inf
   514   sup_SUPR_fold_sup ~> sup_SUP_fold_sup
   515   INFI_fold_inf ~> INF_fold_inf
   516   SUPR_fold_sup ~> SUP_fold_sup
   517   union_set ~> union_set_fold
   518   minus_set ~> minus_set_fold
   519   INFI_set_fold ~> INF_set_fold
   520   SUPR_set_fold ~> SUP_set_fold
   521   INF_code ~> INF_set_foldr
   522   SUP_code ~> SUP_set_foldr
   523   foldr.simps ~> foldr.simps (in point-free formulation)
   524   foldr_fold_rev ~> foldr_conv_fold
   525   foldl_fold ~> foldl_conv_fold
   526   foldr_foldr ~> foldr_conv_foldl
   527   foldl_foldr ~> foldl_conv_foldr
   528   fold_set_remdups ~> fold_set_fold_remdups
   529   fold_set ~> fold_set_fold
   530   fold1_set ~> fold1_set_fold
   531 
   532 INCOMPATIBILITY.
   533 
   534 * Dropped rarely useful theorems concerning fold combinators:
   535 foldl_apply, foldl_fun_comm, foldl_rev, fold_weak_invariant,
   536 rev_foldl_cons, fold_set_remdups, fold_set, fold_set1,
   537 concat_conv_foldl, foldl_weak_invariant, foldl_invariant,
   538 foldr_invariant, foldl_absorb0, foldl_foldr1_lemma, foldl_foldr1,
   539 listsum_conv_fold, listsum_foldl, sort_foldl_insort, foldl_assoc,
   540 foldr_conv_foldl, start_le_sum, elem_le_sum, sum_eq_0_conv.
   541 INCOMPATIBILITY.  For the common phrases "%xs. List.foldr plus xs 0"
   542 and "List.foldl plus 0", prefer "List.listsum".  Otherwise it can be
   543 useful to boil down "List.foldr" and "List.foldl" to "List.fold" by
   544 unfolding "foldr_conv_fold" and "foldl_conv_fold".
   545 
   546 * Dropped lemmas minus_set_foldr, union_set_foldr, union_coset_foldr,
   547 inter_coset_foldr, Inf_fin_set_foldr, Sup_fin_set_foldr,
   548 Min_fin_set_foldr, Max_fin_set_foldr, Inf_set_foldr, Sup_set_foldr,
   549 INF_set_foldr, SUP_set_foldr.  INCOMPATIBILITY.  Prefer corresponding
   550 lemmas over fold rather than foldr, or make use of lemmas
   551 fold_conv_foldr and fold_rev.
   552 
   553 * Congruence rules Option.map_cong and Option.bind_cong for recursion
   554 through option types.
   555 
   556 * "Transitive_Closure.ntrancl": bounded transitive closure on
   557 relations.
   558 
   559 * Constant "Set.not_member" now qualified.  INCOMPATIBILITY.
   560 
   561 * Theory Int: Discontinued many legacy theorems specific to type int.
   562 INCOMPATIBILITY, use the corresponding generic theorems instead.
   563 
   564   zminus_zminus ~> minus_minus
   565   zminus_0 ~> minus_zero
   566   zminus_zadd_distrib ~> minus_add_distrib
   567   zadd_commute ~> add_commute
   568   zadd_assoc ~> add_assoc
   569   zadd_left_commute ~> add_left_commute
   570   zadd_ac ~> add_ac
   571   zmult_ac ~> mult_ac
   572   zadd_0 ~> add_0_left
   573   zadd_0_right ~> add_0_right
   574   zadd_zminus_inverse2 ~> left_minus
   575   zmult_zminus ~> mult_minus_left
   576   zmult_commute ~> mult_commute
   577   zmult_assoc ~> mult_assoc
   578   zadd_zmult_distrib ~> left_distrib
   579   zadd_zmult_distrib2 ~> right_distrib
   580   zdiff_zmult_distrib ~> left_diff_distrib
   581   zdiff_zmult_distrib2 ~> right_diff_distrib
   582   zmult_1 ~> mult_1_left
   583   zmult_1_right ~> mult_1_right
   584   zle_refl ~> order_refl
   585   zle_trans ~> order_trans
   586   zle_antisym ~> order_antisym
   587   zle_linear ~> linorder_linear
   588   zless_linear ~> linorder_less_linear
   589   zadd_left_mono ~> add_left_mono
   590   zadd_strict_right_mono ~> add_strict_right_mono
   591   zadd_zless_mono ~> add_less_le_mono
   592   int_0_less_1 ~> zero_less_one
   593   int_0_neq_1 ~> zero_neq_one
   594   zless_le ~> less_le
   595   zpower_zadd_distrib ~> power_add
   596   zero_less_zpower_abs_iff ~> zero_less_power_abs_iff
   597   zero_le_zpower_abs ~> zero_le_power_abs
   598 
   599 * Theory Deriv: Renamed
   600 
   601   DERIV_nonneg_imp_nonincreasing ~> DERIV_nonneg_imp_nondecreasing
   602 
   603 * Theory Library/Multiset: Improved code generation of multisets.
   604 
   605 * Theory HOL/Library/Set_Algebras: Addition and multiplication on sets
   606 are expressed via type classes again. The special syntax
   607 \<oplus>/\<otimes> has been replaced by plain +/*. Removed constant
   608 setsum_set, which is now subsumed by Big_Operators.setsum.
   609 INCOMPATIBILITY.
   610 
   611 * Theory HOL/Library/Diagonalize has been removed. INCOMPATIBILITY,
   612 use theory HOL/Library/Nat_Bijection instead.
   613 
   614 * Theory HOL/Library/RBT_Impl: Backing implementation of red-black
   615 trees is now inside a type class context.  Names of affected
   616 operations and lemmas have been prefixed by rbt_.  INCOMPATIBILITY for
   617 theories working directly with raw red-black trees, adapt the names as
   618 follows:
   619 
   620   Operations:
   621   bulkload -> rbt_bulkload
   622   del_from_left -> rbt_del_from_left
   623   del_from_right -> rbt_del_from_right
   624   del -> rbt_del
   625   delete -> rbt_delete
   626   ins -> rbt_ins
   627   insert -> rbt_insert
   628   insertw -> rbt_insert_with
   629   insert_with_key -> rbt_insert_with_key
   630   map_entry -> rbt_map_entry
   631   lookup -> rbt_lookup
   632   sorted -> rbt_sorted
   633   tree_greater -> rbt_greater
   634   tree_less -> rbt_less
   635   tree_less_symbol -> rbt_less_symbol
   636   union -> rbt_union
   637   union_with -> rbt_union_with
   638   union_with_key -> rbt_union_with_key
   639 
   640   Lemmas:
   641   balance_left_sorted -> balance_left_rbt_sorted
   642   balance_left_tree_greater -> balance_left_rbt_greater
   643   balance_left_tree_less -> balance_left_rbt_less
   644   balance_right_sorted -> balance_right_rbt_sorted
   645   balance_right_tree_greater -> balance_right_rbt_greater
   646   balance_right_tree_less -> balance_right_rbt_less
   647   balance_sorted -> balance_rbt_sorted
   648   balance_tree_greater -> balance_rbt_greater
   649   balance_tree_less -> balance_rbt_less
   650   bulkload_is_rbt -> rbt_bulkload_is_rbt
   651   combine_sorted -> combine_rbt_sorted
   652   combine_tree_greater -> combine_rbt_greater
   653   combine_tree_less -> combine_rbt_less
   654   delete_in_tree -> rbt_delete_in_tree
   655   delete_is_rbt -> rbt_delete_is_rbt
   656   del_from_left_tree_greater -> rbt_del_from_left_rbt_greater
   657   del_from_left_tree_less -> rbt_del_from_left_rbt_less
   658   del_from_right_tree_greater -> rbt_del_from_right_rbt_greater
   659   del_from_right_tree_less -> rbt_del_from_right_rbt_less
   660   del_in_tree -> rbt_del_in_tree
   661   del_inv1_inv2 -> rbt_del_inv1_inv2
   662   del_sorted -> rbt_del_rbt_sorted
   663   del_tree_greater -> rbt_del_rbt_greater
   664   del_tree_less -> rbt_del_rbt_less
   665   dom_lookup_Branch -> dom_rbt_lookup_Branch
   666   entries_lookup -> entries_rbt_lookup
   667   finite_dom_lookup -> finite_dom_rbt_lookup
   668   insert_sorted -> rbt_insert_rbt_sorted
   669   insertw_is_rbt -> rbt_insertw_is_rbt
   670   insertwk_is_rbt -> rbt_insertwk_is_rbt
   671   insertwk_sorted -> rbt_insertwk_rbt_sorted
   672   insertw_sorted -> rbt_insertw_rbt_sorted
   673   ins_sorted -> ins_rbt_sorted
   674   ins_tree_greater -> ins_rbt_greater
   675   ins_tree_less -> ins_rbt_less
   676   is_rbt_sorted -> is_rbt_rbt_sorted
   677   lookup_balance -> rbt_lookup_balance
   678   lookup_bulkload -> rbt_lookup_rbt_bulkload
   679   lookup_delete -> rbt_lookup_rbt_delete
   680   lookup_Empty -> rbt_lookup_Empty
   681   lookup_from_in_tree -> rbt_lookup_from_in_tree
   682   lookup_in_tree -> rbt_lookup_in_tree
   683   lookup_ins -> rbt_lookup_ins
   684   lookup_insert -> rbt_lookup_rbt_insert
   685   lookup_insertw -> rbt_lookup_rbt_insertw
   686   lookup_insertwk -> rbt_lookup_rbt_insertwk
   687   lookup_keys -> rbt_lookup_keys
   688   lookup_map -> rbt_lookup_map
   689   lookup_map_entry -> rbt_lookup_rbt_map_entry
   690   lookup_tree_greater -> rbt_lookup_rbt_greater
   691   lookup_tree_less -> rbt_lookup_rbt_less
   692   lookup_union -> rbt_lookup_rbt_union
   693   map_entry_color_of -> rbt_map_entry_color_of
   694   map_entry_inv1 -> rbt_map_entry_inv1
   695   map_entry_inv2 -> rbt_map_entry_inv2
   696   map_entry_is_rbt -> rbt_map_entry_is_rbt
   697   map_entry_sorted -> rbt_map_entry_rbt_sorted
   698   map_entry_tree_greater -> rbt_map_entry_rbt_greater
   699   map_entry_tree_less -> rbt_map_entry_rbt_less
   700   map_tree_greater -> map_rbt_greater
   701   map_tree_less -> map_rbt_less
   702   map_sorted -> map_rbt_sorted
   703   paint_sorted -> paint_rbt_sorted
   704   paint_lookup -> paint_rbt_lookup
   705   paint_tree_greater -> paint_rbt_greater
   706   paint_tree_less -> paint_rbt_less
   707   sorted_entries -> rbt_sorted_entries
   708   tree_greater_eq_trans -> rbt_greater_eq_trans
   709   tree_greater_nit -> rbt_greater_nit
   710   tree_greater_prop -> rbt_greater_prop
   711   tree_greater_simps -> rbt_greater_simps
   712   tree_greater_trans -> rbt_greater_trans
   713   tree_less_eq_trans -> rbt_less_eq_trans
   714   tree_less_nit -> rbt_less_nit
   715   tree_less_prop -> rbt_less_prop
   716   tree_less_simps -> rbt_less_simps
   717   tree_less_trans -> rbt_less_trans
   718   tree_ord_props -> rbt_ord_props
   719   union_Branch -> rbt_union_Branch
   720   union_is_rbt -> rbt_union_is_rbt
   721   unionw_is_rbt -> rbt_unionw_is_rbt
   722   unionwk_is_rbt -> rbt_unionwk_is_rbt
   723   unionwk_sorted -> rbt_unionwk_rbt_sorted
   724 
   725 * Theory HOL/Library/Float: Floating point numbers are now defined as
   726 a subset of the real numbers.  All operations are defined using the
   727 lifing-framework and proofs use the transfer method.  INCOMPATIBILITY.
   728 
   729   Changed Operations:
   730   float_abs -> abs
   731   float_nprt -> nprt
   732   float_pprt -> pprt
   733   pow2 -> use powr
   734   round_down -> float_round_down
   735   round_up -> float_round_up
   736   scale -> exponent
   737 
   738   Removed Operations:
   739   ceiling_fl, lb_mult, lb_mod, ub_mult, ub_mod
   740 
   741   Renamed Lemmas:
   742   abs_float_def -> Float.compute_float_abs
   743   bitlen_ge0 -> bitlen_nonneg
   744   bitlen.simps -> Float.compute_bitlen
   745   float_components -> Float_mantissa_exponent
   746   float_divl.simps -> Float.compute_float_divl
   747   float_divr.simps -> Float.compute_float_divr
   748   float_eq_odd -> mult_powr_eq_mult_powr_iff
   749   float_power -> real_of_float_power
   750   lapprox_posrat_def -> Float.compute_lapprox_posrat
   751   lapprox_rat.simps -> Float.compute_lapprox_rat
   752   le_float_def' -> Float.compute_float_le
   753   le_float_def -> less_eq_float.rep_eq
   754   less_float_def' -> Float.compute_float_less
   755   less_float_def -> less_float.rep_eq
   756   normfloat_def -> Float.compute_normfloat
   757   normfloat_imp_odd_or_zero -> mantissa_not_dvd and mantissa_noteq_0
   758   normfloat -> normfloat_def
   759   normfloat_unique -> use normfloat_def
   760   number_of_float_Float -> Float.compute_float_numeral, Float.compute_float_neg_numeral
   761   one_float_def -> Float.compute_float_one
   762   plus_float_def -> Float.compute_float_plus
   763   rapprox_posrat_def -> Float.compute_rapprox_posrat
   764   rapprox_rat.simps -> Float.compute_rapprox_rat
   765   real_of_float_0 -> zero_float.rep_eq
   766   real_of_float_1 -> one_float.rep_eq
   767   real_of_float_abs -> abs_float.rep_eq
   768   real_of_float_add -> plus_float.rep_eq
   769   real_of_float_minus -> uminus_float.rep_eq
   770   real_of_float_mult -> times_float.rep_eq
   771   real_of_float_simp -> Float.rep_eq
   772   real_of_float_sub -> minus_float.rep_eq
   773   round_down.simps -> Float.compute_float_round_down
   774   round_up.simps -> Float.compute_float_round_up
   775   times_float_def -> Float.compute_float_times
   776   uminus_float_def -> Float.compute_float_uminus
   777   zero_float_def -> Float.compute_float_zero
   778 
   779   Lemmas not necessary anymore, use the transfer method:
   780   bitlen_B0, bitlen_B1, bitlen_ge1, bitlen_Min, bitlen_Pls, float_divl,
   781   float_divr, float_le_simp, float_less1_mantissa_bound,
   782   float_less_simp, float_less_zero, float_le_zero,
   783   float_pos_less1_e_neg, float_pos_m_pos, float_split, float_split2,
   784   floor_pos_exp, lapprox_posrat, lapprox_posrat_bottom, lapprox_rat,
   785   lapprox_rat_bottom, normalized_float, rapprox_posrat,
   786   rapprox_posrat_le1, rapprox_rat, real_of_float_ge0_exp,
   787   real_of_float_neg_exp, real_of_float_nge0_exp, round_down floor_fl,
   788   round_up, zero_le_float, zero_less_float
   789 
   790 * New theory HOL/Library/DAList provides an abstract type for
   791 association lists with distinct keys.
   792 
   793 * Session HOL/IMP: Added new theory of abstract interpretation of
   794 annotated commands.
   795 
   796 * Session HOL-Import: Re-implementation from scratch is faster,
   797 simpler, and more scalable.  Requires a proof bundle, which is
   798 available as an external component.  Discontinued old (and mostly
   799 dead) Importer for HOL4 and HOL Light.  INCOMPATIBILITY.
   800 
   801 * Session HOL-Word: Discontinued many redundant theorems specific to
   802 type 'a word. INCOMPATIBILITY, use the corresponding generic theorems
   803 instead.
   804 
   805   word_sub_alt ~> word_sub_wi
   806   word_add_alt ~> word_add_def
   807   word_mult_alt ~> word_mult_def
   808   word_minus_alt ~> word_minus_def
   809   word_0_alt ~> word_0_wi
   810   word_1_alt ~> word_1_wi
   811   word_add_0 ~> add_0_left
   812   word_add_0_right ~> add_0_right
   813   word_mult_1 ~> mult_1_left
   814   word_mult_1_right ~> mult_1_right
   815   word_add_commute ~> add_commute
   816   word_add_assoc ~> add_assoc
   817   word_add_left_commute ~> add_left_commute
   818   word_mult_commute ~> mult_commute
   819   word_mult_assoc ~> mult_assoc
   820   word_mult_left_commute ~> mult_left_commute
   821   word_left_distrib ~> left_distrib
   822   word_right_distrib ~> right_distrib
   823   word_left_minus ~> left_minus
   824   word_diff_0_right ~> diff_0_right
   825   word_diff_self ~> diff_self
   826   word_sub_def ~> diff_minus
   827   word_diff_minus ~> diff_minus
   828   word_add_ac ~> add_ac
   829   word_mult_ac ~> mult_ac
   830   word_plus_ac0 ~> add_0_left add_0_right add_ac
   831   word_times_ac1 ~> mult_1_left mult_1_right mult_ac
   832   word_order_trans ~> order_trans
   833   word_order_refl ~> order_refl
   834   word_order_antisym ~> order_antisym
   835   word_order_linear ~> linorder_linear
   836   lenw1_zero_neq_one ~> zero_neq_one
   837   word_number_of_eq ~> number_of_eq
   838   word_of_int_add_hom ~> wi_hom_add
   839   word_of_int_sub_hom ~> wi_hom_sub
   840   word_of_int_mult_hom ~> wi_hom_mult
   841   word_of_int_minus_hom ~> wi_hom_neg
   842   word_of_int_succ_hom ~> wi_hom_succ
   843   word_of_int_pred_hom ~> wi_hom_pred
   844   word_of_int_0_hom ~> word_0_wi
   845   word_of_int_1_hom ~> word_1_wi
   846 
   847 * Session HOL-Word: New proof method "word_bitwise" for splitting
   848 machine word equalities and inequalities into logical circuits,
   849 defined in HOL/Word/WordBitwise.thy.  Supports addition, subtraction,
   850 multiplication, shifting by constants, bitwise operators and numeric
   851 constants.  Requires fixed-length word types, not 'a word.  Solves
   852 many standard word identities outright and converts more into first
   853 order problems amenable to blast or similar.  See also examples in
   854 HOL/Word/Examples/WordExamples.thy.
   855 
   856 * Session HOL-Probability: Introduced the type "'a measure" to
   857 represent measures, this replaces the records 'a algebra and 'a
   858 measure_space.  The locales based on subset_class now have two
   859 locale-parameters the space \<Omega> and the set of measurable sets M.
   860 The product of probability spaces uses now the same constant as the
   861 finite product of sigma-finite measure spaces "PiM :: ('i => 'a)
   862 measure".  Most constants are defined now outside of locales and gain
   863 an additional parameter, like null_sets, almost_eventually or \<mu>'.
   864 Measure space constructions for distributions and densities now got
   865 their own constants distr and density.  Instead of using locales to
   866 describe measure spaces with a finite space, the measure count_space
   867 and point_measure is introduced.  INCOMPATIBILITY.
   868 
   869   Renamed constants:
   870   measure -> emeasure
   871   finite_measure.\<mu>' -> measure
   872   product_algebra_generator -> prod_algebra
   873   product_prob_space.emb -> prod_emb
   874   product_prob_space.infprod_algebra -> PiM
   875 
   876   Removed locales:
   877   completeable_measure_space
   878   finite_measure_space
   879   finite_prob_space
   880   finite_product_finite_prob_space
   881   finite_product_sigma_algebra
   882   finite_sigma_algebra
   883   measure_space
   884   pair_finite_prob_space
   885   pair_finite_sigma_algebra
   886   pair_finite_space
   887   pair_sigma_algebra
   888   product_sigma_algebra
   889 
   890   Removed constants:
   891   conditional_space
   892   distribution -> use distr measure, or distributed predicate
   893   image_space
   894   joint_distribution -> use distr measure, or distributed predicate
   895   pair_measure_generator
   896   product_prob_space.infprod_algebra -> use PiM
   897   subvimage
   898 
   899   Replacement theorems:
   900   finite_additivity_sufficient -> ring_of_sets.countably_additiveI_finite
   901   finite_measure.empty_measure -> measure_empty
   902   finite_measure.finite_continuity_from_above -> finite_measure.finite_Lim_measure_decseq
   903   finite_measure.finite_continuity_from_below -> finite_measure.finite_Lim_measure_incseq
   904   finite_measure.finite_measure_countably_subadditive -> finite_measure.finite_measure_subadditive_countably
   905   finite_measure.finite_measure_eq -> finite_measure.emeasure_eq_measure
   906   finite_measure.finite_measure -> finite_measure.emeasure_finite
   907   finite_measure.finite_measure_finite_singleton -> finite_measure.finite_measure_eq_setsum_singleton
   908   finite_measure.positive_measure' -> measure_nonneg
   909   finite_measure.real_measure -> finite_measure.emeasure_real
   910   finite_product_prob_space.finite_measure_times -> finite_product_prob_space.finite_measure_PiM_emb
   911   finite_product_sigma_algebra.in_P -> sets_PiM_I_finite
   912   finite_product_sigma_algebra.P_empty -> space_PiM_empty, sets_PiM_empty
   913   information_space.conditional_entropy_eq -> information_space.conditional_entropy_simple_distributed
   914   information_space.conditional_entropy_positive -> information_space.conditional_entropy_nonneg_simple
   915   information_space.conditional_mutual_information_eq_mutual_information -> information_space.conditional_mutual_information_eq_mutual_information_simple
   916   information_space.conditional_mutual_information_generic_positive -> information_space.conditional_mutual_information_nonneg_simple
   917   information_space.conditional_mutual_information_positive -> information_space.conditional_mutual_information_nonneg_simple
   918   information_space.entropy_commute -> information_space.entropy_commute_simple
   919   information_space.entropy_eq -> information_space.entropy_simple_distributed
   920   information_space.entropy_generic_eq -> information_space.entropy_simple_distributed
   921   information_space.entropy_positive -> information_space.entropy_nonneg_simple
   922   information_space.entropy_uniform_max -> information_space.entropy_uniform
   923   information_space.KL_eq_0_imp -> information_space.KL_eq_0_iff_eq
   924   information_space.KL_eq_0 -> information_space.KL_same_eq_0
   925   information_space.KL_ge_0 -> information_space.KL_nonneg
   926   information_space.mutual_information_eq -> information_space.mutual_information_simple_distributed
   927   information_space.mutual_information_positive -> information_space.mutual_information_nonneg_simple
   928   Int_stable_cuboids -> Int_stable_atLeastAtMost
   929   Int_stable_product_algebra_generator -> positive_integral
   930   measure_preserving -> equality "distr M N f = N" "f : measurable M N"
   931   measure_space.additive -> emeasure_additive
   932   measure_space.AE_iff_null_set -> AE_iff_null
   933   measure_space.almost_everywhere_def -> eventually_ae_filter
   934   measure_space.almost_everywhere_vimage -> AE_distrD
   935   measure_space.continuity_from_above -> INF_emeasure_decseq
   936   measure_space.continuity_from_above_Lim -> Lim_emeasure_decseq
   937   measure_space.continuity_from_below_Lim -> Lim_emeasure_incseq
   938   measure_space.continuity_from_below -> SUP_emeasure_incseq
   939   measure_space_density -> emeasure_density
   940   measure_space.density_is_absolutely_continuous -> absolutely_continuousI_density
   941   measure_space.integrable_vimage -> integrable_distr
   942   measure_space.integral_translated_density -> integral_density
   943   measure_space.integral_vimage -> integral_distr
   944   measure_space.measure_additive -> plus_emeasure
   945   measure_space.measure_compl -> emeasure_compl
   946   measure_space.measure_countable_increasing -> emeasure_countable_increasing
   947   measure_space.measure_countably_subadditive -> emeasure_subadditive_countably
   948   measure_space.measure_decseq -> decseq_emeasure
   949   measure_space.measure_Diff -> emeasure_Diff
   950   measure_space.measure_Diff_null_set -> emeasure_Diff_null_set
   951   measure_space.measure_eq_0 -> emeasure_eq_0
   952   measure_space.measure_finitely_subadditive -> emeasure_subadditive_finite
   953   measure_space.measure_finite_singleton -> emeasure_eq_setsum_singleton
   954   measure_space.measure_incseq -> incseq_emeasure
   955   measure_space.measure_insert -> emeasure_insert
   956   measure_space.measure_mono -> emeasure_mono
   957   measure_space.measure_not_negative -> emeasure_not_MInf
   958   measure_space.measure_preserving_Int_stable -> measure_eqI_generator_eq
   959   measure_space.measure_setsum -> setsum_emeasure
   960   measure_space.measure_setsum_split -> setsum_emeasure_cover
   961   measure_space.measure_space_vimage -> emeasure_distr
   962   measure_space.measure_subadditive_finite -> emeasure_subadditive_finite
   963   measure_space.measure_subadditive -> subadditive
   964   measure_space.measure_top -> emeasure_space
   965   measure_space.measure_UN_eq_0 -> emeasure_UN_eq_0
   966   measure_space.measure_Un_null_set -> emeasure_Un_null_set
   967   measure_space.positive_integral_translated_density -> positive_integral_density
   968   measure_space.positive_integral_vimage -> positive_integral_distr
   969   measure_space.real_continuity_from_above -> Lim_measure_decseq
   970   measure_space.real_continuity_from_below -> Lim_measure_incseq
   971   measure_space.real_measure_countably_subadditive -> measure_subadditive_countably
   972   measure_space.real_measure_Diff -> measure_Diff
   973   measure_space.real_measure_finite_Union -> measure_finite_Union
   974   measure_space.real_measure_setsum_singleton -> measure_eq_setsum_singleton
   975   measure_space.real_measure_subadditive -> measure_subadditive
   976   measure_space.real_measure_Union -> measure_Union
   977   measure_space.real_measure_UNION -> measure_UNION
   978   measure_space.simple_function_vimage -> simple_function_comp
   979   measure_space.simple_integral_vimage -> simple_integral_distr
   980   measure_space.simple_integral_vimage -> simple_integral_distr
   981   measure_unique_Int_stable -> measure_eqI_generator_eq
   982   measure_unique_Int_stable_vimage -> measure_eqI_generator_eq
   983   pair_sigma_algebra.measurable_cut_fst -> sets_Pair1
   984   pair_sigma_algebra.measurable_cut_snd -> sets_Pair2
   985   pair_sigma_algebra.measurable_pair_image_fst -> measurable_Pair1
   986   pair_sigma_algebra.measurable_pair_image_snd -> measurable_Pair2
   987   pair_sigma_algebra.measurable_product_swap -> measurable_pair_swap_iff
   988   pair_sigma_algebra.pair_sigma_algebra_measurable -> measurable_pair_swap
   989   pair_sigma_algebra.pair_sigma_algebra_swap_measurable -> measurable_pair_swap'
   990   pair_sigma_algebra.sets_swap -> sets_pair_swap
   991   pair_sigma_finite.measure_cut_measurable_fst -> pair_sigma_finite.measurable_emeasure_Pair1
   992   pair_sigma_finite.measure_cut_measurable_snd -> pair_sigma_finite.measurable_emeasure_Pair2
   993   pair_sigma_finite.measure_preserving_swap -> pair_sigma_finite.distr_pair_swap
   994   pair_sigma_finite.pair_measure_alt2 -> pair_sigma_finite.emeasure_pair_measure_alt2
   995   pair_sigma_finite.pair_measure_alt -> pair_sigma_finite.emeasure_pair_measure_alt
   996   pair_sigma_finite.pair_measure_times -> pair_sigma_finite.emeasure_pair_measure_Times
   997   prob_space.indep_distribution_eq_measure -> prob_space.indep_vars_iff_distr_eq_PiM
   998   prob_space.indep_var_distributionD -> prob_space.indep_var_distribution_eq
   999   prob_space.measure_space_1 -> prob_space.emeasure_space_1
  1000   prob_space.prob_space_vimage -> prob_space_distr
  1001   prob_space.random_variable_restrict -> measurable_restrict
  1002   prob_space_unique_Int_stable -> measure_eqI_prob_space
  1003   product_algebraE -> prod_algebraE_all
  1004   product_algebra_generator_der -> prod_algebra_eq_finite
  1005   product_algebra_generator_into_space -> prod_algebra_sets_into_space
  1006   product_algebraI -> sets_PiM_I_finite
  1007   product_measure_exists -> product_sigma_finite.sigma_finite
  1008   product_prob_space.finite_index_eq_finite_product -> product_prob_space.sets_PiM_generator
  1009   product_prob_space.finite_measure_infprod_emb_Pi -> product_prob_space.measure_PiM_emb
  1010   product_prob_space.infprod_spec -> product_prob_space.emeasure_PiM_emb_not_empty
  1011   product_prob_space.measurable_component -> measurable_component_singleton
  1012   product_prob_space.measurable_emb -> measurable_prod_emb
  1013   product_prob_space.measurable_into_infprod_algebra -> measurable_PiM_single
  1014   product_prob_space.measurable_singleton_infprod -> measurable_component_singleton
  1015   product_prob_space.measure_emb -> emeasure_prod_emb
  1016   product_prob_space.measure_preserving_restrict -> product_prob_space.distr_restrict
  1017   product_sigma_algebra.product_algebra_into_space -> space_closed
  1018   product_sigma_finite.measure_fold -> product_sigma_finite.distr_merge
  1019   product_sigma_finite.measure_preserving_component_singelton -> product_sigma_finite.distr_singleton
  1020   product_sigma_finite.measure_preserving_merge -> product_sigma_finite.distr_merge
  1021   sequence_space.measure_infprod -> sequence_space.measure_PiM_countable
  1022   sets_product_algebra -> sets_PiM
  1023   sigma_algebra.measurable_sigma -> measurable_measure_of
  1024   sigma_finite_measure.disjoint_sigma_finite -> sigma_finite_disjoint
  1025   sigma_finite_measure.RN_deriv_vimage -> sigma_finite_measure.RN_deriv_distr
  1026   sigma_product_algebra_sigma_eq -> sigma_prod_algebra_sigma_eq
  1027   space_product_algebra -> space_PiM
  1028 
  1029 * Session HOL-TPTP: support to parse and import TPTP problems (all
  1030 languages) into Isabelle/HOL.
  1031 
  1032 
  1033 *** FOL ***
  1034 
  1035 * New "case_product" attribute (see HOL).
  1036 
  1037 
  1038 *** ZF ***
  1039 
  1040 * Greater support for structured proofs involving induction or case
  1041 analysis.
  1042 
  1043 * Much greater use of mathematical symbols.
  1044 
  1045 * Removal of many ML theorem bindings.  INCOMPATIBILITY.
  1046 
  1047 
  1048 *** ML ***
  1049 
  1050 * Antiquotation @{keyword "name"} produces a parser for outer syntax
  1051 from a minor keyword introduced via theory header declaration.
  1052 
  1053 * Antiquotation @{command_spec "name"} produces the
  1054 Outer_Syntax.command_spec from a major keyword introduced via theory
  1055 header declaration; it can be passed to Outer_Syntax.command etc.
  1056 
  1057 * Local_Theory.define no longer hard-wires default theorem name
  1058 "foo_def", but retains the binding as given.  If that is Binding.empty
  1059 / Attrib.empty_binding, the result is not registered as user-level
  1060 fact.  The Local_Theory.define_internal variant allows to specify a
  1061 non-empty name (used for the foundation in the background theory),
  1062 while omitting the fact binding in the user-context.  Potential
  1063 INCOMPATIBILITY for derived definitional packages: need to specify
  1064 naming policy for primitive definitions more explicitly.
  1065 
  1066 * Renamed Thm.capply to Thm.apply, and Thm.cabs to Thm.lambda in
  1067 conformance with similar operations in structure Term and Logic.
  1068 
  1069 * Antiquotation @{attributes [...]} embeds attribute source
  1070 representation into the ML text, which is particularly useful with
  1071 declarations like Local_Theory.note.
  1072 
  1073 * Structure Proof_Context follows standard naming scheme.  Old
  1074 ProofContext has been discontinued.  INCOMPATIBILITY.
  1075 
  1076 * Refined Local_Theory.declaration {syntax, pervasive}, with subtle
  1077 change of semantics: update is applied to auxiliary local theory
  1078 context as well.
  1079 
  1080 * Modernized some old-style infix operations:
  1081 
  1082   addeqcongs    ~> Simplifier.add_eqcong
  1083   deleqcongs    ~> Simplifier.del_eqcong
  1084   addcongs      ~> Simplifier.add_cong
  1085   delcongs      ~> Simplifier.del_cong
  1086   setmksimps    ~> Simplifier.set_mksimps
  1087   setmkcong     ~> Simplifier.set_mkcong
  1088   setmksym      ~> Simplifier.set_mksym
  1089   setmkeqTrue   ~> Simplifier.set_mkeqTrue
  1090   settermless   ~> Simplifier.set_termless
  1091   setsubgoaler  ~> Simplifier.set_subgoaler
  1092   addsplits     ~> Splitter.add_split
  1093   delsplits     ~> Splitter.del_split
  1094 
  1095 
  1096 *** System ***
  1097 
  1098 * USER_HOME settings variable points to cross-platform user home
  1099 directory, which coincides with HOME on POSIX systems only.  Likewise,
  1100 the Isabelle path specification "~" now expands to $USER_HOME, instead
  1101 of former $HOME.  A different default for USER_HOME may be set
  1102 explicitly in shell environment, before Isabelle settings are
  1103 evaluated.  Minor INCOMPATIBILITY: need to adapt Isabelle path where
  1104 the generic user home was intended.
  1105 
  1106 * ISABELLE_HOME_WINDOWS refers to ISABELLE_HOME in windows file name
  1107 notation, which is useful for the jEdit file browser, for example.
  1108 
  1109 * ISABELLE_JDK_HOME settings variable points to JDK with javac and jar
  1110 (not just JRE).
  1111 
  1112 
  1113 
  1114 New in Isabelle2011-1 (October 2011)
  1115 ------------------------------------
  1116 
  1117 *** General ***
  1118 
  1119 * Improved Isabelle/jEdit Prover IDE (PIDE), which can be invoked as
  1120 "isabelle jedit" or "ISABELLE_HOME/Isabelle" on the command line.
  1121 
  1122   - Management of multiple theory files directly from the editor
  1123     buffer store -- bypassing the file-system (no requirement to save
  1124     files for checking).
  1125 
  1126   - Markup of formal entities within the text buffer, with semantic
  1127     highlighting, tooltips and hyperlinks to jump to defining source
  1128     positions.
  1129 
  1130   - Improved text rendering, with sub/superscripts in the source
  1131     buffer (including support for copy/paste wrt. output panel, HTML
  1132     theory output and other non-Isabelle text boxes).
  1133 
  1134   - Refined scheduling of proof checking and printing of results,
  1135     based on interactive editor view.  (Note: jEdit folding and
  1136     narrowing allows to restrict buffer perspectives explicitly.)
  1137 
  1138   - Reduced CPU performance requirements, usable on machines with few
  1139     cores.
  1140 
  1141   - Reduced memory requirements due to pruning of unused document
  1142     versions (garbage collection).
  1143 
  1144 See also ~~/src/Tools/jEdit/README.html for further information,
  1145 including some remaining limitations.
  1146 
  1147 * Theory loader: source files are exclusively located via the master
  1148 directory of each theory node (where the .thy file itself resides).
  1149 The global load path (such as src/HOL/Library) has been discontinued.
  1150 Note that the path element ~~ may be used to reference theories in the
  1151 Isabelle home folder -- for instance, "~~/src/HOL/Library/FuncSet".
  1152 INCOMPATIBILITY.
  1153 
  1154 * Theory loader: source files are identified by content via SHA1
  1155 digests.  Discontinued former path/modtime identification and optional
  1156 ISABELLE_FILE_IDENT plugin scripts.
  1157 
  1158 * Parallelization of nested Isar proofs is subject to
  1159 Goal.parallel_proofs_threshold (default 100).  See also isabelle
  1160 usedir option -Q.
  1161 
  1162 * Name space: former unsynchronized references are now proper
  1163 configuration options, with more conventional names:
  1164 
  1165   long_names   ~> names_long
  1166   short_names  ~> names_short
  1167   unique_names ~> names_unique
  1168 
  1169 Minor INCOMPATIBILITY, need to declare options in context like this:
  1170 
  1171   declare [[names_unique = false]]
  1172 
  1173 * Literal facts `prop` may contain dummy patterns, e.g. `_ = _`.  Note
  1174 that the result needs to be unique, which means fact specifications
  1175 may have to be refined after enriching a proof context.
  1176 
  1177 * Attribute "case_names" has been refined: the assumptions in each case
  1178 can be named now by following the case name with [name1 name2 ...].
  1179 
  1180 * Isabelle/Isar reference manual has been updated and extended:
  1181   - "Synopsis" provides a catalog of main Isar language concepts.
  1182   - Formal references in syntax diagrams, via @{rail} antiquotation.
  1183   - Updated material from classic "ref" manual, notably about
  1184     "Classical Reasoner".
  1185 
  1186 
  1187 *** HOL ***
  1188 
  1189 * Class bot and top require underlying partial order rather than
  1190 preorder: uniqueness of bot and top is guaranteed.  INCOMPATIBILITY.
  1191 
  1192 * Class complete_lattice: generalized a couple of lemmas from sets;
  1193 generalized theorems INF_cong and SUP_cong.  New type classes for
  1194 complete boolean algebras and complete linear orders.  Lemmas
  1195 Inf_less_iff, less_Sup_iff, INF_less_iff, less_SUP_iff now reside in
  1196 class complete_linorder.
  1197 
  1198 Changed proposition of lemmas Inf_bool_def, Sup_bool_def, Inf_fun_def,
  1199 Sup_fun_def, Inf_apply, Sup_apply.
  1200 
  1201 Removed redundant lemmas (the right hand side gives hints how to
  1202 replace them for (metis ...), or (simp only: ...) proofs):
  1203 
  1204   Inf_singleton ~> Inf_insert [where A="{}", unfolded Inf_empty inf_top_right]
  1205   Sup_singleton ~> Sup_insert [where A="{}", unfolded Sup_empty sup_bot_right]
  1206   Inf_binary ~> Inf_insert, Inf_empty, and inf_top_right
  1207   Sup_binary ~> Sup_insert, Sup_empty, and sup_bot_right
  1208   Int_eq_Inter ~> Inf_insert, Inf_empty, and inf_top_right
  1209   Un_eq_Union ~> Sup_insert, Sup_empty, and sup_bot_right
  1210   Inter_def ~> INF_def, image_def
  1211   Union_def ~> SUP_def, image_def
  1212   INT_eq ~> INF_def, and image_def
  1213   UN_eq ~> SUP_def, and image_def
  1214   INF_subset ~> INF_superset_mono [OF _ order_refl]
  1215 
  1216 More consistent and comprehensive names:
  1217 
  1218   INTER_eq_Inter_image ~> INF_def
  1219   UNION_eq_Union_image ~> SUP_def
  1220   INFI_def ~> INF_def
  1221   SUPR_def ~> SUP_def
  1222   INF_leI ~> INF_lower
  1223   INF_leI2 ~> INF_lower2
  1224   le_INFI ~> INF_greatest
  1225   le_SUPI ~> SUP_upper
  1226   le_SUPI2 ~> SUP_upper2
  1227   SUP_leI ~> SUP_least
  1228   INFI_bool_eq ~> INF_bool_eq
  1229   SUPR_bool_eq ~> SUP_bool_eq
  1230   INFI_apply ~> INF_apply
  1231   SUPR_apply ~> SUP_apply
  1232   INTER_def ~> INTER_eq
  1233   UNION_def ~> UNION_eq
  1234 
  1235 INCOMPATIBILITY.
  1236 
  1237 * Renamed theory Complete_Lattice to Complete_Lattices.
  1238 INCOMPATIBILITY.
  1239 
  1240 * Theory Complete_Lattices: lemmas Inf_eq_top_iff, INF_eq_top_iff,
  1241 INF_image, Inf_insert, INF_top, Inf_top_conv, INF_top_conv, SUP_bot,
  1242 Sup_bot_conv, SUP_bot_conv, Sup_eq_top_iff, SUP_eq_top_iff, SUP_image,
  1243 Sup_insert are now declared as [simp].  INCOMPATIBILITY.
  1244 
  1245 * Theory Lattice: lemmas compl_inf_bot, compl_le_comp_iff,
  1246 compl_sup_top, inf_idem, inf_left_idem, inf_sup_absorb, sup_idem,
  1247 sup_inf_absob, sup_left_idem are now declared as [simp].  Minor
  1248 INCOMPATIBILITY.
  1249 
  1250 * Added syntactic classes "inf" and "sup" for the respective
  1251 constants.  INCOMPATIBILITY: Changes in the argument order of the
  1252 (mostly internal) locale predicates for some derived classes.
  1253 
  1254 * Theorem collections ball_simps and bex_simps do not contain theorems
  1255 referring to UNION any longer; these have been moved to collection
  1256 UN_ball_bex_simps.  INCOMPATIBILITY.
  1257 
  1258 * Theory Archimedean_Field: floor now is defined as parameter of a
  1259 separate type class floor_ceiling.
  1260 
  1261 * Theory Finite_Set: more coherent development of fold_set locales:
  1262 
  1263     locale fun_left_comm ~> locale comp_fun_commute
  1264     locale fun_left_comm_idem ~> locale comp_fun_idem
  1265 
  1266 Both use point-free characterization; interpretation proofs may need
  1267 adjustment.  INCOMPATIBILITY.
  1268 
  1269 * Theory Limits: Type "'a net" has been renamed to "'a filter", in
  1270 accordance with standard mathematical terminology. INCOMPATIBILITY.
  1271 
  1272 * Theory Complex_Main: The locale interpretations for the
  1273 bounded_linear and bounded_bilinear locales have been removed, in
  1274 order to reduce the number of duplicate lemmas. Users must use the
  1275 original names for distributivity theorems, potential INCOMPATIBILITY.
  1276 
  1277   divide.add ~> add_divide_distrib
  1278   divide.diff ~> diff_divide_distrib
  1279   divide.setsum ~> setsum_divide_distrib
  1280   mult.add_right ~> right_distrib
  1281   mult.diff_right ~> right_diff_distrib
  1282   mult_right.setsum ~> setsum_right_distrib
  1283   mult_left.diff ~> left_diff_distrib
  1284 
  1285 * Theory Complex_Main: Several redundant theorems have been removed or
  1286 replaced by more general versions. INCOMPATIBILITY.
  1287 
  1288   real_diff_def ~> minus_real_def
  1289   real_divide_def ~> divide_real_def
  1290   real_less_def ~> less_le
  1291   real_abs_def ~> abs_real_def
  1292   real_sgn_def ~> sgn_real_def
  1293   real_mult_commute ~> mult_commute
  1294   real_mult_assoc ~> mult_assoc
  1295   real_mult_1 ~> mult_1_left
  1296   real_add_mult_distrib ~> left_distrib
  1297   real_zero_not_eq_one ~> zero_neq_one
  1298   real_mult_inverse_left ~> left_inverse
  1299   INVERSE_ZERO ~> inverse_zero
  1300   real_le_refl ~> order_refl
  1301   real_le_antisym ~> order_antisym
  1302   real_le_trans ~> order_trans
  1303   real_le_linear ~> linear
  1304   real_le_eq_diff ~> le_iff_diff_le_0
  1305   real_add_left_mono ~> add_left_mono
  1306   real_mult_order ~> mult_pos_pos
  1307   real_mult_less_mono2 ~> mult_strict_left_mono
  1308   real_of_int_real_of_nat ~> real_of_int_of_nat_eq
  1309   real_0_le_divide_iff ~> zero_le_divide_iff
  1310   realpow_two_disj ~> power2_eq_iff
  1311   real_squared_diff_one_factored ~> square_diff_one_factored
  1312   realpow_two_diff ~> square_diff_square_factored
  1313   reals_complete2 ~> complete_real
  1314   real_sum_squared_expand ~> power2_sum
  1315   exp_ln_eq ~> ln_unique
  1316   expi_add ~> exp_add
  1317   expi_zero ~> exp_zero
  1318   lemma_DERIV_subst ~> DERIV_cong
  1319   LIMSEQ_Zfun_iff ~> tendsto_Zfun_iff
  1320   LIMSEQ_const ~> tendsto_const
  1321   LIMSEQ_norm ~> tendsto_norm
  1322   LIMSEQ_add ~> tendsto_add
  1323   LIMSEQ_minus ~> tendsto_minus
  1324   LIMSEQ_minus_cancel ~> tendsto_minus_cancel
  1325   LIMSEQ_diff ~> tendsto_diff
  1326   bounded_linear.LIMSEQ ~> bounded_linear.tendsto
  1327   bounded_bilinear.LIMSEQ ~> bounded_bilinear.tendsto
  1328   LIMSEQ_mult ~> tendsto_mult
  1329   LIMSEQ_inverse ~> tendsto_inverse
  1330   LIMSEQ_divide ~> tendsto_divide
  1331   LIMSEQ_pow ~> tendsto_power
  1332   LIMSEQ_setsum ~> tendsto_setsum
  1333   LIMSEQ_setprod ~> tendsto_setprod
  1334   LIMSEQ_norm_zero ~> tendsto_norm_zero_iff
  1335   LIMSEQ_rabs_zero ~> tendsto_rabs_zero_iff
  1336   LIMSEQ_imp_rabs ~> tendsto_rabs
  1337   LIMSEQ_add_minus ~> tendsto_add [OF _ tendsto_minus]
  1338   LIMSEQ_add_const ~> tendsto_add [OF _ tendsto_const]
  1339   LIMSEQ_diff_const ~> tendsto_diff [OF _ tendsto_const]
  1340   LIMSEQ_Complex ~> tendsto_Complex
  1341   LIM_ident ~> tendsto_ident_at
  1342   LIM_const ~> tendsto_const
  1343   LIM_add ~> tendsto_add
  1344   LIM_add_zero ~> tendsto_add_zero
  1345   LIM_minus ~> tendsto_minus
  1346   LIM_diff ~> tendsto_diff
  1347   LIM_norm ~> tendsto_norm
  1348   LIM_norm_zero ~> tendsto_norm_zero
  1349   LIM_norm_zero_cancel ~> tendsto_norm_zero_cancel
  1350   LIM_norm_zero_iff ~> tendsto_norm_zero_iff
  1351   LIM_rabs ~> tendsto_rabs
  1352   LIM_rabs_zero ~> tendsto_rabs_zero
  1353   LIM_rabs_zero_cancel ~> tendsto_rabs_zero_cancel
  1354   LIM_rabs_zero_iff ~> tendsto_rabs_zero_iff
  1355   LIM_compose ~> tendsto_compose
  1356   LIM_mult ~> tendsto_mult
  1357   LIM_scaleR ~> tendsto_scaleR
  1358   LIM_of_real ~> tendsto_of_real
  1359   LIM_power ~> tendsto_power
  1360   LIM_inverse ~> tendsto_inverse
  1361   LIM_sgn ~> tendsto_sgn
  1362   isCont_LIM_compose ~> isCont_tendsto_compose
  1363   bounded_linear.LIM ~> bounded_linear.tendsto
  1364   bounded_linear.LIM_zero ~> bounded_linear.tendsto_zero
  1365   bounded_bilinear.LIM ~> bounded_bilinear.tendsto
  1366   bounded_bilinear.LIM_prod_zero ~> bounded_bilinear.tendsto_zero
  1367   bounded_bilinear.LIM_left_zero ~> bounded_bilinear.tendsto_left_zero
  1368   bounded_bilinear.LIM_right_zero ~> bounded_bilinear.tendsto_right_zero
  1369   LIM_inverse_fun ~> tendsto_inverse [OF tendsto_ident_at]
  1370 
  1371 * Theory Complex_Main: The definition of infinite series was
  1372 generalized.  Now it is defined on the type class {topological_space,
  1373 comm_monoid_add}.  Hence it is useable also for extended real numbers.
  1374 
  1375 * Theory Complex_Main: The complex exponential function "expi" is now
  1376 a type-constrained abbreviation for "exp :: complex => complex"; thus
  1377 several polymorphic lemmas about "exp" are now applicable to "expi".
  1378 
  1379 * Code generation:
  1380 
  1381   - Theory Library/Code_Char_ord provides native ordering of
  1382     characters in the target language.
  1383 
  1384   - Commands code_module and code_library are legacy, use export_code
  1385     instead.
  1386 
  1387   - Method "evaluation" is legacy, use method "eval" instead.
  1388 
  1389   - Legacy evaluator "SML" is deactivated by default.  May be
  1390     reactivated by the following theory command:
  1391 
  1392       setup {* Value.add_evaluator ("SML", Codegen.eval_term) *}
  1393 
  1394 * Declare ext [intro] by default.  Rare INCOMPATIBILITY.
  1395 
  1396 * New proof method "induction" that gives induction hypotheses the
  1397 name "IH", thus distinguishing them from further hypotheses that come
  1398 from rule induction.  The latter are still called "hyps".  Method
  1399 "induction" is a thin wrapper around "induct" and follows the same
  1400 syntax.
  1401 
  1402 * Method "fastsimp" has been renamed to "fastforce", but "fastsimp" is
  1403 still available as a legacy feature for some time.
  1404 
  1405 * Nitpick:
  1406   - Added "need" and "total_consts" options.
  1407   - Reintroduced "show_skolems" option by popular demand.
  1408   - Renamed attribute: nitpick_def ~> nitpick_unfold.
  1409     INCOMPATIBILITY.
  1410 
  1411 * Sledgehammer:
  1412   - Use quasi-sound (and efficient) translations by default.
  1413   - Added support for the following provers: E-ToFoF, LEO-II,
  1414     Satallax, SNARK, Waldmeister, and Z3 with TPTP syntax.
  1415   - Automatically preplay and minimize proofs before showing them if
  1416     this can be done within reasonable time.
  1417   - sledgehammer available_provers ~> sledgehammer supported_provers.
  1418     INCOMPATIBILITY.
  1419   - Added "preplay_timeout", "slicing", "type_enc", "sound",
  1420     "max_mono_iters", and "max_new_mono_instances" options.
  1421   - Removed "explicit_apply" and "full_types" options as well as "Full
  1422     Types" Proof General menu item. INCOMPATIBILITY.
  1423 
  1424 * Metis:
  1425   - Removed "metisF" -- use "metis" instead. INCOMPATIBILITY.
  1426   - Obsoleted "metisFT" -- use "metis (full_types)" instead.
  1427     INCOMPATIBILITY.
  1428 
  1429 * Command 'try':
  1430   - Renamed 'try_methods' and added "simp:", "intro:", "dest:", and
  1431     "elim:" options. INCOMPATIBILITY.
  1432   - Introduced 'try' that not only runs 'try_methods' but also
  1433     'solve_direct', 'sledgehammer', 'quickcheck', and 'nitpick'.
  1434 
  1435 * Quickcheck:
  1436   - Added "eval" option to evaluate terms for the found counterexample
  1437     (currently only supported by the default (exhaustive) tester).
  1438   - Added post-processing of terms to obtain readable counterexamples
  1439     (currently only supported by the default (exhaustive) tester).
  1440   - New counterexample generator quickcheck[narrowing] enables
  1441     narrowing-based testing.  Requires the Glasgow Haskell compiler
  1442     with its installation location defined in the Isabelle settings
  1443     environment as ISABELLE_GHC.
  1444   - Removed quickcheck tester "SML" based on the SML code generator
  1445     (formly in HOL/Library).
  1446 
  1447 * Function package: discontinued option "tailrec".  INCOMPATIBILITY,
  1448 use 'partial_function' instead.
  1449 
  1450 * Theory Library/Extended_Reals replaces now the positive extended
  1451 reals found in probability theory. This file is extended by
  1452 Multivariate_Analysis/Extended_Real_Limits.
  1453 
  1454 * Theory Library/Old_Recdef: old 'recdef' package has been moved here,
  1455 from where it must be imported explicitly if it is really required.
  1456 INCOMPATIBILITY.
  1457 
  1458 * Theory Library/Wfrec: well-founded recursion combinator "wfrec" has
  1459 been moved here.  INCOMPATIBILITY.
  1460 
  1461 * Theory Library/Saturated provides type of numbers with saturated
  1462 arithmetic.
  1463 
  1464 * Theory Library/Product_Lattice defines a pointwise ordering for the
  1465 product type 'a * 'b, and provides instance proofs for various order
  1466 and lattice type classes.
  1467 
  1468 * Theory Library/Countable now provides the "countable_datatype" proof
  1469 method for proving "countable" class instances for datatypes.
  1470 
  1471 * Theory Library/Cset_Monad allows do notation for computable sets
  1472 (cset) via the generic monad ad-hoc overloading facility.
  1473 
  1474 * Library: Theories of common data structures are split into theories
  1475 for implementation, an invariant-ensuring type, and connection to an
  1476 abstract type. INCOMPATIBILITY.
  1477 
  1478   - RBT is split into RBT and RBT_Mapping.
  1479   - AssocList is split and renamed into AList and AList_Mapping.
  1480   - DList is split into DList_Impl, DList, and DList_Cset.
  1481   - Cset is split into Cset and List_Cset.
  1482 
  1483 * Theory Library/Nat_Infinity has been renamed to
  1484 Library/Extended_Nat, with name changes of the following types and
  1485 constants:
  1486 
  1487   type inat   ~> type enat
  1488   Fin         ~> enat
  1489   Infty       ~> infinity (overloaded)
  1490   iSuc        ~> eSuc
  1491   the_Fin     ~> the_enat
  1492 
  1493 Every theorem name containing "inat", "Fin", "Infty", or "iSuc" has
  1494 been renamed accordingly. INCOMPATIBILITY.
  1495 
  1496 * Session Multivariate_Analysis: The euclidean_space type class now
  1497 fixes a constant "Basis :: 'a set" consisting of the standard
  1498 orthonormal basis for the type. Users now have the option of
  1499 quantifying over this set instead of using the "basis" function, e.g.
  1500 "ALL x:Basis. P x" vs "ALL i<DIM('a). P (basis i)".
  1501 
  1502 * Session Multivariate_Analysis: Type "('a, 'b) cart" has been renamed
  1503 to "('a, 'b) vec" (the syntax "'a ^ 'b" remains unaffected). Constants
  1504 "Cart_nth" and "Cart_lambda" have been respectively renamed to
  1505 "vec_nth" and "vec_lambda"; theorems mentioning those names have
  1506 changed to match. Definition theorems for overloaded constants now use
  1507 the standard "foo_vec_def" naming scheme. A few other theorems have
  1508 been renamed as follows (INCOMPATIBILITY):
  1509 
  1510   Cart_eq          ~> vec_eq_iff
  1511   dist_nth_le_cart ~> dist_vec_nth_le
  1512   tendsto_vector   ~> vec_tendstoI
  1513   Cauchy_vector    ~> vec_CauchyI
  1514 
  1515 * Session Multivariate_Analysis: Several duplicate theorems have been
  1516 removed, and other theorems have been renamed or replaced with more
  1517 general versions. INCOMPATIBILITY.
  1518 
  1519   finite_choice ~> finite_set_choice
  1520   eventually_conjI ~> eventually_conj
  1521   eventually_and ~> eventually_conj_iff
  1522   eventually_false ~> eventually_False
  1523   setsum_norm ~> norm_setsum
  1524   Lim_sequentially ~> LIMSEQ_def
  1525   Lim_ident_at ~> LIM_ident
  1526   Lim_const ~> tendsto_const
  1527   Lim_cmul ~> tendsto_scaleR [OF tendsto_const]
  1528   Lim_neg ~> tendsto_minus
  1529   Lim_add ~> tendsto_add
  1530   Lim_sub ~> tendsto_diff
  1531   Lim_mul ~> tendsto_scaleR
  1532   Lim_vmul ~> tendsto_scaleR [OF _ tendsto_const]
  1533   Lim_null_norm ~> tendsto_norm_zero_iff [symmetric]
  1534   Lim_linear ~> bounded_linear.tendsto
  1535   Lim_component ~> tendsto_euclidean_component
  1536   Lim_component_cart ~> tendsto_vec_nth
  1537   Lim_inner ~> tendsto_inner [OF tendsto_const]
  1538   dot_lsum ~> inner_setsum_left
  1539   dot_rsum ~> inner_setsum_right
  1540   continuous_cmul ~> continuous_scaleR [OF continuous_const]
  1541   continuous_neg ~> continuous_minus
  1542   continuous_sub ~> continuous_diff
  1543   continuous_vmul ~> continuous_scaleR [OF _ continuous_const]
  1544   continuous_mul ~> continuous_scaleR
  1545   continuous_inv ~> continuous_inverse
  1546   continuous_at_within_inv ~> continuous_at_within_inverse
  1547   continuous_at_inv ~> continuous_at_inverse
  1548   continuous_at_norm ~> continuous_norm [OF continuous_at_id]
  1549   continuous_at_infnorm ~> continuous_infnorm [OF continuous_at_id]
  1550   continuous_at_component ~> continuous_component [OF continuous_at_id]
  1551   continuous_on_neg ~> continuous_on_minus
  1552   continuous_on_sub ~> continuous_on_diff
  1553   continuous_on_cmul ~> continuous_on_scaleR [OF continuous_on_const]
  1554   continuous_on_vmul ~> continuous_on_scaleR [OF _ continuous_on_const]
  1555   continuous_on_mul ~> continuous_on_scaleR
  1556   continuous_on_mul_real ~> continuous_on_mult
  1557   continuous_on_inner ~> continuous_on_inner [OF continuous_on_const]
  1558   continuous_on_norm ~> continuous_on_norm [OF continuous_on_id]
  1559   continuous_on_inverse ~> continuous_on_inv
  1560   uniformly_continuous_on_neg ~> uniformly_continuous_on_minus
  1561   uniformly_continuous_on_sub ~> uniformly_continuous_on_diff
  1562   subset_interior ~> interior_mono
  1563   subset_closure ~> closure_mono
  1564   closure_univ ~> closure_UNIV
  1565   real_arch_lt ~> reals_Archimedean2
  1566   real_arch ~> reals_Archimedean3
  1567   real_abs_norm ~> abs_norm_cancel
  1568   real_abs_sub_norm ~> norm_triangle_ineq3
  1569   norm_cauchy_schwarz_abs ~> Cauchy_Schwarz_ineq2
  1570 
  1571 * Session HOL-Probability:
  1572   - Caratheodory's extension lemma is now proved for ring_of_sets.
  1573   - Infinite products of probability measures are now available.
  1574   - Sigma closure is independent, if the generator is independent
  1575   - Use extended reals instead of positive extended
  1576     reals. INCOMPATIBILITY.
  1577 
  1578 * Session HOLCF: Discontinued legacy theorem names, INCOMPATIBILITY.
  1579 
  1580   expand_fun_below ~> fun_below_iff
  1581   below_fun_ext ~> fun_belowI
  1582   expand_cfun_eq ~> cfun_eq_iff
  1583   ext_cfun ~> cfun_eqI
  1584   expand_cfun_below ~> cfun_below_iff
  1585   below_cfun_ext ~> cfun_belowI
  1586   monofun_fun_fun ~> fun_belowD
  1587   monofun_fun_arg ~> monofunE
  1588   monofun_lub_fun ~> adm_monofun [THEN admD]
  1589   cont_lub_fun ~> adm_cont [THEN admD]
  1590   cont2cont_Rep_CFun ~> cont2cont_APP
  1591   cont_Rep_CFun_app ~> cont_APP_app
  1592   cont_Rep_CFun_app_app ~> cont_APP_app_app
  1593   cont_cfun_fun ~> cont_Rep_cfun1 [THEN contE]
  1594   cont_cfun_arg ~> cont_Rep_cfun2 [THEN contE]
  1595   contlub_cfun ~> lub_APP [symmetric]
  1596   contlub_LAM ~> lub_LAM [symmetric]
  1597   thelubI ~> lub_eqI
  1598   UU_I ~> bottomI
  1599   lift_distinct1 ~> lift.distinct(1)
  1600   lift_distinct2 ~> lift.distinct(2)
  1601   Def_not_UU ~> lift.distinct(2)
  1602   Def_inject ~> lift.inject
  1603   below_UU_iff ~> below_bottom_iff
  1604   eq_UU_iff ~> eq_bottom_iff
  1605 
  1606 
  1607 *** Document preparation ***
  1608 
  1609 * Antiquotation @{rail} layouts railroad syntax diagrams, see also
  1610 isar-ref manual, both for description and actual application of the
  1611 same.
  1612 
  1613 * Antiquotation @{value} evaluates the given term and presents its
  1614 result.
  1615 
  1616 * Antiquotations: term style "isub" provides ad-hoc conversion of
  1617 variables x1, y23 into subscripted form x\<^isub>1,
  1618 y\<^isub>2\<^isub>3.
  1619 
  1620 * Predefined LaTeX macros for Isabelle symbols \<bind> and \<then>
  1621 (e.g. see ~~/src/HOL/Library/Monad_Syntax.thy).
  1622 
  1623 * Localized \isabellestyle switch can be used within blocks or groups
  1624 like this:
  1625 
  1626   \isabellestyle{it}  %preferred default
  1627   {\isabellestylett @{text "typewriter stuff"}}
  1628 
  1629 * Discontinued special treatment of hard tabulators.  Implicit
  1630 tab-width is now defined as 1.  Potential INCOMPATIBILITY for visual
  1631 layouts.
  1632 
  1633 
  1634 *** ML ***
  1635 
  1636 * The inner syntax of sort/type/term/prop supports inlined YXML
  1637 representations within quoted string tokens.  By encoding logical
  1638 entities via Term_XML (in ML or Scala) concrete syntax can be
  1639 bypassed, which is particularly useful for producing bits of text
  1640 under external program control.
  1641 
  1642 * Antiquotations for ML and document preparation are managed as theory
  1643 data, which requires explicit setup.
  1644 
  1645 * Isabelle_Process.is_active allows tools to check if the official
  1646 process wrapper is running (Isabelle/Scala/jEdit) or the old TTY loop
  1647 (better known as Proof General).
  1648 
  1649 * Structure Proof_Context follows standard naming scheme.  Old
  1650 ProofContext is still available for some time as legacy alias.
  1651 
  1652 * Structure Timing provides various operations for timing; supersedes
  1653 former start_timing/end_timing etc.
  1654 
  1655 * Path.print is the official way to show file-system paths to users
  1656 (including quotes etc.).
  1657 
  1658 * Inner syntax: identifiers in parse trees of generic categories
  1659 "logic", "aprop", "idt" etc. carry position information (disguised as
  1660 type constraints).  Occasional INCOMPATIBILITY with non-compliant
  1661 translations that choke on unexpected type constraints.  Positions can
  1662 be stripped in ML translations via Syntax.strip_positions /
  1663 Syntax.strip_positions_ast, or via the syntax constant
  1664 "_strip_positions" within parse trees.  As last resort, positions can
  1665 be disabled via the configuration option Syntax.positions, which is
  1666 called "syntax_positions" in Isar attribute syntax.
  1667 
  1668 * Discontinued special status of various ML structures that contribute
  1669 to structure Syntax (Ast, Lexicon, Mixfix, Parser, Printer etc.): less
  1670 pervasive content, no inclusion in structure Syntax.  INCOMPATIBILITY,
  1671 refer directly to Ast.Constant, Lexicon.is_identifier,
  1672 Syntax_Trans.mk_binder_tr etc.
  1673 
  1674 * Typed print translation: discontinued show_sorts argument, which is
  1675 already available via context of "advanced" translation.
  1676 
  1677 * Refined PARALLEL_GOALS tactical: degrades gracefully for schematic
  1678 goal states; body tactic needs to address all subgoals uniformly.
  1679 
  1680 * Slightly more special eq_list/eq_set, with shortcut involving
  1681 pointer equality (assumes that eq relation is reflexive).
  1682 
  1683 * Classical tactics use proper Proof.context instead of historic types
  1684 claset/clasimpset.  Old-style declarations like addIs, addEs, addDs
  1685 operate directly on Proof.context.  Raw type claset retains its use as
  1686 snapshot of the classical context, which can be recovered via
  1687 (put_claset HOL_cs) etc.  Type clasimpset has been discontinued.
  1688 INCOMPATIBILITY, classical tactics and derived proof methods require
  1689 proper Proof.context.
  1690 
  1691 
  1692 *** System ***
  1693 
  1694 * Discontinued support for Poly/ML 5.2, which was the last version
  1695 without proper multithreading and TimeLimit implementation.
  1696 
  1697 * Discontinued old lib/scripts/polyml-platform, which has been
  1698 obsolete since Isabelle2009-2.
  1699 
  1700 * Various optional external tools are referenced more robustly and
  1701 uniformly by explicit Isabelle settings as follows:
  1702 
  1703   ISABELLE_CSDP   (formerly CSDP_EXE)
  1704   ISABELLE_GHC    (formerly EXEC_GHC or GHC_PATH)
  1705   ISABELLE_OCAML  (formerly EXEC_OCAML)
  1706   ISABELLE_SWIPL  (formerly EXEC_SWIPL)
  1707   ISABELLE_YAP    (formerly EXEC_YAP)
  1708 
  1709 Note that automated detection from the file-system or search path has
  1710 been discontinued.  INCOMPATIBILITY.
  1711 
  1712 * Scala layer provides JVM method invocation service for static
  1713 methods of type (String)String, see Invoke_Scala.method in ML.  For
  1714 example:
  1715 
  1716   Invoke_Scala.method "java.lang.System.getProperty" "java.home"
  1717 
  1718 Together with YXML.string_of_body/parse_body and XML.Encode/Decode
  1719 this allows to pass structured values between ML and Scala.
  1720 
  1721 * The IsabelleText fonts includes some further glyphs to support the
  1722 Prover IDE.  Potential INCOMPATIBILITY: users who happen to have
  1723 installed a local copy (which is normally *not* required) need to
  1724 delete or update it from ~~/lib/fonts/.
  1725 
  1726 
  1727 
  1728 New in Isabelle2011 (January 2011)
  1729 ----------------------------------
  1730 
  1731 *** General ***
  1732 
  1733 * Experimental Prover IDE based on Isabelle/Scala and jEdit (see
  1734 src/Tools/jEdit).  This also serves as IDE for Isabelle/ML, with
  1735 useful tooltips and hyperlinks produced from its static analysis.  The
  1736 bundled component provides an executable Isabelle tool that can be run
  1737 like this:
  1738 
  1739   Isabelle2011/bin/isabelle jedit
  1740 
  1741 * Significantly improved Isabelle/Isar implementation manual.
  1742 
  1743 * System settings: ISABELLE_HOME_USER now includes ISABELLE_IDENTIFIER
  1744 (and thus refers to something like $HOME/.isabelle/Isabelle2011),
  1745 while the default heap location within that directory lacks that extra
  1746 suffix.  This isolates multiple Isabelle installations from each
  1747 other, avoiding problems with old settings in new versions.
  1748 INCOMPATIBILITY, need to copy/upgrade old user settings manually.
  1749 
  1750 * Source files are always encoded as UTF-8, instead of old-fashioned
  1751 ISO-Latin-1.  INCOMPATIBILITY.  Isabelle LaTeX documents might require
  1752 the following package declarations:
  1753 
  1754   \usepackage[utf8]{inputenc}
  1755   \usepackage{textcomp}
  1756 
  1757 * Explicit treatment of UTF-8 sequences as Isabelle symbols, such that
  1758 a Unicode character is treated as a single symbol, not a sequence of
  1759 non-ASCII bytes as before.  Since Isabelle/ML string literals may
  1760 contain symbols without further backslash escapes, Unicode can now be
  1761 used here as well.  Recall that Symbol.explode in ML provides a
  1762 consistent view on symbols, while raw explode (or String.explode)
  1763 merely give a byte-oriented representation.
  1764 
  1765 * Theory loader: source files are primarily located via the master
  1766 directory of each theory node (where the .thy file itself resides).
  1767 The global load path is still partially available as legacy feature.
  1768 Minor INCOMPATIBILITY due to subtle change in file lookup: use
  1769 explicit paths, relatively to the theory.
  1770 
  1771 * Special treatment of ML file names has been discontinued.
  1772 Historically, optional extensions .ML or .sml were added on demand --
  1773 at the cost of clarity of file dependencies.  Recall that Isabelle/ML
  1774 files exclusively use the .ML extension.  Minor INCOMPATIBILTY.
  1775 
  1776 * Various options that affect pretty printing etc. are now properly
  1777 handled within the context via configuration options, instead of
  1778 unsynchronized references or print modes.  There are both ML Config.T
  1779 entities and Isar declaration attributes to access these.
  1780 
  1781   ML (Config.T)                 Isar (attribute)
  1782 
  1783   eta_contract                  eta_contract
  1784   show_brackets                 show_brackets
  1785   show_sorts                    show_sorts
  1786   show_types                    show_types
  1787   show_question_marks           show_question_marks
  1788   show_consts                   show_consts
  1789   show_abbrevs                  show_abbrevs
  1790 
  1791   Syntax.ast_trace              syntax_ast_trace
  1792   Syntax.ast_stat               syntax_ast_stat
  1793   Syntax.ambiguity_level        syntax_ambiguity_level
  1794 
  1795   Goal_Display.goals_limit      goals_limit
  1796   Goal_Display.show_main_goal   show_main_goal
  1797 
  1798   Method.rule_trace             rule_trace
  1799 
  1800   Thy_Output.display            thy_output_display
  1801   Thy_Output.quotes             thy_output_quotes
  1802   Thy_Output.indent             thy_output_indent
  1803   Thy_Output.source             thy_output_source
  1804   Thy_Output.break              thy_output_break
  1805 
  1806 Note that corresponding "..._default" references in ML may only be
  1807 changed globally at the ROOT session setup, but *not* within a theory.
  1808 The option "show_abbrevs" supersedes the former print mode
  1809 "no_abbrevs" with inverted meaning.
  1810 
  1811 * More systematic naming of some configuration options.
  1812 INCOMPATIBILITY.
  1813 
  1814   trace_simp  ~>  simp_trace
  1815   debug_simp  ~>  simp_debug
  1816 
  1817 * Support for real valued configuration options, using simplistic
  1818 floating-point notation that coincides with the inner syntax for
  1819 float_token.
  1820 
  1821 * Support for real valued preferences (with approximative PGIP type):
  1822 front-ends need to accept "pgint" values in float notation.
  1823 INCOMPATIBILITY.
  1824 
  1825 * The IsabelleText font now includes Cyrillic, Hebrew, Arabic from
  1826 DejaVu Sans.
  1827 
  1828 * Discontinued support for Poly/ML 5.0 and 5.1 versions.
  1829 
  1830 
  1831 *** Pure ***
  1832 
  1833 * Command 'type_synonym' (with single argument) replaces somewhat
  1834 outdated 'types', which is still available as legacy feature for some
  1835 time.
  1836 
  1837 * Command 'nonterminal' (with 'and' separated list of arguments)
  1838 replaces somewhat outdated 'nonterminals'.  INCOMPATIBILITY.
  1839 
  1840 * Command 'notepad' replaces former 'example_proof' for
  1841 experimentation in Isar without any result.  INCOMPATIBILITY.
  1842 
  1843 * Locale interpretation commands 'interpret' and 'sublocale' accept
  1844 lists of equations to map definitions in a locale to appropriate
  1845 entities in the context of the interpretation.  The 'interpretation'
  1846 command already provided this functionality.
  1847 
  1848 * Diagnostic command 'print_dependencies' prints the locale instances
  1849 that would be activated if the specified expression was interpreted in
  1850 the current context.  Variant "print_dependencies!" assumes a context
  1851 without interpretations.
  1852 
  1853 * Diagnostic command 'print_interps' prints interpretations in proofs
  1854 in addition to interpretations in theories.
  1855 
  1856 * Discontinued obsolete 'global' and 'local' commands to manipulate
  1857 the theory name space.  Rare INCOMPATIBILITY.  The ML functions
  1858 Sign.root_path and Sign.local_path may be applied directly where this
  1859 feature is still required for historical reasons.
  1860 
  1861 * Discontinued obsolete 'constdefs' command.  INCOMPATIBILITY, use
  1862 'definition' instead.
  1863 
  1864 * The "prems" fact, which refers to the accidental collection of
  1865 foundational premises in the context, is now explicitly marked as
  1866 legacy feature and will be discontinued soon.  Consider using "assms"
  1867 of the head statement or reference facts by explicit names.
  1868 
  1869 * Document antiquotations @{class} and @{type} print classes and type
  1870 constructors.
  1871 
  1872 * Document antiquotation @{file} checks file/directory entries within
  1873 the local file system.
  1874 
  1875 
  1876 *** HOL ***
  1877 
  1878 * Coercive subtyping: functions can be declared as coercions and type
  1879 inference will add them as necessary upon input of a term.  Theory
  1880 Complex_Main declares real :: nat => real and real :: int => real as
  1881 coercions. A coercion function f is declared like this:
  1882 
  1883   declare [[coercion f]]
  1884 
  1885 To lift coercions through type constructors (e.g. from nat => real to
  1886 nat list => real list), map functions can be declared, e.g.
  1887 
  1888   declare [[coercion_map map]]
  1889 
  1890 Currently coercion inference is activated only in theories including
  1891 real numbers, i.e. descendants of Complex_Main.  This is controlled by
  1892 the configuration option "coercion_enabled", e.g. it can be enabled in
  1893 other theories like this:
  1894 
  1895   declare [[coercion_enabled]]
  1896 
  1897 * Command 'partial_function' provides basic support for recursive
  1898 function definitions over complete partial orders.  Concrete instances
  1899 are provided for i) the option type, ii) tail recursion on arbitrary
  1900 types, and iii) the heap monad of Imperative_HOL.  See
  1901 src/HOL/ex/Fundefs.thy and src/HOL/Imperative_HOL/ex/Linked_Lists.thy
  1902 for examples.
  1903 
  1904 * Function package: f.psimps rules are no longer implicitly declared
  1905 as [simp].  INCOMPATIBILITY.
  1906 
  1907 * Datatype package: theorems generated for executable equality (class
  1908 "eq") carry proper names and are treated as default code equations.
  1909 
  1910 * Inductive package: now offers command 'inductive_simps' to
  1911 automatically derive instantiated and simplified equations for
  1912 inductive predicates, similar to 'inductive_cases'.
  1913 
  1914 * Command 'enriched_type' allows to register properties of the
  1915 functorial structure of types.
  1916 
  1917 * Improved infrastructure for term evaluation using code generator
  1918 techniques, in particular static evaluation conversions.
  1919 
  1920 * Code generator: Scala (2.8 or higher) has been added to the target
  1921 languages.
  1922 
  1923 * Code generator: globbing constant expressions "*" and "Theory.*"
  1924 have been replaced by the more idiomatic "_" and "Theory._".
  1925 INCOMPATIBILITY.
  1926 
  1927 * Code generator: export_code without explicit file declaration prints
  1928 to standard output.  INCOMPATIBILITY.
  1929 
  1930 * Code generator: do not print function definitions for case
  1931 combinators any longer.
  1932 
  1933 * Code generator: simplification with rules determined with
  1934 src/Tools/Code/code_simp.ML and method "code_simp".
  1935 
  1936 * Code generator for records: more idiomatic representation of record
  1937 types.  Warning: records are not covered by ancient SML code
  1938 generation any longer.  INCOMPATIBILITY.  In cases of need, a suitable
  1939 rep_datatype declaration helps to succeed then:
  1940 
  1941   record 'a foo = ...
  1942   ...
  1943   rep_datatype foo_ext ...
  1944 
  1945 * Records: logical foundation type for records does not carry a
  1946 '_type' suffix any longer (obsolete due to authentic syntax).
  1947 INCOMPATIBILITY.
  1948 
  1949 * Quickcheck now by default uses exhaustive testing instead of random
  1950 testing.  Random testing can be invoked by "quickcheck [random]",
  1951 exhaustive testing by "quickcheck [exhaustive]".
  1952 
  1953 * Quickcheck instantiates polymorphic types with small finite
  1954 datatypes by default. This enables a simple execution mechanism to
  1955 handle quantifiers and function equality over the finite datatypes.
  1956 
  1957 * Quickcheck random generator has been renamed from "code" to
  1958 "random".  INCOMPATIBILITY.
  1959 
  1960 * Quickcheck now has a configurable time limit which is set to 30
  1961 seconds by default. This can be changed by adding [timeout = n] to the
  1962 quickcheck command. The time limit for Auto Quickcheck is still set
  1963 independently.
  1964 
  1965 * Quickcheck in locales considers interpretations of that locale for
  1966 counter example search.
  1967 
  1968 * Sledgehammer:
  1969   - Added "smt" and "remote_smt" provers based on the "smt" proof
  1970     method. See the Sledgehammer manual for details ("isabelle doc
  1971     sledgehammer").
  1972   - Renamed commands:
  1973     sledgehammer atp_info ~> sledgehammer running_provers
  1974     sledgehammer atp_kill ~> sledgehammer kill_provers
  1975     sledgehammer available_atps ~> sledgehammer available_provers
  1976     INCOMPATIBILITY.
  1977   - Renamed options:
  1978     sledgehammer [atps = ...] ~> sledgehammer [provers = ...]
  1979     sledgehammer [atp = ...] ~> sledgehammer [prover = ...]
  1980     sledgehammer [timeout = 77 s] ~> sledgehammer [timeout = 77]
  1981     (and "ms" and "min" are no longer supported)
  1982     INCOMPATIBILITY.
  1983 
  1984 * Nitpick:
  1985   - Renamed options:
  1986     nitpick [timeout = 77 s] ~> nitpick [timeout = 77]
  1987     nitpick [tac_timeout = 777 ms] ~> nitpick [tac_timeout = 0.777]
  1988     INCOMPATIBILITY.
  1989   - Added support for partial quotient types.
  1990   - Added local versions of the "Nitpick.register_xxx" functions.
  1991   - Added "whack" option.
  1992   - Allow registration of quotient types as codatatypes.
  1993   - Improved "merge_type_vars" option to merge more types.
  1994   - Removed unsound "fast_descrs" option.
  1995   - Added custom symmetry breaking for datatypes, making it possible to reach
  1996     higher cardinalities.
  1997   - Prevent the expansion of too large definitions.
  1998 
  1999 * Proof methods "metis" and "meson" now have configuration options
  2000 "meson_trace", "metis_trace", and "metis_verbose" that can be enabled
  2001 to diagnose these tools. E.g.
  2002 
  2003     using [[metis_trace = true]]
  2004 
  2005 * Auto Solve: Renamed "Auto Solve Direct".  The tool is now available
  2006 manually as command 'solve_direct'.
  2007 
  2008 * The default SMT solver Z3 must be enabled explicitly (due to
  2009 licensing issues) by setting the environment variable
  2010 Z3_NON_COMMERCIAL in etc/settings of the component, for example.  For
  2011 commercial applications, the SMT solver CVC3 is provided as fall-back;
  2012 changing the SMT solver is done via the configuration option
  2013 "smt_solver".
  2014 
  2015 * Remote SMT solvers need to be referred to by the "remote_" prefix,
  2016 i.e. "remote_cvc3" and "remote_z3".
  2017 
  2018 * Added basic SMT support for datatypes, records, and typedefs using
  2019 the oracle mode (no proofs).  Direct support of pairs has been dropped
  2020 in exchange (pass theorems fst_conv snd_conv pair_collapse to the SMT
  2021 support for a similar behavior).  Minor INCOMPATIBILITY.
  2022 
  2023 * Changed SMT configuration options:
  2024   - Renamed:
  2025     z3_proofs ~> smt_oracle (with inverted meaning)
  2026     z3_trace_assms ~> smt_trace_used_facts
  2027     INCOMPATIBILITY.
  2028   - Added:
  2029     smt_verbose
  2030     smt_random_seed
  2031     smt_datatypes
  2032     smt_infer_triggers
  2033     smt_monomorph_limit
  2034     cvc3_options
  2035     remote_cvc3_options
  2036     remote_z3_options
  2037     yices_options
  2038 
  2039 * Boogie output files (.b2i files) need to be declared in the theory
  2040 header.
  2041 
  2042 * Simplification procedure "list_to_set_comprehension" rewrites list
  2043 comprehensions applied to List.set to set comprehensions.  Occasional
  2044 INCOMPATIBILITY, may be deactivated like this:
  2045 
  2046   declare [[simproc del: list_to_set_comprehension]]
  2047 
  2048 * Removed old version of primrec package.  INCOMPATIBILITY.
  2049 
  2050 * Removed simplifier congruence rule of "prod_case", as has for long
  2051 been the case with "split".  INCOMPATIBILITY.
  2052 
  2053 * String.literal is a type, but not a datatype.  INCOMPATIBILITY.
  2054 
  2055 * Removed [split_format ... and ... and ...] version of
  2056 [split_format].  Potential INCOMPATIBILITY.
  2057 
  2058 * Predicate "sorted" now defined inductively, with nice induction
  2059 rules.  INCOMPATIBILITY: former sorted.simps now named sorted_simps.
  2060 
  2061 * Constant "contents" renamed to "the_elem", to free the generic name
  2062 contents for other uses.  INCOMPATIBILITY.
  2063 
  2064 * Renamed class eq and constant eq (for code generation) to class
  2065 equal and constant equal, plus renaming of related facts and various
  2066 tuning.  INCOMPATIBILITY.
  2067 
  2068 * Dropped type classes mult_mono and mult_mono1.  INCOMPATIBILITY.
  2069 
  2070 * Removed output syntax "'a ~=> 'b" for "'a => 'b option".
  2071 INCOMPATIBILITY.
  2072 
  2073 * Renamed theory Fset to Cset, type Fset.fset to Cset.set, in order to
  2074 avoid confusion with finite sets.  INCOMPATIBILITY.
  2075 
  2076 * Abandoned locales equiv, congruent and congruent2 for equivalence
  2077 relations.  INCOMPATIBILITY: use equivI rather than equiv_intro (same
  2078 for congruent(2)).
  2079 
  2080 * Some previously unqualified names have been qualified:
  2081 
  2082   types
  2083     bool ~> HOL.bool
  2084     nat ~> Nat.nat
  2085 
  2086   constants
  2087     Trueprop ~> HOL.Trueprop
  2088     True ~> HOL.True
  2089     False ~> HOL.False
  2090     op & ~> HOL.conj
  2091     op | ~> HOL.disj
  2092     op --> ~> HOL.implies
  2093     op = ~> HOL.eq
  2094     Not ~> HOL.Not
  2095     The ~> HOL.The
  2096     All ~> HOL.All
  2097     Ex ~> HOL.Ex
  2098     Ex1 ~> HOL.Ex1
  2099     Let ~> HOL.Let
  2100     If ~> HOL.If
  2101     Ball ~> Set.Ball
  2102     Bex ~> Set.Bex
  2103     Suc ~> Nat.Suc
  2104     Pair ~> Product_Type.Pair
  2105     fst ~> Product_Type.fst
  2106     snd ~> Product_Type.snd
  2107     curry ~> Product_Type.curry
  2108     op : ~> Set.member
  2109     Collect ~> Set.Collect
  2110 
  2111 INCOMPATIBILITY.
  2112 
  2113 * More canonical naming convention for some fundamental definitions:
  2114 
  2115     bot_bool_eq ~> bot_bool_def
  2116     top_bool_eq ~> top_bool_def
  2117     inf_bool_eq ~> inf_bool_def
  2118     sup_bool_eq ~> sup_bool_def
  2119     bot_fun_eq  ~> bot_fun_def
  2120     top_fun_eq  ~> top_fun_def
  2121     inf_fun_eq  ~> inf_fun_def
  2122     sup_fun_eq  ~> sup_fun_def
  2123 
  2124 INCOMPATIBILITY.
  2125 
  2126 * More stylized fact names:
  2127 
  2128   expand_fun_eq ~> fun_eq_iff
  2129   expand_set_eq ~> set_eq_iff
  2130   set_ext       ~> set_eqI
  2131   nat_number    ~> eval_nat_numeral
  2132 
  2133 INCOMPATIBILITY.
  2134 
  2135 * Refactoring of code-generation specific operations in theory List:
  2136 
  2137   constants
  2138     null ~> List.null
  2139 
  2140   facts
  2141     mem_iff ~> member_def
  2142     null_empty ~> null_def
  2143 
  2144 INCOMPATIBILITY.  Note that these were not supposed to be used
  2145 regularly unless for striking reasons; their main purpose was code
  2146 generation.
  2147 
  2148 Various operations from the Haskell prelude are used for generating
  2149 Haskell code.
  2150 
  2151 * Term "bij f" is now an abbreviation of "bij_betw f UNIV UNIV".  Term
  2152 "surj f" is now an abbreviation of "range f = UNIV".  The theorems
  2153 bij_def and surj_def are unchanged.  INCOMPATIBILITY.
  2154 
  2155 * Abolished some non-alphabetic type names: "prod" and "sum" replace
  2156 "*" and "+" respectively.  INCOMPATIBILITY.
  2157 
  2158 * Name "Plus" of disjoint sum operator "<+>" is now hidden.  Write
  2159 "Sum_Type.Plus" instead.
  2160 
  2161 * Constant "split" has been merged with constant "prod_case"; names of
  2162 ML functions, facts etc. involving split have been retained so far,
  2163 though.  INCOMPATIBILITY.
  2164 
  2165 * Dropped old infix syntax "_ mem _" for List.member; use "_ : set _"
  2166 instead.  INCOMPATIBILITY.
  2167 
  2168 * Removed lemma "Option.is_none_none" which duplicates "is_none_def".
  2169 INCOMPATIBILITY.
  2170 
  2171 * Former theory Library/Enum is now part of the HOL-Main image.
  2172 INCOMPATIBILITY: all constants of the Enum theory now have to be
  2173 referred to by its qualified name.
  2174 
  2175   enum    ~>  Enum.enum
  2176   nlists  ~>  Enum.nlists
  2177   product ~>  Enum.product
  2178 
  2179 * Theory Library/Monad_Syntax provides do-syntax for monad types.
  2180 Syntax in Library/State_Monad has been changed to avoid ambiguities.
  2181 INCOMPATIBILITY.
  2182 
  2183 * Theory Library/SetsAndFunctions has been split into
  2184 Library/Function_Algebras and Library/Set_Algebras; canonical names
  2185 for instance definitions for functions; various improvements.
  2186 INCOMPATIBILITY.
  2187 
  2188 * Theory Library/Multiset provides stable quicksort implementation of
  2189 sort_key.
  2190 
  2191 * Theory Library/Multiset: renamed empty_idemp ~> empty_neutral.
  2192 INCOMPATIBILITY.
  2193 
  2194 * Session Multivariate_Analysis: introduced a type class for euclidean
  2195 space.  Most theorems are now stated in terms of euclidean spaces
  2196 instead of finite cartesian products.
  2197 
  2198   types
  2199     real ^ 'n ~>  'a::real_vector
  2200               ~>  'a::euclidean_space
  2201               ~>  'a::ordered_euclidean_space
  2202         (depends on your needs)
  2203 
  2204   constants
  2205      _ $ _        ~> _ $$ _
  2206      \<chi> x. _  ~> \<chi>\<chi> x. _
  2207      CARD('n)     ~> DIM('a)
  2208 
  2209 Also note that the indices are now natural numbers and not from some
  2210 finite type. Finite cartesian products of euclidean spaces, products
  2211 of euclidean spaces the real and complex numbers are instantiated to
  2212 be euclidean_spaces.  INCOMPATIBILITY.
  2213 
  2214 * Session Probability: introduced pextreal as positive extended real
  2215 numbers.  Use pextreal as value for measures.  Introduce the
  2216 Radon-Nikodym derivative, product spaces and Fubini's theorem for
  2217 arbitrary sigma finite measures.  Introduces Lebesgue measure based on
  2218 the integral in Multivariate Analysis.  INCOMPATIBILITY.
  2219 
  2220 * Session Imperative_HOL: revamped, corrected dozens of inadequacies.
  2221 INCOMPATIBILITY.
  2222 
  2223 * Session SPARK (with image HOL-SPARK) provides commands to load and
  2224 prove verification conditions generated by the SPARK Ada program
  2225 verifier.  See also src/HOL/SPARK and src/HOL/SPARK/Examples.
  2226 
  2227 
  2228 *** HOL-Algebra ***
  2229 
  2230 * Theorems for additive ring operations (locale abelian_monoid and
  2231 descendants) are generated by interpretation from their multiplicative
  2232 counterparts.  Names (in particular theorem names) have the mandatory
  2233 qualifier 'add'.  Previous theorem names are redeclared for
  2234 compatibility.
  2235 
  2236 * Structure "int_ring" is now an abbreviation (previously a
  2237 definition).  This fits more natural with advanced interpretations.
  2238 
  2239 
  2240 *** HOLCF ***
  2241 
  2242 * The domain package now runs in definitional mode by default: The
  2243 former command 'new_domain' is now called 'domain'.  To use the domain
  2244 package in its original axiomatic mode, use 'domain (unsafe)'.
  2245 INCOMPATIBILITY.
  2246 
  2247 * The new class "domain" is now the default sort.  Class "predomain"
  2248 is an unpointed version of "domain". Theories can be updated by
  2249 replacing sort annotations as shown below.  INCOMPATIBILITY.
  2250 
  2251   'a::type ~> 'a::countable
  2252   'a::cpo  ~> 'a::predomain
  2253   'a::pcpo ~> 'a::domain
  2254 
  2255 * The old type class "rep" has been superseded by class "domain".
  2256 Accordingly, users of the definitional package must remove any
  2257 "default_sort rep" declarations.  INCOMPATIBILITY.
  2258 
  2259 * The domain package (definitional mode) now supports unpointed
  2260 predomain argument types, as long as they are marked 'lazy'. (Strict
  2261 arguments must be in class "domain".) For example, the following
  2262 domain definition now works:
  2263 
  2264   domain natlist = nil | cons (lazy "nat discr") (lazy "natlist")
  2265 
  2266 * Theory HOLCF/Library/HOL_Cpo provides cpo and predomain class
  2267 instances for types from main HOL: bool, nat, int, char, 'a + 'b,
  2268 'a option, and 'a list.  Additionally, it configures fixrec and the
  2269 domain package to work with these types.  For example:
  2270 
  2271   fixrec isInl :: "('a + 'b) u -> tr"
  2272     where "isInl$(up$(Inl x)) = TT" | "isInl$(up$(Inr y)) = FF"
  2273 
  2274   domain V = VFun (lazy "V -> V") | VCon (lazy "nat") (lazy "V list")
  2275 
  2276 * The "(permissive)" option of fixrec has been replaced with a
  2277 per-equation "(unchecked)" option. See
  2278 src/HOL/HOLCF/Tutorial/Fixrec_ex.thy for examples. INCOMPATIBILITY.
  2279 
  2280 * The "bifinite" class no longer fixes a constant "approx"; the class
  2281 now just asserts that such a function exists.  INCOMPATIBILITY.
  2282 
  2283 * Former type "alg_defl" has been renamed to "defl".  HOLCF no longer
  2284 defines an embedding of type 'a defl into udom by default; instances
  2285 of "bifinite" and "domain" classes are available in
  2286 src/HOL/HOLCF/Library/Defl_Bifinite.thy.
  2287 
  2288 * The syntax "REP('a)" has been replaced with "DEFL('a)".
  2289 
  2290 * The predicate "directed" has been removed.  INCOMPATIBILITY.
  2291 
  2292 * The type class "finite_po" has been removed.  INCOMPATIBILITY.
  2293 
  2294 * The function "cprod_map" has been renamed to "prod_map".
  2295 INCOMPATIBILITY.
  2296 
  2297 * The monadic bind operator on each powerdomain has new binder syntax
  2298 similar to sets, e.g. "\<Union>\<sharp>x\<in>xs. t" represents
  2299 "upper_bind\<cdot>xs\<cdot>(\<Lambda> x. t)".
  2300 
  2301 * The infix syntax for binary union on each powerdomain has changed
  2302 from e.g. "+\<sharp>" to "\<union>\<sharp>", for consistency with set
  2303 syntax.  INCOMPATIBILITY.
  2304 
  2305 * The constant "UU" has been renamed to "bottom".  The syntax "UU" is
  2306 still supported as an input translation.
  2307 
  2308 * Renamed some theorems (the original names are also still available).
  2309 
  2310   expand_fun_below   ~> fun_below_iff
  2311   below_fun_ext      ~> fun_belowI
  2312   expand_cfun_eq     ~> cfun_eq_iff
  2313   ext_cfun           ~> cfun_eqI
  2314   expand_cfun_below  ~> cfun_below_iff
  2315   below_cfun_ext     ~> cfun_belowI
  2316   cont2cont_Rep_CFun ~> cont2cont_APP
  2317 
  2318 * The Abs and Rep functions for various types have changed names.
  2319 Related theorem names have also changed to match. INCOMPATIBILITY.
  2320 
  2321   Rep_CFun  ~> Rep_cfun
  2322   Abs_CFun  ~> Abs_cfun
  2323   Rep_Sprod ~> Rep_sprod
  2324   Abs_Sprod ~> Abs_sprod
  2325   Rep_Ssum  ~> Rep_ssum
  2326   Abs_Ssum  ~> Abs_ssum
  2327 
  2328 * Lemmas with names of the form *_defined_iff or *_strict_iff have
  2329 been renamed to *_bottom_iff.  INCOMPATIBILITY.
  2330 
  2331 * Various changes to bisimulation/coinduction with domain package:
  2332 
  2333   - Definitions of "bisim" constants no longer mention definedness.
  2334   - With mutual recursion, "bisim" predicate is now curried.
  2335   - With mutual recursion, each type gets a separate coind theorem.
  2336   - Variable names in bisim_def and coinduct rules have changed.
  2337 
  2338 INCOMPATIBILITY.
  2339 
  2340 * Case combinators generated by the domain package for type "foo" are
  2341 now named "foo_case" instead of "foo_when".  INCOMPATIBILITY.
  2342 
  2343 * Several theorems have been renamed to more accurately reflect the
  2344 names of constants and types involved.  INCOMPATIBILITY.
  2345 
  2346   thelub_const    ~> lub_const
  2347   lub_const       ~> is_lub_const
  2348   thelubI         ~> lub_eqI
  2349   is_lub_lub      ~> is_lubD2
  2350   lubI            ~> is_lub_lub
  2351   unique_lub      ~> is_lub_unique
  2352   is_ub_lub       ~> is_lub_rangeD1
  2353   lub_bin_chain   ~> is_lub_bin_chain
  2354   lub_fun         ~> is_lub_fun
  2355   thelub_fun      ~> lub_fun
  2356   thelub_cfun     ~> lub_cfun
  2357   thelub_Pair     ~> lub_Pair
  2358   lub_cprod       ~> is_lub_prod
  2359   thelub_cprod    ~> lub_prod
  2360   minimal_cprod   ~> minimal_prod
  2361   inst_cprod_pcpo ~> inst_prod_pcpo
  2362   UU_I            ~> bottomI
  2363   compact_UU      ~> compact_bottom
  2364   deflation_UU    ~> deflation_bottom
  2365   finite_deflation_UU ~> finite_deflation_bottom
  2366 
  2367 * Many legacy theorem names have been discontinued.  INCOMPATIBILITY.
  2368 
  2369   sq_ord_less_eq_trans ~> below_eq_trans
  2370   sq_ord_eq_less_trans ~> eq_below_trans
  2371   refl_less            ~> below_refl
  2372   trans_less           ~> below_trans
  2373   antisym_less         ~> below_antisym
  2374   antisym_less_inverse ~> po_eq_conv [THEN iffD1]
  2375   box_less             ~> box_below
  2376   rev_trans_less       ~> rev_below_trans
  2377   not_less2not_eq      ~> not_below2not_eq
  2378   less_UU_iff          ~> below_UU_iff
  2379   flat_less_iff        ~> flat_below_iff
  2380   adm_less             ~> adm_below
  2381   adm_not_less         ~> adm_not_below
  2382   adm_compact_not_less ~> adm_compact_not_below
  2383   less_fun_def         ~> below_fun_def
  2384   expand_fun_less      ~> fun_below_iff
  2385   less_fun_ext         ~> fun_belowI
  2386   less_discr_def       ~> below_discr_def
  2387   discr_less_eq        ~> discr_below_eq
  2388   less_unit_def        ~> below_unit_def
  2389   less_cprod_def       ~> below_prod_def
  2390   prod_lessI           ~> prod_belowI
  2391   Pair_less_iff        ~> Pair_below_iff
  2392   fst_less_iff         ~> fst_below_iff
  2393   snd_less_iff         ~> snd_below_iff
  2394   expand_cfun_less     ~> cfun_below_iff
  2395   less_cfun_ext        ~> cfun_belowI
  2396   injection_less       ~> injection_below
  2397   less_up_def          ~> below_up_def
  2398   not_Iup_less         ~> not_Iup_below
  2399   Iup_less             ~> Iup_below
  2400   up_less              ~> up_below
  2401   Def_inject_less_eq   ~> Def_below_Def
  2402   Def_less_is_eq       ~> Def_below_iff
  2403   spair_less_iff       ~> spair_below_iff
  2404   less_sprod           ~> below_sprod
  2405   spair_less           ~> spair_below
  2406   sfst_less_iff        ~> sfst_below_iff
  2407   ssnd_less_iff        ~> ssnd_below_iff
  2408   fix_least_less       ~> fix_least_below
  2409   dist_less_one        ~> dist_below_one
  2410   less_ONE             ~> below_ONE
  2411   ONE_less_iff         ~> ONE_below_iff
  2412   less_sinlD           ~> below_sinlD
  2413   less_sinrD           ~> below_sinrD
  2414 
  2415 
  2416 *** FOL and ZF ***
  2417 
  2418 * All constant names are now qualified internally and use proper
  2419 identifiers, e.g. "IFOL.eq" instead of "op =".  INCOMPATIBILITY.
  2420 
  2421 
  2422 *** ML ***
  2423 
  2424 * Antiquotation @{assert} inlines a function bool -> unit that raises
  2425 Fail if the argument is false.  Due to inlining the source position of
  2426 failed assertions is included in the error output.
  2427 
  2428 * Discontinued antiquotation @{theory_ref}, which is obsolete since ML
  2429 text is in practice always evaluated with a stable theory checkpoint.
  2430 Minor INCOMPATIBILITY, use (Theory.check_thy @{theory}) instead.
  2431 
  2432 * Antiquotation @{theory A} refers to theory A from the ancestry of
  2433 the current context, not any accidental theory loader state as before.
  2434 Potential INCOMPATIBILITY, subtle change in semantics.
  2435 
  2436 * Syntax.pretty_priority (default 0) configures the required priority
  2437 of pretty-printed output and thus affects insertion of parentheses.
  2438 
  2439 * Syntax.default_root (default "any") configures the inner syntax
  2440 category (nonterminal symbol) for parsing of terms.
  2441 
  2442 * Former exception Library.UnequalLengths now coincides with
  2443 ListPair.UnequalLengths.
  2444 
  2445 * Renamed structure MetaSimplifier to Raw_Simplifier.  Note that the
  2446 main functionality is provided by structure Simplifier.
  2447 
  2448 * Renamed raw "explode" function to "raw_explode" to emphasize its
  2449 meaning.  Note that internally to Isabelle, Symbol.explode is used in
  2450 almost all situations.
  2451 
  2452 * Discontinued obsolete function sys_error and exception SYS_ERROR.
  2453 See implementation manual for further details on exceptions in
  2454 Isabelle/ML.
  2455 
  2456 * Renamed setmp_noncritical to Unsynchronized.setmp to emphasize its
  2457 meaning.
  2458 
  2459 * Renamed structure PureThy to Pure_Thy and moved most of its
  2460 operations to structure Global_Theory, to emphasize that this is
  2461 rarely-used global-only stuff.
  2462 
  2463 * Discontinued Output.debug.  Minor INCOMPATIBILITY, use plain writeln
  2464 instead (or tracing for high-volume output).
  2465 
  2466 * Configuration option show_question_marks only affects regular pretty
  2467 printing of types and terms, not raw Term.string_of_vname.
  2468 
  2469 * ML_Context.thm and ML_Context.thms are no longer pervasive.  Rare
  2470 INCOMPATIBILITY, superseded by static antiquotations @{thm} and
  2471 @{thms} for most purposes.
  2472 
  2473 * ML structure Unsynchronized is never opened, not even in Isar
  2474 interaction mode as before.  Old Unsynchronized.set etc. have been
  2475 discontinued -- use plain := instead.  This should be *rare* anyway,
  2476 since modern tools always work via official context data, notably
  2477 configuration options.
  2478 
  2479 * Parallel and asynchronous execution requires special care concerning
  2480 interrupts.  Structure Exn provides some convenience functions that
  2481 avoid working directly with raw Interrupt.  User code must not absorb
  2482 interrupts -- intermediate handling (for cleanup etc.) needs to be
  2483 followed by re-raising of the original exception.  Another common
  2484 source of mistakes are "handle _" patterns, which make the meaning of
  2485 the program subject to physical effects of the environment.
  2486 
  2487 
  2488 
  2489 New in Isabelle2009-2 (June 2010)
  2490 ---------------------------------
  2491 
  2492 *** General ***
  2493 
  2494 * Authentic syntax for *all* logical entities (type classes, type
  2495 constructors, term constants): provides simple and robust
  2496 correspondence between formal entities and concrete syntax.  Within
  2497 the parse tree / AST representations, "constants" are decorated by
  2498 their category (class, type, const) and spelled out explicitly with
  2499 their full internal name.
  2500 
  2501 Substantial INCOMPATIBILITY concerning low-level syntax declarations
  2502 and translations (translation rules and translation functions in ML).
  2503 Some hints on upgrading:
  2504 
  2505   - Many existing uses of 'syntax' and 'translations' can be replaced
  2506     by more modern 'type_notation', 'notation' and 'abbreviation',
  2507     which are independent of this issue.
  2508 
  2509   - 'translations' require markup within the AST; the term syntax
  2510     provides the following special forms:
  2511 
  2512       CONST c   -- produces syntax version of constant c from context
  2513       XCONST c  -- literally c, checked as constant from context
  2514       c         -- literally c, if declared by 'syntax'
  2515 
  2516     Plain identifiers are treated as AST variables -- occasionally the
  2517     system indicates accidental variables via the error "rhs contains
  2518     extra variables".
  2519 
  2520     Type classes and type constructors are marked according to their
  2521     concrete syntax.  Some old translations rules need to be written
  2522     for the "type" category, using type constructor application
  2523     instead of pseudo-term application of the default category
  2524     "logic".
  2525 
  2526   - 'parse_translation' etc. in ML may use the following
  2527     antiquotations:
  2528 
  2529       @{class_syntax c}   -- type class c within parse tree / AST
  2530       @{term_syntax c}    -- type constructor c within parse tree / AST
  2531       @{const_syntax c}   -- ML version of "CONST c" above
  2532       @{syntax_const c}   -- literally c (checked wrt. 'syntax' declarations)
  2533 
  2534   - Literal types within 'typed_print_translations', i.e. those *not*
  2535     represented as pseudo-terms are represented verbatim.  Use @{class
  2536     c} or @{type_name c} here instead of the above syntax
  2537     antiquotations.
  2538 
  2539 Note that old non-authentic syntax was based on unqualified base
  2540 names, so all of the above "constant" names would coincide.  Recall
  2541 that 'print_syntax' and ML_command "set Syntax.trace_ast" help to
  2542 diagnose syntax problems.
  2543 
  2544 * Type constructors admit general mixfix syntax, not just infix.
  2545 
  2546 * Concrete syntax may be attached to local entities without a proof
  2547 body, too.  This works via regular mixfix annotations for 'fix',
  2548 'def', 'obtain' etc. or via the explicit 'write' command, which is
  2549 similar to the 'notation' command in theory specifications.
  2550 
  2551 * Discontinued unnamed infix syntax (legacy feature for many years) --
  2552 need to specify constant name and syntax separately.  Internal ML
  2553 datatype constructors have been renamed from InfixName to Infix etc.
  2554 Minor INCOMPATIBILITY.
  2555 
  2556 * Schematic theorem statements need to be explicitly markup as such,
  2557 via commands 'schematic_lemma', 'schematic_theorem',
  2558 'schematic_corollary'.  Thus the relevance of the proof is made
  2559 syntactically clear, which impacts performance in a parallel or
  2560 asynchronous interactive environment.  Minor INCOMPATIBILITY.
  2561 
  2562 * Use of cumulative prems via "!" in some proof methods has been
  2563 discontinued (old legacy feature).
  2564 
  2565 * References 'trace_simp' and 'debug_simp' have been replaced by
  2566 configuration options stored in the context. Enabling tracing (the
  2567 case of debugging is similar) in proofs works via
  2568 
  2569   using [[trace_simp = true]]
  2570 
  2571 Tracing is then active for all invocations of the simplifier in
  2572 subsequent goal refinement steps. Tracing may also still be enabled or
  2573 disabled via the ProofGeneral settings menu.
  2574 
  2575 * Separate commands 'hide_class', 'hide_type', 'hide_const',
  2576 'hide_fact' replace the former 'hide' KIND command.  Minor
  2577 INCOMPATIBILITY.
  2578 
  2579 * Improved parallelism of proof term normalization: usedir -p2 -q0 is
  2580 more efficient than combinations with -q1 or -q2.
  2581 
  2582 
  2583 *** Pure ***
  2584 
  2585 * Proofterms record type-class reasoning explicitly, using the
  2586 "unconstrain" operation internally.  This eliminates all sort
  2587 constraints from a theorem and proof, introducing explicit
  2588 OFCLASS-premises.  On the proof term level, this operation is
  2589 automatically applied at theorem boundaries, such that closed proofs
  2590 are always free of sort constraints.  INCOMPATIBILITY for tools that
  2591 inspect proof terms.
  2592 
  2593 * Local theory specifications may depend on extra type variables that
  2594 are not present in the result type -- arguments TYPE('a) :: 'a itself
  2595 are added internally.  For example:
  2596 
  2597   definition unitary :: bool where "unitary = (ALL (x::'a) y. x = y)"
  2598 
  2599 * Predicates of locales introduced by classes carry a mandatory
  2600 "class" prefix.  INCOMPATIBILITY.
  2601 
  2602 * Vacuous class specifications observe default sort.  INCOMPATIBILITY.
  2603 
  2604 * Old 'axclass' command has been discontinued.  INCOMPATIBILITY, use
  2605 'class' instead.
  2606 
  2607 * Command 'code_reflect' allows to incorporate generated ML code into
  2608 runtime environment; replaces immature code_datatype antiquotation.
  2609 INCOMPATIBILITY.
  2610 
  2611 * Code generator: simple concept for abstract datatypes obeying
  2612 invariants.
  2613 
  2614 * Code generator: details of internal data cache have no impact on the
  2615 user space functionality any longer.
  2616 
  2617 * Methods "unfold_locales" and "intro_locales" ignore non-locale
  2618 subgoals.  This is more appropriate for interpretations with 'where'.
  2619 INCOMPATIBILITY.
  2620 
  2621 * Command 'example_proof' opens an empty proof body.  This allows to
  2622 experiment with Isar, without producing any persistent result.
  2623 
  2624 * Commands 'type_notation' and 'no_type_notation' declare type syntax
  2625 within a local theory context, with explicit checking of the
  2626 constructors involved (in contrast to the raw 'syntax' versions).
  2627 
  2628 * Commands 'types' and 'typedecl' now work within a local theory
  2629 context -- without introducing dependencies on parameters or
  2630 assumptions, which is not possible in Isabelle/Pure.
  2631 
  2632 * Command 'defaultsort' has been renamed to 'default_sort', it works
  2633 within a local theory context.  Minor INCOMPATIBILITY.
  2634 
  2635 
  2636 *** HOL ***
  2637 
  2638 * Command 'typedef' now works within a local theory context -- without
  2639 introducing dependencies on parameters or assumptions, which is not
  2640 possible in Isabelle/Pure/HOL.  Note that the logical environment may
  2641 contain multiple interpretations of local typedefs (with different
  2642 non-emptiness proofs), even in a global theory context.
  2643 
  2644 * New package for quotient types.  Commands 'quotient_type' and
  2645 'quotient_definition' may be used for defining types and constants by
  2646 quotient constructions.  An example is the type of integers created by
  2647 quotienting pairs of natural numbers:
  2648 
  2649   fun
  2650     intrel :: "(nat * nat) => (nat * nat) => bool"
  2651   where
  2652     "intrel (x, y) (u, v) = (x + v = u + y)"
  2653 
  2654   quotient_type int = "nat * nat" / intrel
  2655     by (auto simp add: equivp_def expand_fun_eq)
  2656 
  2657   quotient_definition
  2658     "0::int" is "(0::nat, 0::nat)"
  2659 
  2660 The method "lifting" can be used to lift of theorems from the
  2661 underlying "raw" type to the quotient type.  The example
  2662 src/HOL/Quotient_Examples/FSet.thy includes such a quotient
  2663 construction and provides a reasoning infrastructure for finite sets.
  2664 
  2665 * Renamed Library/Quotient.thy to Library/Quotient_Type.thy to avoid
  2666 clash with new theory Quotient in Main HOL.
  2667 
  2668 * Moved the SMT binding into the main HOL session, eliminating
  2669 separate HOL-SMT session.
  2670 
  2671 * List membership infix mem operation is only an input abbreviation.
  2672 INCOMPATIBILITY.
  2673 
  2674 * Theory Library/Word.thy has been removed.  Use library Word/Word.thy
  2675 for future developements; former Library/Word.thy is still present in
  2676 the AFP entry RSAPPS.
  2677 
  2678 * Theorem Int.int_induct renamed to Int.int_of_nat_induct and is no
  2679 longer shadowed.  INCOMPATIBILITY.
  2680 
  2681 * Dropped theorem duplicate comp_arith; use semiring_norm instead.
  2682 INCOMPATIBILITY.
  2683 
  2684 * Dropped theorem RealPow.real_sq_order; use power2_le_imp_le instead.
  2685 INCOMPATIBILITY.
  2686 
  2687 * Dropped normalizing_semiring etc; use the facts in semiring classes
  2688 instead.  INCOMPATIBILITY.
  2689 
  2690 * Dropped several real-specific versions of lemmas about floor and
  2691 ceiling; use the generic lemmas from theory "Archimedean_Field"
  2692 instead.  INCOMPATIBILITY.
  2693 
  2694   floor_number_of_eq         ~> floor_number_of
  2695   le_floor_eq_number_of      ~> number_of_le_floor
  2696   le_floor_eq_zero           ~> zero_le_floor
  2697   le_floor_eq_one            ~> one_le_floor
  2698   floor_less_eq_number_of    ~> floor_less_number_of
  2699   floor_less_eq_zero         ~> floor_less_zero
  2700   floor_less_eq_one          ~> floor_less_one
  2701   less_floor_eq_number_of    ~> number_of_less_floor
  2702   less_floor_eq_zero         ~> zero_less_floor
  2703   less_floor_eq_one          ~> one_less_floor
  2704   floor_le_eq_number_of      ~> floor_le_number_of
  2705   floor_le_eq_zero           ~> floor_le_zero
  2706   floor_le_eq_one            ~> floor_le_one
  2707   floor_subtract_number_of   ~> floor_diff_number_of
  2708   floor_subtract_one         ~> floor_diff_one
  2709   ceiling_number_of_eq       ~> ceiling_number_of
  2710   ceiling_le_eq_number_of    ~> ceiling_le_number_of
  2711   ceiling_le_zero_eq         ~> ceiling_le_zero
  2712   ceiling_le_eq_one          ~> ceiling_le_one
  2713   less_ceiling_eq_number_of  ~> number_of_less_ceiling
  2714   less_ceiling_eq_zero       ~> zero_less_ceiling
  2715   less_ceiling_eq_one        ~> one_less_ceiling
  2716   ceiling_less_eq_number_of  ~> ceiling_less_number_of
  2717   ceiling_less_eq_zero       ~> ceiling_less_zero
  2718   ceiling_less_eq_one        ~> ceiling_less_one
  2719   le_ceiling_eq_number_of    ~> number_of_le_ceiling
  2720   le_ceiling_eq_zero         ~> zero_le_ceiling
  2721   le_ceiling_eq_one          ~> one_le_ceiling
  2722   ceiling_subtract_number_of ~> ceiling_diff_number_of
  2723   ceiling_subtract_one       ~> ceiling_diff_one
  2724 
  2725 * Theory "Finite_Set": various folding_XXX locales facilitate the
  2726 application of the various fold combinators on finite sets.
  2727 
  2728 * Library theory "RBT" renamed to "RBT_Impl"; new library theory "RBT"
  2729 provides abstract red-black tree type which is backed by "RBT_Impl" as
  2730 implementation.  INCOMPATIBILTY.
  2731 
  2732 * Theory Library/Coinductive_List has been removed -- superseded by
  2733 AFP/thys/Coinductive.
  2734 
  2735 * Theory PReal, including the type "preal" and related operations, has
  2736 been removed.  INCOMPATIBILITY.
  2737 
  2738 * Real: new development using Cauchy Sequences.
  2739 
  2740 * Split off theory "Big_Operators" containing setsum, setprod,
  2741 Inf_fin, Sup_fin, Min, Max from theory Finite_Set.  INCOMPATIBILITY.
  2742 
  2743 * Theory "Rational" renamed to "Rat", for consistency with "Nat",
  2744 "Int" etc.  INCOMPATIBILITY.
  2745 
  2746 * Constant Rat.normalize needs to be qualified.  INCOMPATIBILITY.
  2747 
  2748 * New set of rules "ac_simps" provides combined assoc / commute
  2749 rewrites for all interpretations of the appropriate generic locales.
  2750 
  2751 * Renamed theory "OrderedGroup" to "Groups" and split theory
  2752 "Ring_and_Field" into theories "Rings" and "Fields"; for more
  2753 appropriate and more consistent names suitable for name prefixes
  2754 within the HOL theories.  INCOMPATIBILITY.
  2755 
  2756 * Some generic constants have been put to appropriate theories:
  2757   - less_eq, less: Orderings
  2758   - zero, one, plus, minus, uminus, times, abs, sgn: Groups
  2759   - inverse, divide: Rings
  2760 INCOMPATIBILITY.
  2761 
  2762 * More consistent naming of type classes involving orderings (and
  2763 lattices):
  2764 
  2765     lower_semilattice                   ~> semilattice_inf
  2766     upper_semilattice                   ~> semilattice_sup
  2767 
  2768     dense_linear_order                  ~> dense_linorder
  2769 
  2770     pordered_ab_group_add               ~> ordered_ab_group_add
  2771     pordered_ab_group_add_abs           ~> ordered_ab_group_add_abs
  2772     pordered_ab_semigroup_add           ~> ordered_ab_semigroup_add
  2773     pordered_ab_semigroup_add_imp_le    ~> ordered_ab_semigroup_add_imp_le
  2774     pordered_cancel_ab_semigroup_add    ~> ordered_cancel_ab_semigroup_add
  2775     pordered_cancel_comm_semiring       ~> ordered_cancel_comm_semiring
  2776     pordered_cancel_semiring            ~> ordered_cancel_semiring
  2777     pordered_comm_monoid_add            ~> ordered_comm_monoid_add
  2778     pordered_comm_ring                  ~> ordered_comm_ring
  2779     pordered_comm_semiring              ~> ordered_comm_semiring
  2780     pordered_ring                       ~> ordered_ring
  2781     pordered_ring_abs                   ~> ordered_ring_abs
  2782     pordered_semiring                   ~> ordered_semiring
  2783 
  2784     ordered_ab_group_add                ~> linordered_ab_group_add
  2785     ordered_ab_semigroup_add            ~> linordered_ab_semigroup_add
  2786     ordered_cancel_ab_semigroup_add     ~> linordered_cancel_ab_semigroup_add
  2787     ordered_comm_semiring_strict        ~> linordered_comm_semiring_strict
  2788     ordered_field                       ~> linordered_field
  2789     ordered_field_no_lb                 ~> linordered_field_no_lb
  2790     ordered_field_no_ub                 ~> linordered_field_no_ub
  2791     ordered_field_dense_linear_order    ~> dense_linordered_field
  2792     ordered_idom                        ~> linordered_idom
  2793     ordered_ring                        ~> linordered_ring
  2794     ordered_ring_le_cancel_factor       ~> linordered_ring_le_cancel_factor
  2795     ordered_ring_less_cancel_factor     ~> linordered_ring_less_cancel_factor
  2796     ordered_ring_strict                 ~> linordered_ring_strict
  2797     ordered_semidom                     ~> linordered_semidom
  2798     ordered_semiring                    ~> linordered_semiring
  2799     ordered_semiring_1                  ~> linordered_semiring_1
  2800     ordered_semiring_1_strict           ~> linordered_semiring_1_strict
  2801     ordered_semiring_strict             ~> linordered_semiring_strict
  2802 
  2803   The following slightly odd type classes have been moved to a
  2804   separate theory Library/Lattice_Algebras:
  2805 
  2806     lordered_ab_group_add               ~> lattice_ab_group_add
  2807     lordered_ab_group_add_abs           ~> lattice_ab_group_add_abs
  2808     lordered_ab_group_add_meet          ~> semilattice_inf_ab_group_add
  2809     lordered_ab_group_add_join          ~> semilattice_sup_ab_group_add
  2810     lordered_ring                       ~> lattice_ring
  2811 
  2812 INCOMPATIBILITY.
  2813 
  2814 * Refined field classes:
  2815   - classes division_ring_inverse_zero, field_inverse_zero,
  2816     linordered_field_inverse_zero include rule inverse 0 = 0 --
  2817     subsumes former division_by_zero class;
  2818   - numerous lemmas have been ported from field to division_ring.
  2819 INCOMPATIBILITY.
  2820 
  2821 * Refined algebra theorem collections:
  2822   - dropped theorem group group_simps, use algebra_simps instead;
  2823   - dropped theorem group ring_simps, use field_simps instead;
  2824   - proper theorem collection field_simps subsumes former theorem
  2825     groups field_eq_simps and field_simps;
  2826   - dropped lemma eq_minus_self_iff which is a duplicate for
  2827     equal_neg_zero.
  2828 INCOMPATIBILITY.
  2829 
  2830 * Theory Finite_Set and List: some lemmas have been generalized from
  2831 sets to lattices:
  2832 
  2833   fun_left_comm_idem_inter      ~> fun_left_comm_idem_inf
  2834   fun_left_comm_idem_union      ~> fun_left_comm_idem_sup
  2835   inter_Inter_fold_inter        ~> inf_Inf_fold_inf
  2836   union_Union_fold_union        ~> sup_Sup_fold_sup
  2837   Inter_fold_inter              ~> Inf_fold_inf
  2838   Union_fold_union              ~> Sup_fold_sup
  2839   inter_INTER_fold_inter        ~> inf_INFI_fold_inf
  2840   union_UNION_fold_union        ~> sup_SUPR_fold_sup
  2841   INTER_fold_inter              ~> INFI_fold_inf
  2842   UNION_fold_union              ~> SUPR_fold_sup
  2843 
  2844 * Theory "Complete_Lattice": lemmas top_def and bot_def have been
  2845 replaced by the more convenient lemmas Inf_empty and Sup_empty.
  2846 Dropped lemmas Inf_insert_simp and Sup_insert_simp, which are subsumed
  2847 by Inf_insert and Sup_insert.  Lemmas Inf_UNIV and Sup_UNIV replace
  2848 former Inf_Univ and Sup_Univ.  Lemmas inf_top_right and sup_bot_right
  2849 subsume inf_top and sup_bot respectively.  INCOMPATIBILITY.
  2850 
  2851 * Reorganized theory Multiset: swapped notation of pointwise and
  2852 multiset order:
  2853 
  2854   - pointwise ordering is instance of class order with standard syntax
  2855     <= and <;
  2856   - multiset ordering has syntax <=# and <#; partial order properties
  2857     are provided by means of interpretation with prefix
  2858     multiset_order;
  2859   - less duplication, less historical organization of sections,
  2860     conversion from associations lists to multisets, rudimentary code
  2861     generation;
  2862   - use insert_DiffM2 [symmetric] instead of elem_imp_eq_diff_union,
  2863     if needed.
  2864 
  2865 Renamed:
  2866 
  2867   multiset_eq_conv_count_eq  ~>  multiset_ext_iff
  2868   multi_count_ext  ~>  multiset_ext
  2869   diff_union_inverse2  ~>  diff_union_cancelR
  2870 
  2871 INCOMPATIBILITY.
  2872 
  2873 * Theory Permutation: replaced local "remove" by List.remove1.
  2874 
  2875 * Code generation: ML and OCaml code is decorated with signatures.
  2876 
  2877 * Theory List: added transpose.
  2878 
  2879 * Library/Nat_Bijection.thy is a collection of bijective functions
  2880 between nat and other types, which supersedes the older libraries
  2881 Library/Nat_Int_Bij.thy and HOLCF/NatIso.thy.  INCOMPATIBILITY.
  2882 
  2883   Constants:
  2884   Nat_Int_Bij.nat2_to_nat         ~> prod_encode
  2885   Nat_Int_Bij.nat_to_nat2         ~> prod_decode
  2886   Nat_Int_Bij.int_to_nat_bij      ~> int_encode
  2887   Nat_Int_Bij.nat_to_int_bij      ~> int_decode
  2888   Countable.pair_encode           ~> prod_encode
  2889   NatIso.prod2nat                 ~> prod_encode
  2890   NatIso.nat2prod                 ~> prod_decode
  2891   NatIso.sum2nat                  ~> sum_encode
  2892   NatIso.nat2sum                  ~> sum_decode
  2893   NatIso.list2nat                 ~> list_encode
  2894   NatIso.nat2list                 ~> list_decode
  2895   NatIso.set2nat                  ~> set_encode
  2896   NatIso.nat2set                  ~> set_decode
  2897 
  2898   Lemmas:
  2899   Nat_Int_Bij.bij_nat_to_int_bij  ~> bij_int_decode
  2900   Nat_Int_Bij.nat2_to_nat_inj     ~> inj_prod_encode
  2901   Nat_Int_Bij.nat2_to_nat_surj    ~> surj_prod_encode
  2902   Nat_Int_Bij.nat_to_nat2_inj     ~> inj_prod_decode
  2903   Nat_Int_Bij.nat_to_nat2_surj    ~> surj_prod_decode
  2904   Nat_Int_Bij.i2n_n2i_id          ~> int_encode_inverse
  2905   Nat_Int_Bij.n2i_i2n_id          ~> int_decode_inverse
  2906   Nat_Int_Bij.surj_nat_to_int_bij ~> surj_int_encode
  2907   Nat_Int_Bij.surj_int_to_nat_bij ~> surj_int_decode
  2908   Nat_Int_Bij.inj_nat_to_int_bij  ~> inj_int_encode
  2909   Nat_Int_Bij.inj_int_to_nat_bij  ~> inj_int_decode
  2910   Nat_Int_Bij.bij_nat_to_int_bij  ~> bij_int_encode
  2911   Nat_Int_Bij.bij_int_to_nat_bij  ~> bij_int_decode
  2912 
  2913 * Sledgehammer:
  2914   - Renamed ATP commands:
  2915     atp_info     ~> sledgehammer running_atps
  2916     atp_kill     ~> sledgehammer kill_atps
  2917     atp_messages ~> sledgehammer messages
  2918     atp_minimize ~> sledgehammer minimize
  2919     print_atps   ~> sledgehammer available_atps
  2920     INCOMPATIBILITY.
  2921   - Added user's manual ("isabelle doc sledgehammer").
  2922   - Added option syntax and "sledgehammer_params" to customize
  2923     Sledgehammer's behavior.  See the manual for details.
  2924   - Modified the Isar proof reconstruction code so that it produces
  2925     direct proofs rather than proofs by contradiction.  (This feature
  2926     is still experimental.)
  2927   - Made Isar proof reconstruction work for SPASS, remote ATPs, and in
  2928     full-typed mode.
  2929   - Added support for TPTP syntax for SPASS via the "spass_tptp" ATP.
  2930 
  2931 * Nitpick:
  2932   - Added and implemented "binary_ints" and "bits" options.
  2933   - Added "std" option and implemented support for nonstandard models.
  2934   - Added and implemented "finitize" option to improve the precision
  2935     of infinite datatypes based on a monotonicity analysis.
  2936   - Added support for quotient types.
  2937   - Added support for "specification" and "ax_specification"
  2938     constructs.
  2939   - Added support for local definitions (for "function" and
  2940     "termination" proofs).
  2941   - Added support for term postprocessors.
  2942   - Optimized "Multiset.multiset" and "FinFun.finfun".
  2943   - Improved efficiency of "destroy_constrs" optimization.
  2944   - Fixed soundness bugs related to "destroy_constrs" optimization and
  2945     record getters.
  2946   - Fixed soundness bug related to higher-order constructors.
  2947   - Fixed soundness bug when "full_descrs" is enabled.
  2948   - Improved precision of set constructs.
  2949   - Added "atoms" option.
  2950   - Added cache to speed up repeated Kodkod invocations on the same
  2951     problems.
  2952   - Renamed "MiniSatJNI", "zChaffJNI", "BerkMinAlloy", and
  2953     "SAT4JLight" to "MiniSat_JNI", "zChaff_JNI", "BerkMin_Alloy", and
  2954     "SAT4J_Light".  INCOMPATIBILITY.
  2955   - Removed "skolemize", "uncurry", "sym_break", "flatten_prop",
  2956     "sharing_depth", and "show_skolems" options.  INCOMPATIBILITY.
  2957   - Removed "nitpick_intro" attribute.  INCOMPATIBILITY.
  2958 
  2959 * Method "induct" now takes instantiations of the form t, where t is not
  2960   a variable, as a shorthand for "x == t", where x is a fresh variable.
  2961   If this is not intended, t has to be enclosed in parentheses.
  2962   By default, the equalities generated by definitional instantiations
  2963   are pre-simplified, which may cause parameters of inductive cases
  2964   to disappear, or may even delete some of the inductive cases.
  2965   Use "induct (no_simp)" instead of "induct" to restore the old
  2966   behaviour. The (no_simp) option is also understood by the "cases"
  2967   and "nominal_induct" methods, which now perform pre-simplification, too.
  2968   INCOMPATIBILITY.
  2969 
  2970 
  2971 *** HOLCF ***
  2972 
  2973 * Variable names in lemmas generated by the domain package have
  2974 changed; the naming scheme is now consistent with the HOL datatype
  2975 package.  Some proof scripts may be affected, INCOMPATIBILITY.
  2976 
  2977 * The domain package no longer defines the function "foo_copy" for
  2978 recursive domain "foo".  The reach lemma is now stated directly in
  2979 terms of "foo_take".  Lemmas and proofs that mention "foo_copy" must
  2980 be reformulated in terms of "foo_take", INCOMPATIBILITY.
  2981 
  2982 * Most definedness lemmas generated by the domain package (previously
  2983 of the form "x ~= UU ==> foo$x ~= UU") now have an if-and-only-if form
  2984 like "foo$x = UU <-> x = UU", which works better as a simp rule.
  2985 Proofs that used definedness lemmas as intro rules may break,
  2986 potential INCOMPATIBILITY.
  2987 
  2988 * Induction and casedist rules generated by the domain package now
  2989 declare proper case_names (one called "bottom", and one named for each
  2990 constructor).  INCOMPATIBILITY.
  2991 
  2992 * For mutually-recursive domains, separate "reach" and "take_lemma"
  2993 rules are generated for each domain, INCOMPATIBILITY.
  2994 
  2995   foo_bar.reach       ~> foo.reach  bar.reach
  2996   foo_bar.take_lemmas ~> foo.take_lemma  bar.take_lemma
  2997 
  2998 * Some lemmas generated by the domain package have been renamed for
  2999 consistency with the datatype package, INCOMPATIBILITY.
  3000 
  3001   foo.ind        ~> foo.induct
  3002   foo.finite_ind ~> foo.finite_induct
  3003   foo.coind      ~> foo.coinduct
  3004   foo.casedist   ~> foo.exhaust
  3005   foo.exhaust    ~> foo.nchotomy
  3006 
  3007 * For consistency with other definition packages, the fixrec package
  3008 now generates qualified theorem names, INCOMPATIBILITY.
  3009 
  3010   foo_simps  ~> foo.simps
  3011   foo_unfold ~> foo.unfold
  3012   foo_induct ~> foo.induct
  3013 
  3014 * The "fixrec_simp" attribute has been removed.  The "fixrec_simp"
  3015 method and internal fixrec proofs now use the default simpset instead.
  3016 INCOMPATIBILITY.
  3017 
  3018 * The "contlub" predicate has been removed.  Proof scripts should use
  3019 lemma contI2 in place of monocontlub2cont, INCOMPATIBILITY.
  3020 
  3021 * The "admw" predicate has been removed, INCOMPATIBILITY.
  3022 
  3023 * The constants cpair, cfst, and csnd have been removed in favor of
  3024 Pair, fst, and snd from Isabelle/HOL, INCOMPATIBILITY.
  3025 
  3026 
  3027 *** ML ***
  3028 
  3029 * Antiquotations for basic formal entities:
  3030 
  3031     @{class NAME}         -- type class
  3032     @{class_syntax NAME}  -- syntax representation of the above
  3033 
  3034     @{type_name NAME}     -- logical type
  3035     @{type_abbrev NAME}   -- type abbreviation
  3036     @{nonterminal NAME}   -- type of concrete syntactic category
  3037     @{type_syntax NAME}   -- syntax representation of any of the above
  3038 
  3039     @{const_name NAME}    -- logical constant (INCOMPATIBILITY)
  3040     @{const_abbrev NAME}  -- abbreviated constant
  3041     @{const_syntax NAME}  -- syntax representation of any of the above
  3042 
  3043 * Antiquotation @{syntax_const NAME} ensures that NAME refers to a raw
  3044 syntax constant (cf. 'syntax' command).
  3045 
  3046 * Antiquotation @{make_string} inlines a function to print arbitrary
  3047 values similar to the ML toplevel.  The result is compiler dependent
  3048 and may fall back on "?" in certain situations.
  3049 
  3050 * Diagnostic commands 'ML_val' and 'ML_command' may refer to
  3051 antiquotations @{Isar.state} and @{Isar.goal}.  This replaces impure
  3052 Isar.state() and Isar.goal(), which belong to the old TTY loop and do
  3053 not work with the asynchronous Isar document model.
  3054 
  3055 * Configuration options now admit dynamic default values, depending on
  3056 the context or even global references.
  3057 
  3058 * SHA1.digest digests strings according to SHA-1 (see RFC 3174).  It
  3059 uses an efficient external library if available (for Poly/ML).
  3060 
  3061 * Renamed some important ML structures, while keeping the old names
  3062 for some time as aliases within the structure Legacy:
  3063 
  3064   OuterKeyword  ~>  Keyword
  3065   OuterLex      ~>  Token
  3066   OuterParse    ~>  Parse
  3067   OuterSyntax   ~>  Outer_Syntax
  3068   PrintMode     ~>  Print_Mode
  3069   SpecParse     ~>  Parse_Spec
  3070   ThyInfo       ~>  Thy_Info
  3071   ThyLoad       ~>  Thy_Load
  3072   ThyOutput     ~>  Thy_Output
  3073   TypeInfer     ~>  Type_Infer
  3074 
  3075 Note that "open Legacy" simplifies porting of sources, but forgetting
  3076 to remove it again will complicate porting again in the future.
  3077 
  3078 * Most operations that refer to a global context are named
  3079 accordingly, e.g. Simplifier.global_context or
  3080 ProofContext.init_global.  There are some situations where a global
  3081 context actually works, but under normal circumstances one needs to
  3082 pass the proper local context through the code!
  3083 
  3084 * Discontinued old TheoryDataFun with its copy/init operation -- data
  3085 needs to be pure.  Functor Theory_Data_PP retains the traditional
  3086 Pretty.pp argument to merge, which is absent in the standard
  3087 Theory_Data version.
  3088 
  3089 * Sorts.certify_sort and derived "cert" operations for types and terms
  3090 no longer minimize sorts.  Thus certification at the boundary of the
  3091 inference kernel becomes invariant under addition of class relations,
  3092 which is an important monotonicity principle.  Sorts are now minimized
  3093 in the syntax layer only, at the boundary between the end-user and the
  3094 system.  Subtle INCOMPATIBILITY, may have to use Sign.minimize_sort
  3095 explicitly in rare situations.
  3096 
  3097 * Renamed old-style Drule.standard to Drule.export_without_context, to
  3098 emphasize that this is in no way a standard operation.
  3099 INCOMPATIBILITY.
  3100 
  3101 * Subgoal.FOCUS (and variants): resulting goal state is normalized as
  3102 usual for resolution.  Rare INCOMPATIBILITY.
  3103 
  3104 * Renamed varify/unvarify operations to varify_global/unvarify_global
  3105 to emphasize that these only work in a global situation (which is
  3106 quite rare).
  3107 
  3108 * Curried take and drop in library.ML; negative length is interpreted
  3109 as infinity (as in chop).  Subtle INCOMPATIBILITY.
  3110 
  3111 * Proof terms: type substitutions on proof constants now use canonical
  3112 order of type variables.  INCOMPATIBILITY for tools working with proof
  3113 terms.
  3114 
  3115 * Raw axioms/defs may no longer carry sort constraints, and raw defs
  3116 may no longer carry premises.  User-level specifications are
  3117 transformed accordingly by Thm.add_axiom/add_def.
  3118 
  3119 
  3120 *** System ***
  3121 
  3122 * Discontinued special HOL_USEDIR_OPTIONS for the main HOL image;
  3123 ISABELLE_USEDIR_OPTIONS applies uniformly to all sessions.  Note that
  3124 proof terms are enabled unconditionally in the new HOL-Proofs image.
  3125 
  3126 * Discontinued old ISABELLE and ISATOOL environment settings (legacy
  3127 feature since Isabelle2009).  Use ISABELLE_PROCESS and ISABELLE_TOOL,
  3128 respectively.
  3129 
  3130 * Old lib/scripts/polyml-platform is superseded by the
  3131 ISABELLE_PLATFORM setting variable, which defaults to the 32 bit
  3132 variant, even on a 64 bit machine.  The following example setting
  3133 prefers 64 bit if available:
  3134 
  3135   ML_PLATFORM="${ISABELLE_PLATFORM64:-$ISABELLE_PLATFORM}"
  3136 
  3137 * The preliminary Isabelle/jEdit application demonstrates the emerging
  3138 Isabelle/Scala layer for advanced prover interaction and integration.
  3139 See src/Tools/jEdit or "isabelle jedit" provided by the properly built
  3140 component.
  3141 
  3142 * "IsabelleText" is a Unicode font derived from Bitstream Vera Mono
  3143 and Bluesky TeX fonts.  It provides the usual Isabelle symbols,
  3144 similar to the default assignment of the document preparation system
  3145 (cf. isabellesym.sty).  The Isabelle/Scala class Isabelle_System
  3146 provides some operations for direct access to the font without asking
  3147 the user for manual installation.
  3148 
  3149 
  3150 
  3151 New in Isabelle2009-1 (December 2009)
  3152 -------------------------------------
  3153 
  3154 *** General ***
  3155 
  3156 * Discontinued old form of "escaped symbols" such as \\<forall>.  Only
  3157 one backslash should be used, even in ML sources.
  3158 
  3159 
  3160 *** Pure ***
  3161 
  3162 * Locale interpretation propagates mixins along the locale hierarchy.
  3163 The currently only available mixins are the equations used to map
  3164 local definitions to terms of the target domain of an interpretation.
  3165 
  3166 * Reactivated diagnostic command 'print_interps'.  Use "print_interps
  3167 loc" to print all interpretations of locale "loc" in the theory.
  3168 Interpretations in proofs are not shown.
  3169 
  3170 * Thoroughly revised locales tutorial.  New section on conditional
  3171 interpretation.
  3172 
  3173 * On instantiation of classes, remaining undefined class parameters
  3174 are formally declared.  INCOMPATIBILITY.
  3175 
  3176 
  3177 *** Document preparation ***
  3178 
  3179 * New generalized style concept for printing terms: @{foo (style) ...}
  3180 instead of @{foo_style style ...}  (old form is still retained for
  3181 backward compatibility).  Styles can be also applied for
  3182 antiquotations prop, term_type and typeof.
  3183 
  3184 
  3185 *** HOL ***
  3186 
  3187 * New proof method "smt" for a combination of first-order logic with
  3188 equality, linear and nonlinear (natural/integer/real) arithmetic, and
  3189 fixed-size bitvectors; there is also basic support for higher-order
  3190 features (esp. lambda abstractions).  It is an incomplete decision
  3191 procedure based on external SMT solvers using the oracle mechanism;
  3192 for the SMT solver Z3, this method is proof-producing.  Certificates
  3193 are provided to avoid calling the external solvers solely for
  3194 re-checking proofs.  Due to a remote SMT service there is no need for
  3195 installing SMT solvers locally.  See src/HOL/SMT.
  3196 
  3197 * New commands to load and prove verification conditions generated by
  3198 the Boogie program verifier or derived systems (e.g. the Verifying C
  3199 Compiler (VCC) or Spec#).  See src/HOL/Boogie.
  3200 
  3201 * New counterexample generator tool 'nitpick' based on the Kodkod
  3202 relational model finder.  See src/HOL/Tools/Nitpick and
  3203 src/HOL/Nitpick_Examples.
  3204 
  3205 * New commands 'code_pred' and 'values' to invoke the predicate
  3206 compiler and to enumerate values of inductive predicates.
  3207 
  3208 * A tabled implementation of the reflexive transitive closure.
  3209 
  3210 * New implementation of quickcheck uses generic code generator;
  3211 default generators are provided for all suitable HOL types, records
  3212 and datatypes.  Old quickcheck can be re-activated importing theory
  3213 Library/SML_Quickcheck.
  3214 
  3215 * New testing tool Mirabelle for automated proof tools.  Applies
  3216 several tools and tactics like sledgehammer, metis, or quickcheck, to
  3217 every proof step in a theory.  To be used in batch mode via the
  3218 "mirabelle" utility.
  3219 
  3220 * New proof method "sos" (sum of squares) for nonlinear real
  3221 arithmetic (originally due to John Harison). It requires theory
  3222 Library/Sum_Of_Squares.  It is not a complete decision procedure but
  3223 works well in practice on quantifier-free real arithmetic with +, -,
  3224 *, ^, =, <= and <, i.e. boolean combinations of equalities and
  3225 inequalities between polynomials.  It makes use of external
  3226 semidefinite programming solvers.  Method "sos" generates a
  3227 certificate that can be pasted into the proof thus avoiding the need
  3228 to call an external tool every time the proof is checked.  See
  3229 src/HOL/Library/Sum_Of_Squares.
  3230 
  3231 * New method "linarith" invokes existing linear arithmetic decision
  3232 procedure only.
  3233 
  3234 * New command 'atp_minimal' reduces result produced by Sledgehammer.
  3235 
  3236 * New Sledgehammer option "Full Types" in Proof General settings menu.
  3237 Causes full type information to be output to the ATPs.  This slows
  3238 ATPs down considerably but eliminates a source of unsound "proofs"
  3239 that fail later.
  3240 
  3241 * New method "metisFT": A version of metis that uses full type
  3242 information in order to avoid failures of proof reconstruction.
  3243 
  3244 * New evaluator "approximate" approximates an real valued term using
  3245 the same method as the approximation method.
  3246 
  3247 * Method "approximate" now supports arithmetic expressions as
  3248 boundaries of intervals and implements interval splitting and Taylor
  3249 series expansion.
  3250 
  3251 * ML antiquotation @{code_datatype} inserts definition of a datatype
  3252 generated by the code generator; e.g. see src/HOL/Predicate.thy.
  3253 
  3254 * New theory SupInf of the supremum and infimum operators for sets of
  3255 reals.
  3256 
  3257 * New theory Probability, which contains a development of measure
  3258 theory, eventually leading to Lebesgue integration and probability.
  3259 
  3260 * Extended Multivariate Analysis to include derivation and Brouwer's
  3261 fixpoint theorem.
  3262 
  3263 * Reorganization of number theory, INCOMPATIBILITY:
  3264   - new number theory development for nat and int, in theories Divides
  3265     and GCD as well as in new session Number_Theory
  3266   - some constants and facts now suffixed with _nat and _int
  3267     accordingly
  3268   - former session NumberTheory now named Old_Number_Theory, including
  3269     theories Legacy_GCD and Primes (prefer Number_Theory if possible)
  3270   - moved theory Pocklington from src/HOL/Library to
  3271     src/HOL/Old_Number_Theory
  3272 
  3273 * Theory GCD includes functions Gcd/GCD and Lcm/LCM for the gcd and
  3274 lcm of finite and infinite sets. It is shown that they form a complete
  3275 lattice.
  3276 
  3277 * Class semiring_div requires superclass no_zero_divisors and proof of
  3278 div_mult_mult1; theorems div_mult_mult1, div_mult_mult2,
  3279 div_mult_mult1_if, div_mult_mult1 and div_mult_mult2 have been
  3280 generalized to class semiring_div, subsuming former theorems
  3281 zdiv_zmult_zmult1, zdiv_zmult_zmult1_if, zdiv_zmult_zmult1 and
  3282 zdiv_zmult_zmult2.  div_mult_mult1 is now [simp] by default.
  3283 INCOMPATIBILITY.
  3284 
  3285 * Refinements to lattice classes and sets:
  3286   - less default intro/elim rules in locale variant, more default
  3287     intro/elim rules in class variant: more uniformity
  3288   - lemma ge_sup_conv renamed to le_sup_iff, in accordance with
  3289     le_inf_iff
  3290   - dropped lemma alias inf_ACI for inf_aci (same for sup_ACI and
  3291     sup_aci)
  3292   - renamed ACI to inf_sup_aci
  3293   - new class "boolean_algebra"
  3294   - class "complete_lattice" moved to separate theory
  3295     "Complete_Lattice"; corresponding constants (and abbreviations)
  3296     renamed and with authentic syntax:
  3297     Set.Inf ~>    Complete_Lattice.Inf
  3298     Set.Sup ~>    Complete_Lattice.Sup
  3299     Set.INFI ~>   Complete_Lattice.INFI
  3300     Set.SUPR ~>   Complete_Lattice.SUPR
  3301     Set.Inter ~>  Complete_Lattice.Inter
  3302     Set.Union ~>  Complete_Lattice.Union
  3303     Set.INTER ~>  Complete_Lattice.INTER
  3304     Set.UNION ~>  Complete_Lattice.UNION
  3305   - authentic syntax for
  3306     Set.Pow
  3307     Set.image
  3308   - mere abbreviations:
  3309     Set.empty               (for bot)
  3310     Set.UNIV                (for top)
  3311     Set.inter               (for inf, formerly Set.Int)
  3312     Set.union               (for sup, formerly Set.Un)
  3313     Complete_Lattice.Inter  (for Inf)
  3314     Complete_Lattice.Union  (for Sup)
  3315     Complete_Lattice.INTER  (for INFI)
  3316     Complete_Lattice.UNION  (for SUPR)
  3317   - object-logic definitions as far as appropriate
  3318 
  3319 INCOMPATIBILITY.  Care is required when theorems Int_subset_iff or
  3320 Un_subset_iff are explicitly deleted as default simp rules; then also
  3321 their lattice counterparts le_inf_iff and le_sup_iff have to be
  3322 deleted to achieve the desired effect.
  3323 
  3324 * Rules inf_absorb1, inf_absorb2, sup_absorb1, sup_absorb2 are no simp
  3325 rules by default any longer; the same applies to min_max.inf_absorb1
  3326 etc.  INCOMPATIBILITY.
  3327 
  3328 * Rules sup_Int_eq and sup_Un_eq are no longer declared as
  3329 pred_set_conv by default.  INCOMPATIBILITY.
  3330 
  3331 * Power operations on relations and functions are now one dedicated
  3332 constant "compow" with infix syntax "^^".  Power operation on
  3333 multiplicative monoids retains syntax "^" and is now defined generic
  3334 in class power.  INCOMPATIBILITY.
  3335 
  3336 * Relation composition "R O S" now has a more standard argument order:
  3337 "R O S = {(x, z). EX y. (x, y) : R & (y, z) : S}".  INCOMPATIBILITY,
  3338 rewrite propositions with "S O R" --> "R O S". Proofs may occasionally
  3339 break, since the O_assoc rule was not rewritten like this.  Fix using
  3340 O_assoc[symmetric].  The same applies to the curried version "R OO S".
  3341 
  3342 * Function "Inv" is renamed to "inv_into" and function "inv" is now an
  3343 abbreviation for "inv_into UNIV".  Lemmas are renamed accordingly.
  3344 INCOMPATIBILITY.
  3345 
  3346 * Most rules produced by inductive and datatype package have mandatory
  3347 prefixes.  INCOMPATIBILITY.
  3348 
  3349 * Changed "DERIV_intros" to a dynamic fact, which can be augmented by
  3350 the attribute of the same name.  Each of the theorems in the list
  3351 DERIV_intros assumes composition with an additional function and
  3352 matches a variable to the derivative, which has to be solved by the
  3353 Simplifier.  Hence (auto intro!: DERIV_intros) computes the derivative
  3354 of most elementary terms.  Former Maclauren.DERIV_tac and
  3355 Maclauren.deriv_tac should be replaced by (auto intro!: DERIV_intros).
  3356 INCOMPATIBILITY.
  3357 
  3358 * Code generator attributes follow the usual underscore convention:
  3359     code_unfold     replaces    code unfold
  3360     code_post       replaces    code post
  3361     etc.
  3362   INCOMPATIBILITY.
  3363 
  3364 * Renamed methods:
  3365     sizechange -> size_change
  3366     induct_scheme -> induction_schema
  3367   INCOMPATIBILITY.
  3368 
  3369 * Discontinued abbreviation "arbitrary" of constant "undefined".
  3370 INCOMPATIBILITY, use "undefined" directly.
  3371 
  3372 * Renamed theorems:
  3373     Suc_eq_add_numeral_1 -> Suc_eq_plus1
  3374     Suc_eq_add_numeral_1_left -> Suc_eq_plus1_left
  3375     Suc_plus1 -> Suc_eq_plus1
  3376     *anti_sym -> *antisym*
  3377     vector_less_eq_def -> vector_le_def
  3378   INCOMPATIBILITY.
  3379 
  3380 * Added theorem List.map_map as [simp].  Removed List.map_compose.
  3381 INCOMPATIBILITY.
  3382 
  3383 * Removed predicate "M hassize n" (<--> card M = n & finite M).
  3384 INCOMPATIBILITY.
  3385 
  3386 
  3387 *** HOLCF ***
  3388 
  3389 * Theory Representable defines a class "rep" of domains that are
  3390 representable (via an ep-pair) in the universal domain type "udom".
  3391 Instances are provided for all type constructors defined in HOLCF.
  3392 
  3393 * The 'new_domain' command is a purely definitional version of the
  3394 domain package, for representable domains.  Syntax is identical to the
  3395 old domain package.  The 'new_domain' package also supports indirect
  3396 recursion using previously-defined type constructors.  See
  3397 src/HOLCF/ex/New_Domain.thy for examples.
  3398 
  3399 * Method "fixrec_simp" unfolds one step of a fixrec-defined constant
  3400 on the left-hand side of an equation, and then performs
  3401 simplification.  Rewriting is done using rules declared with the
  3402 "fixrec_simp" attribute.  The "fixrec_simp" method is intended as a
  3403 replacement for "fixpat"; see src/HOLCF/ex/Fixrec_ex.thy for examples.
  3404 
  3405 * The pattern-match compiler in 'fixrec' can now handle constructors
  3406 with HOL function types.  Pattern-match combinators for the Pair
  3407 constructor are pre-configured.
  3408 
  3409 * The 'fixrec' package now produces better fixed-point induction rules
  3410 for mutually-recursive definitions:  Induction rules have conclusions
  3411 of the form "P foo bar" instead of "P <foo, bar>".
  3412 
  3413 * The constant "sq_le" (with infix syntax "<<" or "\<sqsubseteq>") has
  3414 been renamed to "below".  The name "below" now replaces "less" in many
  3415 theorem names.  (Legacy theorem names using "less" are still supported
  3416 as well.)
  3417 
  3418 * The 'fixrec' package now supports "bottom patterns".  Bottom
  3419 patterns can be used to generate strictness rules, or to make
  3420 functions more strict (much like the bang-patterns supported by the
  3421 Glasgow Haskell Compiler).  See src/HOLCF/ex/Fixrec_ex.thy for
  3422 examples.
  3423 
  3424 
  3425 *** ML ***
  3426 
  3427 * Support for Poly/ML 5.3.0, with improved reporting of compiler
  3428 errors and run-time exceptions, including detailed source positions.
  3429 
  3430 * Structure Name_Space (formerly NameSpace) now manages uniquely
  3431 identified entries, with some additional information such as source
  3432 position, logical grouping etc.
  3433 
  3434 * Theory and context data is now introduced by the simplified and
  3435 modernized functors Theory_Data, Proof_Data, Generic_Data.  Data needs
  3436 to be pure, but the old TheoryDataFun for mutable data (with explicit
  3437 copy operation) is still available for some time.
  3438 
  3439 * Structure Synchronized (cf. src/Pure/Concurrent/synchronized.ML)
  3440 provides a high-level programming interface to synchronized state
  3441 variables with atomic update.  This works via pure function
  3442 application within a critical section -- its runtime should be as
  3443 short as possible; beware of deadlocks if critical code is nested,
  3444 either directly or indirectly via other synchronized variables!
  3445 
  3446 * Structure Unsynchronized (cf. src/Pure/ML-Systems/unsynchronized.ML)
  3447 wraps raw ML references, explicitly indicating their non-thread-safe
  3448 behaviour.  The Isar toplevel keeps this structure open, to
  3449 accommodate Proof General as well as quick and dirty interactive
  3450 experiments with references.
  3451 
  3452 * PARALLEL_CHOICE and PARALLEL_GOALS provide basic support for
  3453 parallel tactical reasoning.
  3454 
  3455 * Tacticals Subgoal.FOCUS, Subgoal.FOCUS_PREMS, Subgoal.FOCUS_PARAMS
  3456 are similar to SUBPROOF, but are slightly more flexible: only the
  3457 specified parts of the subgoal are imported into the context, and the
  3458 body tactic may introduce new subgoals and schematic variables.
  3459 
  3460 * Old tactical METAHYPS, which does not observe the proof context, has
  3461 been renamed to Old_Goals.METAHYPS and awaits deletion.  Use SUBPROOF
  3462 or Subgoal.FOCUS etc.
  3463 
  3464 * Renamed functor TableFun to Table, and GraphFun to Graph.  (Since
  3465 functors have their own ML name space there is no point to mark them
  3466 separately.)  Minor INCOMPATIBILITY.
  3467 
  3468 * Renamed NamedThmsFun to Named_Thms.  INCOMPATIBILITY.
  3469 
  3470 * Renamed several structures FooBar to Foo_Bar.  Occasional,
  3471 INCOMPATIBILITY.
  3472 
  3473 * Operations of structure Skip_Proof no longer require quick_and_dirty
  3474 mode, which avoids critical setmp.
  3475 
  3476 * Eliminated old Attrib.add_attributes, Method.add_methods and related
  3477 combinators for "args".  INCOMPATIBILITY, need to use simplified
  3478 Attrib/Method.setup introduced in Isabelle2009.
  3479 
  3480 * Proper context for simpset_of, claset_of, clasimpset_of.  May fall
  3481 back on global_simpset_of, global_claset_of, global_clasimpset_of as
  3482 last resort.  INCOMPATIBILITY.
  3483 
  3484 * Display.pretty_thm now requires a proper context (cf. former
  3485 ProofContext.pretty_thm).  May fall back on Display.pretty_thm_global
  3486 or even Display.pretty_thm_without_context as last resort.
  3487 INCOMPATIBILITY.
  3488 
  3489 * Discontinued Display.pretty_ctyp/cterm etc.  INCOMPATIBILITY, use
  3490 Syntax.pretty_typ/term directly, preferably with proper context
  3491 instead of global theory.
  3492 
  3493 
  3494 *** System ***
  3495 
  3496 * Further fine tuning of parallel proof checking, scales up to 8 cores
  3497 (max. speedup factor 5.0).  See also Goal.parallel_proofs in ML and
  3498 usedir option -q.
  3499 
  3500 * Support for additional "Isabelle components" via etc/components, see
  3501 also the system manual.
  3502 
  3503 * The isabelle makeall tool now operates on all components with
  3504 IsaMakefile, not just hardwired "logics".
  3505 
  3506 * Removed "compress" option from isabelle-process and isabelle usedir;
  3507 this is always enabled.
  3508 
  3509 * Discontinued support for Poly/ML 4.x versions.
  3510 
  3511 * Isabelle tool "wwwfind" provides web interface for 'find_theorems'
  3512 on a given logic image.  This requires the lighttpd webserver and is
  3513 currently supported on Linux only.
  3514 
  3515 
  3516 
  3517 New in Isabelle2009 (April 2009)
  3518 --------------------------------
  3519 
  3520 *** General ***
  3521 
  3522 * Simplified main Isabelle executables, with less surprises on
  3523 case-insensitive file-systems (such as Mac OS).
  3524 
  3525   - The main Isabelle tool wrapper is now called "isabelle" instead of
  3526     "isatool."
  3527 
  3528   - The former "isabelle" alias for "isabelle-process" has been
  3529     removed (should rarely occur to regular users).
  3530 
  3531   - The former "isabelle-interface" and its alias "Isabelle" have been
  3532     removed (interfaces are now regular Isabelle tools).
  3533 
  3534 Within scripts and make files, the Isabelle environment variables
  3535 ISABELLE_TOOL and ISABELLE_PROCESS replace old ISATOOL and ISABELLE,
  3536 respectively.  (The latter are still available as legacy feature.)
  3537 
  3538 The old isabelle-interface wrapper could react in confusing ways if
  3539 the interface was uninstalled or changed otherwise.  Individual
  3540 interface tool configuration is now more explicit, see also the
  3541 Isabelle system manual.  In particular, Proof General is now available
  3542 via "isabelle emacs".
  3543 
  3544 INCOMPATIBILITY, need to adapt derivative scripts.  Users may need to
  3545 purge installed copies of Isabelle executables and re-run "isabelle
  3546 install -p ...", or use symlinks.
  3547 
  3548 * The default for ISABELLE_HOME_USER is now ~/.isabelle instead of the
  3549 old ~/isabelle, which was slightly non-standard and apt to cause
  3550 surprises on case-insensitive file-systems (such as Mac OS).
  3551 
  3552 INCOMPATIBILITY, need to move existing ~/isabelle/etc,
  3553 ~/isabelle/heaps, ~/isabelle/browser_info to the new place.  Special
  3554 care is required when using older releases of Isabelle.  Note that
  3555 ISABELLE_HOME_USER can be changed in Isabelle/etc/settings of any
  3556 Isabelle distribution, in order to use the new ~/.isabelle uniformly.
  3557 
  3558 * Proofs of fully specified statements are run in parallel on
  3559 multi-core systems.  A speedup factor of 2.5 to 3.2 can be expected on
  3560 a regular 4-core machine, if the initial heap space is made reasonably
  3561 large (cf. Poly/ML option -H).  (Requires Poly/ML 5.2.1 or later.)
  3562 
  3563 * The main reference manuals ("isar-ref", "implementation", and
  3564 "system") have been updated and extended.  Formally checked references
  3565 as hyperlinks are now available uniformly.
  3566 
  3567 
  3568 *** Pure ***
  3569 
  3570 * Complete re-implementation of locales.  INCOMPATIBILITY in several
  3571 respects.  The most important changes are listed below.  See the
  3572 Tutorial on Locales ("locales" manual) for details.
  3573 
  3574 - In locale expressions, instantiation replaces renaming.  Parameters
  3575 must be declared in a for clause.  To aid compatibility with previous
  3576 parameter inheritance, in locale declarations, parameters that are not
  3577 'touched' (instantiation position "_" or omitted) are implicitly added
  3578 with their syntax at the beginning of the for clause.
  3579 
  3580 - Syntax from abbreviations and definitions in locales is available in
  3581 locale expressions and context elements.  The latter is particularly
  3582 useful in locale declarations.
  3583 
  3584 - More flexible mechanisms to qualify names generated by locale
  3585 expressions.  Qualifiers (prefixes) may be specified in locale
  3586 expressions, and can be marked as mandatory (syntax: "name!:") or
  3587 optional (syntax "name?:").  The default depends for plain "name:"
  3588 depends on the situation where a locale expression is used: in
  3589 commands 'locale' and 'sublocale' prefixes are optional, in
  3590 'interpretation' and 'interpret' prefixes are mandatory.  The old
  3591 implicit qualifiers derived from the parameter names of a locale are
  3592 no longer generated.
  3593 
  3594 - Command "sublocale l < e" replaces "interpretation l < e".  The
  3595 instantiation clause in "interpretation" and "interpret" (square
  3596 brackets) is no longer available.  Use locale expressions.
  3597 
  3598 - When converting proof scripts, mandatory qualifiers in
  3599 'interpretation' and 'interpret' should be retained by default, even
  3600 if this is an INCOMPATIBILITY compared to former behavior.  In the
  3601 worst case, use the "name?:" form for non-mandatory ones.  Qualifiers
  3602 in locale expressions range over a single locale instance only.
  3603 
  3604 - Dropped locale element "includes".  This is a major INCOMPATIBILITY.
  3605 In existing theorem specifications replace the includes element by the
  3606 respective context elements of the included locale, omitting those
  3607 that are already present in the theorem specification.  Multiple
  3608 assume elements of a locale should be replaced by a single one
  3609 involving the locale predicate.  In the proof body, declarations (most
  3610 notably theorems) may be regained by interpreting the respective
  3611 locales in the proof context as required (command "interpret").
  3612 
  3613 If using "includes" in replacement of a target solely because the
  3614 parameter types in the theorem are not as general as in the target,
  3615 consider declaring a new locale with additional type constraints on
  3616 the parameters (context element "constrains").
  3617 
  3618 - Discontinued "locale (open)".  INCOMPATIBILITY.
  3619 
  3620 - Locale interpretation commands no longer attempt to simplify goal.
  3621 INCOMPATIBILITY: in rare situations the generated goal differs.  Use
  3622 methods intro_locales and unfold_locales to clarify.
  3623 
  3624 - Locale interpretation commands no longer accept interpretation
  3625 attributes.  INCOMPATIBILITY.
  3626 
  3627 * Class declaration: so-called "base sort" must not be given in import
  3628 list any longer, but is inferred from the specification.  Particularly
  3629 in HOL, write
  3630 
  3631     class foo = ...
  3632 
  3633 instead of
  3634 
  3635     class foo = type + ...
  3636 
  3637 * Class target: global versions of theorems stemming do not carry a
  3638 parameter prefix any longer.  INCOMPATIBILITY.
  3639 
  3640 * Class 'instance' command no longer accepts attached definitions.
  3641 INCOMPATIBILITY, use proper 'instantiation' target instead.
  3642 
  3643 * Recovered hiding of consts, which was accidentally broken in
  3644 Isabelle2007.  Potential INCOMPATIBILITY, ``hide const c'' really
  3645 makes c inaccessible; consider using ``hide (open) const c'' instead.
  3646 
  3647 * Slightly more coherent Pure syntax, with updated documentation in
  3648 isar-ref manual.  Removed locales meta_term_syntax and
  3649 meta_conjunction_syntax: TERM and &&& (formerly &&) are now permanent,
  3650 INCOMPATIBILITY in rare situations.  Note that &&& should not be used
  3651 directly in regular applications.
  3652 
  3653 * There is a new syntactic category "float_const" for signed decimal
  3654 fractions (e.g. 123.45 or -123.45).
  3655 
  3656 * Removed exotic 'token_translation' command.  INCOMPATIBILITY, use ML
  3657 interface with 'setup' command instead.
  3658 
  3659 * Command 'local_setup' is similar to 'setup', but operates on a local
  3660 theory context.
  3661 
  3662 * The 'axiomatization' command now only works within a global theory
  3663 context.  INCOMPATIBILITY.
  3664 
  3665 * Goal-directed proof now enforces strict proof irrelevance wrt. sort
  3666 hypotheses.  Sorts required in the course of reasoning need to be
  3667 covered by the constraints in the initial statement, completed by the
  3668 type instance information of the background theory.  Non-trivial sort
  3669 hypotheses, which rarely occur in practice, may be specified via
  3670 vacuous propositions of the form SORT_CONSTRAINT('a::c).  For example:
  3671 
  3672   lemma assumes "SORT_CONSTRAINT('a::empty)" shows False ...
  3673 
  3674 The result contains an implicit sort hypotheses as before --
  3675 SORT_CONSTRAINT premises are eliminated as part of the canonical rule
  3676 normalization.
  3677 
  3678 * Generalized Isar history, with support for linear undo, direct state
  3679 addressing etc.
  3680 
  3681 * Changed defaults for unify configuration options:
  3682 
  3683   unify_trace_bound = 50 (formerly 25)
  3684   unify_search_bound = 60 (formerly 30)
  3685 
  3686 * Different bookkeeping for code equations (INCOMPATIBILITY):
  3687 
  3688   a) On theory merge, the last set of code equations for a particular
  3689      constant is taken (in accordance with the policy applied by other
  3690      parts of the code generator framework).
  3691 
  3692   b) Code equations stemming from explicit declarations (e.g. code
  3693      attribute) gain priority over default code equations stemming
  3694      from definition, primrec, fun etc.
  3695 
  3696 * Keyword 'code_exception' now named 'code_abort'.  INCOMPATIBILITY.
  3697 
  3698 * Unified theorem tables for both code generators.  Thus [code
  3699 func] has disappeared and only [code] remains.  INCOMPATIBILITY.
  3700 
  3701 * Command 'find_consts' searches for constants based on type and name
  3702 patterns, e.g.
  3703 
  3704     find_consts "_ => bool"
  3705 
  3706 By default, matching is against subtypes, but it may be restricted to
  3707 the whole type.  Searching by name is possible.  Multiple queries are
  3708 conjunctive and queries may be negated by prefixing them with a
  3709 hyphen:
  3710 
  3711     find_consts strict: "_ => bool" name: "Int" -"int => int"
  3712 
  3713 * New 'find_theorems' criterion "solves" matches theorems that
  3714 directly solve the current goal (modulo higher-order unification).
  3715 
  3716 * Auto solve feature for main theorem statements: whenever a new goal
  3717 is stated, "find_theorems solves" is called; any theorems that could
  3718 solve the lemma directly are listed as part of the goal state.
  3719 Cf. associated options in Proof General Isabelle settings menu,
  3720 enabled by default, with reasonable timeout for pathological cases of
  3721 higher-order unification.
  3722 
  3723 
  3724 *** Document preparation ***
  3725 
  3726 * Antiquotation @{lemma} now imitates a regular terminal proof,
  3727 demanding keyword 'by' and supporting the full method expression
  3728 syntax just like the Isar command 'by'.
  3729 
  3730 
  3731 *** HOL ***
  3732 
  3733 * Integrated main parts of former image HOL-Complex with HOL.  Entry
  3734 points Main and Complex_Main remain as before.
  3735 
  3736 * Logic image HOL-Plain provides a minimal HOL with the most important
  3737 tools available (inductive, datatype, primrec, ...).  This facilitates
  3738 experimentation and tool development.  Note that user applications
  3739 (and library theories) should never refer to anything below theory
  3740 Main, as before.
  3741 
  3742 * Logic image HOL-Main stops at theory Main, and thus facilitates
  3743 experimentation due to shorter build times.
  3744 
  3745 * Logic image HOL-NSA contains theories of nonstandard analysis which
  3746 were previously part of former HOL-Complex.  Entry point Hyperreal
  3747 remains valid, but theories formerly using Complex_Main should now use
  3748 new entry point Hypercomplex.
  3749 
  3750 * Generic ATP manager for Sledgehammer, based on ML threads instead of
  3751 Posix processes.  Avoids potentially expensive forking of the ML
  3752 process.  New thread-based implementation also works on non-Unix
  3753 platforms (Cygwin).  Provers are no longer hardwired, but defined
  3754 within the theory via plain ML wrapper functions.  Basic Sledgehammer
  3755 commands are covered in the isar-ref manual.
  3756 
  3757 * Wrapper scripts for remote SystemOnTPTP service allows to use
  3758 sledgehammer without local ATP installation (Vampire etc.). Other
  3759 provers may be included via suitable ML wrappers, see also
  3760 src/HOL/ATP_Linkup.thy.
  3761 
  3762 * ATP selection (E/Vampire/Spass) is now via Proof General's settings
  3763 menu.
  3764 
  3765 * The metis method no longer fails because the theorem is too trivial
  3766 (contains the empty clause).
  3767 
  3768 * The metis method now fails in the usual manner, rather than raising
  3769 an exception, if it determines that it cannot prove the theorem.
  3770 
  3771 * Method "coherent" implements a prover for coherent logic (see also
  3772 src/Tools/coherent.ML).
  3773 
  3774 * Constants "undefined" and "default" replace "arbitrary".  Usually
  3775 "undefined" is the right choice to replace "arbitrary", though
  3776 logically there is no difference.  INCOMPATIBILITY.
  3777 
  3778 * Command "value" now integrates different evaluation mechanisms.  The
  3779 result of the first successful evaluation mechanism is printed.  In
  3780 square brackets a particular named evaluation mechanisms may be
  3781 specified (currently, [SML], [code] or [nbe]).  See further
  3782 src/HOL/ex/Eval_Examples.thy.
  3783 
  3784 * Normalization by evaluation now allows non-leftlinear equations.
  3785 Declare with attribute [code nbe].
  3786 
  3787 * Methods "case_tac" and "induct_tac" now refer to the very same rules
  3788 as the structured Isar versions "cases" and "induct", cf. the
  3789 corresponding "cases" and "induct" attributes.  Mutual induction rules
  3790 are now presented as a list of individual projections
  3791 (e.g. foo_bar.inducts for types foo and bar); the old format with
  3792 explicit HOL conjunction is no longer supported.  INCOMPATIBILITY, in
  3793 rare situations a different rule is selected --- notably nested tuple
  3794 elimination instead of former prod.exhaust: use explicit (case_tac t
  3795 rule: prod.exhaust) here.
  3796 
  3797 * Attributes "cases", "induct", "coinduct" support "del" option.
  3798 
  3799 * Removed fact "case_split_thm", which duplicates "case_split".
  3800 
  3801 * The option datatype has been moved to a new theory Option.  Renamed
  3802 option_map to Option.map, and o2s to Option.set, INCOMPATIBILITY.
  3803 
  3804 * New predicate "strict_mono" classifies strict functions on partial
  3805 orders.  With strict functions on linear orders, reasoning about
  3806 (in)equalities is facilitated by theorems "strict_mono_eq",
  3807 "strict_mono_less_eq" and "strict_mono_less".
  3808 
  3809 * Some set operations are now proper qualified constants with
  3810 authentic syntax.  INCOMPATIBILITY:
  3811 
  3812     op Int ~>   Set.Int
  3813     op Un ~>    Set.Un
  3814     INTER ~>    Set.INTER
  3815     UNION ~>    Set.UNION
  3816     Inter ~>    Set.Inter
  3817     Union ~>    Set.Union
  3818     {} ~>       Set.empty
  3819     UNIV ~>     Set.UNIV
  3820 
  3821 * Class complete_lattice with operations Inf, Sup, INFI, SUPR now in
  3822 theory Set.
  3823 
  3824 * Auxiliary class "itself" has disappeared -- classes without any
  3825 parameter are treated as expected by the 'class' command.
  3826 
  3827 * Leibnitz's Series for Pi and the arcus tangens and logarithm series.
  3828 
  3829 * Common decision procedures (Cooper, MIR, Ferrack, Approximation,
  3830 Dense_Linear_Order) are now in directory HOL/Decision_Procs.
  3831 
  3832 * Theory src/HOL/Decision_Procs/Approximation provides the new proof
  3833 method "approximation".  It proves formulas on real values by using
  3834 interval arithmetic.  In the formulas are also the transcendental
  3835 functions sin, cos, tan, atan, ln, exp and the constant pi are
  3836 allowed. For examples see
  3837 src/HOL/Descision_Procs/ex/Approximation_Ex.thy.
  3838 
  3839 * Theory "Reflection" now resides in HOL/Library.
  3840 
  3841 * Entry point to Word library now simply named "Word".
  3842 INCOMPATIBILITY.
  3843 
  3844 * Made source layout more coherent with logical distribution
  3845 structure:
  3846 
  3847     src/HOL/Library/RType.thy ~> src/HOL/Typerep.thy
  3848     src/HOL/Library/Code_Message.thy ~> src/HOL/
  3849     src/HOL/Library/GCD.thy ~> src/HOL/
  3850     src/HOL/Library/Order_Relation.thy ~> src/HOL/
  3851     src/HOL/Library/Parity.thy ~> src/HOL/
  3852     src/HOL/Library/Univ_Poly.thy ~> src/HOL/
  3853     src/HOL/Real/ContNotDenum.thy ~> src/HOL/Library/
  3854     src/HOL/Real/Lubs.thy ~> src/HOL/
  3855     src/HOL/Real/PReal.thy ~> src/HOL/
  3856     src/HOL/Real/Rational.thy ~> src/HOL/
  3857     src/HOL/Real/RComplete.thy ~> src/HOL/
  3858     src/HOL/Real/RealDef.thy ~> src/HOL/
  3859     src/HOL/Real/RealPow.thy ~> src/HOL/
  3860     src/HOL/Real/Real.thy ~> src/HOL/
  3861     src/HOL/Complex/Complex_Main.thy ~> src/HOL/
  3862     src/HOL/Complex/Complex.thy ~> src/HOL/
  3863     src/HOL/Complex/FrechetDeriv.thy ~> src/HOL/Library/
  3864     src/HOL/Complex/Fundamental_Theorem_Algebra.thy ~> src/HOL/Library/
  3865     src/HOL/Hyperreal/Deriv.thy ~> src/HOL/
  3866     src/HOL/Hyperreal/Fact.thy ~> src/HOL/
  3867     src/HOL/Hyperreal/Integration.thy ~> src/HOL/
  3868     src/HOL/Hyperreal/Lim.thy ~> src/HOL/
  3869     src/HOL/Hyperreal/Ln.thy ~> src/HOL/
  3870     src/HOL/Hyperreal/Log.thy ~> src/HOL/
  3871     src/HOL/Hyperreal/MacLaurin.thy ~> src/HOL/
  3872     src/HOL/Hyperreal/NthRoot.thy ~> src/HOL/
  3873     src/HOL/Hyperreal/Series.thy ~> src/HOL/
  3874     src/HOL/Hyperreal/SEQ.thy ~> src/HOL/
  3875     src/HOL/Hyperreal/Taylor.thy ~> src/HOL/
  3876     src/HOL/Hyperreal/Transcendental.thy ~> src/HOL/
  3877     src/HOL/Real/Float ~> src/HOL/Library/
  3878     src/HOL/Real/HahnBanach ~> src/HOL/HahnBanach
  3879     src/HOL/Real/RealVector.thy ~> src/HOL/
  3880 
  3881     src/HOL/arith_data.ML ~> src/HOL/Tools
  3882     src/HOL/hologic.ML ~> src/HOL/Tools
  3883     src/HOL/simpdata.ML ~> src/HOL/Tools
  3884     src/HOL/int_arith1.ML ~> src/HOL/Tools/int_arith.ML
  3885     src/HOL/int_factor_simprocs.ML ~> src/HOL/Tools
  3886     src/HOL/nat_simprocs.ML ~> src/HOL/Tools
  3887     src/HOL/Real/float_arith.ML ~> src/HOL/Tools
  3888     src/HOL/Real/float_syntax.ML ~> src/HOL/Tools
  3889     src/HOL/Real/rat_arith.ML ~> src/HOL/Tools
  3890     src/HOL/Real/real_arith.ML ~> src/HOL/Tools
  3891 
  3892     src/HOL/Library/Array.thy ~> src/HOL/Imperative_HOL
  3893     src/HOL/Library/Heap_Monad.thy ~> src/HOL/Imperative_HOL
  3894     src/HOL/Library/Heap.thy ~> src/HOL/Imperative_HOL
  3895     src/HOL/Library/Imperative_HOL.thy ~> src/HOL/Imperative_HOL
  3896     src/HOL/Library/Ref.thy ~> src/HOL/Imperative_HOL
  3897     src/HOL/Library/Relational.thy ~> src/HOL/Imperative_HOL
  3898 
  3899 * If methods "eval" and "evaluation" encounter a structured proof
  3900 state with !!/==>, only the conclusion is evaluated to True (if
  3901 possible), avoiding strange error messages.
  3902 
  3903 * Method "sizechange" automates termination proofs using (a
  3904 modification of) the size-change principle.  Requires SAT solver.  See
  3905 src/HOL/ex/Termination.thy for examples.
  3906 
  3907 * Simplifier: simproc for let expressions now unfolds if bound
  3908 variable occurs at most once in let expression body.  INCOMPATIBILITY.
  3909 
  3910 * Method "arith": Linear arithmetic now ignores all inequalities when
  3911 fast_arith_neq_limit is exceeded, instead of giving up entirely.
  3912 
  3913 * New attribute "arith" for facts that should always be used
  3914 automatically by arithmetic. It is intended to be used locally in
  3915 proofs, e.g.
  3916 
  3917   assumes [arith]: "x > 0"
  3918 
  3919 Global usage is discouraged because of possible performance impact.
  3920 
  3921 * New classes "top" and "bot" with corresponding operations "top" and
  3922 "bot" in theory Orderings; instantiation of class "complete_lattice"
  3923 requires instantiation of classes "top" and "bot".  INCOMPATIBILITY.
  3924 
  3925 * Changed definition lemma "less_fun_def" in order to provide an
  3926 instance for preorders on functions; use lemma "less_le" instead.
  3927 INCOMPATIBILITY.
  3928 
  3929 * Theory Orderings: class "wellorder" moved here, with explicit
  3930 induction rule "less_induct" as assumption.  For instantiation of
  3931 "wellorder" by means of predicate "wf", use rule wf_wellorderI.
  3932 INCOMPATIBILITY.
  3933 
  3934 * Theory Orderings: added class "preorder" as superclass of "order".
  3935 INCOMPATIBILITY: Instantiation proofs for order, linorder
  3936 etc. slightly changed.  Some theorems named order_class.* now named
  3937 preorder_class.*.
  3938 
  3939 * Theory Relation: renamed "refl" to "refl_on", "reflexive" to "refl,
  3940 "diag" to "Id_on".
  3941 
  3942 * Theory Finite_Set: added a new fold combinator of type
  3943 
  3944   ('a => 'b => 'b) => 'b => 'a set => 'b
  3945 
  3946 Occasionally this is more convenient than the old fold combinator
  3947 which is now defined in terms of the new one and renamed to
  3948 fold_image.
  3949 
  3950 * Theories Ring_and_Field and OrderedGroup: The lemmas "group_simps"
  3951 and "ring_simps" have been replaced by "algebra_simps" (which can be
  3952 extended with further lemmas!).  At the moment both still exist but
  3953 the former will disappear at some point.
  3954 
  3955 * Theory Power: Lemma power_Suc is now declared as a simp rule in
  3956 class recpower.  Type-specific simp rules for various recpower types
  3957 have been removed.  INCOMPATIBILITY, rename old lemmas as follows:
  3958 
  3959 rat_power_0    -> power_0
  3960 rat_power_Suc  -> power_Suc
  3961 realpow_0      -> power_0
  3962 realpow_Suc    -> power_Suc
  3963 complexpow_0   -> power_0
  3964 complexpow_Suc -> power_Suc
  3965 power_poly_0   -> power_0
  3966 power_poly_Suc -> power_Suc
  3967 
  3968 * Theories Ring_and_Field and Divides: Definition of "op dvd" has been
  3969 moved to separate class dvd in Ring_and_Field; a couple of lemmas on
  3970 dvd has been generalized to class comm_semiring_1.  Likewise a bunch
  3971 of lemmas from Divides has been generalized from nat to class
  3972 semiring_div.  INCOMPATIBILITY.  This involves the following theorem
  3973 renames resulting from duplicate elimination:
  3974 
  3975     dvd_def_mod ~>          dvd_eq_mod_eq_0
  3976     zero_dvd_iff ~>         dvd_0_left_iff
  3977     dvd_0 ~>                dvd_0_right
  3978     DIVISION_BY_ZERO_DIV ~> div_by_0
  3979     DIVISION_BY_ZERO_MOD ~> mod_by_0
  3980     mult_div ~>             div_mult_self2_is_id
  3981     mult_mod ~>             mod_mult_self2_is_0
  3982 
  3983 * Theory IntDiv: removed many lemmas that are instances of class-based
  3984 generalizations (from Divides and Ring_and_Field).  INCOMPATIBILITY,
  3985 rename old lemmas as follows:
  3986 
  3987 dvd_diff               -> nat_dvd_diff
  3988 dvd_zminus_iff         -> dvd_minus_iff
  3989 mod_add1_eq            -> mod_add_eq
  3990 mod_mult1_eq           -> mod_mult_right_eq
  3991 mod_mult1_eq'          -> mod_mult_left_eq
  3992 mod_mult_distrib_mod   -> mod_mult_eq
  3993 nat_mod_add_left_eq    -> mod_add_left_eq
  3994 nat_mod_add_right_eq   -> mod_add_right_eq
  3995 nat_mod_div_trivial    -> mod_div_trivial
  3996 nat_mod_mod_trivial    -> mod_mod_trivial
  3997 zdiv_zadd_self1        -> div_add_self1
  3998 zdiv_zadd_self2        -> div_add_self2
  3999 zdiv_zmult_self1       -> div_mult_self2_is_id
  4000 zdiv_zmult_self2       -> div_mult_self1_is_id
  4001 zdvd_triv_left         -> dvd_triv_left
  4002 zdvd_triv_right        -> dvd_triv_right
  4003 zdvd_zmult_cancel_disj -> dvd_mult_cancel_left
  4004 zmod_eq0_zdvd_iff      -> dvd_eq_mod_eq_0[symmetric]
  4005 zmod_zadd_left_eq      -> mod_add_left_eq
  4006 zmod_zadd_right_eq     -> mod_add_right_eq
  4007 zmod_zadd_self1        -> mod_add_self1
  4008 zmod_zadd_self2        -> mod_add_self2
  4009 zmod_zadd1_eq          -> mod_add_eq
  4010 zmod_zdiff1_eq         -> mod_diff_eq
  4011 zmod_zdvd_zmod         -> mod_mod_cancel
  4012 zmod_zmod_cancel       -> mod_mod_cancel
  4013 zmod_zmult_self1       -> mod_mult_self2_is_0
  4014 zmod_zmult_self2       -> mod_mult_self1_is_0
  4015 zmod_1                 -> mod_by_1
  4016 zdiv_1                 -> div_by_1
  4017 zdvd_abs1              -> abs_dvd_iff
  4018 zdvd_abs2              -> dvd_abs_iff
  4019 zdvd_refl              -> dvd_refl
  4020 zdvd_trans             -> dvd_trans
  4021 zdvd_zadd              -> dvd_add
  4022 zdvd_zdiff             -> dvd_diff
  4023 zdvd_zminus_iff        -> dvd_minus_iff
  4024 zdvd_zminus2_iff       -> minus_dvd_iff
  4025 zdvd_zmultD            -> dvd_mult_right
  4026 zdvd_zmultD2           -> dvd_mult_left
  4027 zdvd_zmult_mono        -> mult_dvd_mono
  4028 zdvd_0_right           -> dvd_0_right
  4029 zdvd_0_left            -> dvd_0_left_iff
  4030 zdvd_1_left            -> one_dvd
  4031 zminus_dvd_iff         -> minus_dvd_iff
  4032 
  4033 * Theory Rational: 'Fract k 0' now equals '0'.  INCOMPATIBILITY.
  4034 
  4035 * The real numbers offer decimal input syntax: 12.34 is translated
  4036 into 1234/10^2. This translation is not reversed upon output.
  4037 
  4038 * Theory Library/Polynomial defines an abstract type 'a poly of
  4039 univariate polynomials with coefficients of type 'a.  In addition to
  4040 the standard ring operations, it also supports div and mod.  Code
  4041 generation is also supported, using list-style constructors.
  4042 
  4043 * Theory Library/Inner_Product defines a class of real_inner for real
  4044 inner product spaces, with an overloaded operation inner :: 'a => 'a
  4045 => real.  Class real_inner is a subclass of real_normed_vector from
  4046 theory RealVector.
  4047 
  4048 * Theory Library/Product_Vector provides instances for the product
  4049 type 'a * 'b of several classes from RealVector and Inner_Product.
  4050 Definitions of addition, subtraction, scalar multiplication, norms,
  4051 and inner products are included.
  4052 
  4053 * Theory Library/Bit defines the field "bit" of integers modulo 2.  In
  4054 addition to the field operations, numerals and case syntax are also
  4055 supported.
  4056 
  4057 * Theory Library/Diagonalize provides constructive version of Cantor's
  4058 first diagonalization argument.
  4059 
  4060 * Theory Library/GCD: Curried operations gcd, lcm (for nat) and zgcd,
  4061 zlcm (for int); carried together from various gcd/lcm developements in
  4062 the HOL Distribution.  Constants zgcd and zlcm replace former igcd and
  4063 ilcm; corresponding theorems renamed accordingly.  INCOMPATIBILITY,
  4064 may recover tupled syntax as follows:
  4065 
  4066     hide (open) const gcd
  4067     abbreviation gcd where
  4068       "gcd == (%(a, b). GCD.gcd a b)"
  4069     notation (output)
  4070       GCD.gcd ("gcd '(_, _')")
  4071 
  4072 The same works for lcm, zgcd, zlcm.
  4073 
  4074 * Theory Library/Nat_Infinity: added addition, numeral syntax and more
  4075 instantiations for algebraic structures.  Removed some duplicate
  4076 theorems.  Changes in simp rules.  INCOMPATIBILITY.
  4077 
  4078 * ML antiquotation @{code} takes a constant as argument and generates
  4079 corresponding code in background and inserts name of the corresponding
  4080 resulting ML value/function/datatype constructor binding in place.
  4081 All occurrences of @{code} with a single ML block are generated
  4082 simultaneously.  Provides a generic and safe interface for
  4083 instrumentalizing code generation.  See
  4084 src/HOL/Decision_Procs/Ferrack.thy for a more ambitious application.
  4085 In future you ought to refrain from ad-hoc compiling generated SML
  4086 code on the ML toplevel.  Note that (for technical reasons) @{code}
  4087 cannot refer to constants for which user-defined serializations are
  4088 set.  Refer to the corresponding ML counterpart directly in that
  4089 cases.
  4090 
  4091 * Command 'rep_datatype': instead of theorem names the command now
  4092 takes a list of terms denoting the constructors of the type to be
  4093 represented as datatype.  The characteristic theorems have to be
  4094 proven.  INCOMPATIBILITY.  Also observe that the following theorems
  4095 have disappeared in favour of existing ones:
  4096 
  4097     unit_induct                 ~> unit.induct
  4098     prod_induct                 ~> prod.induct
  4099     sum_induct                  ~> sum.induct
  4100     Suc_Suc_eq                  ~> nat.inject
  4101     Suc_not_Zero Zero_not_Suc   ~> nat.distinct
  4102 
  4103 
  4104 *** HOL-Algebra ***
  4105 
  4106 * New locales for orders and lattices where the equivalence relation
  4107 is not restricted to equality.  INCOMPATIBILITY: all order and lattice
  4108 locales use a record structure with field eq for the equivalence.
  4109 
  4110 * New theory of factorial domains.
  4111 
  4112 * Units_l_inv and Units_r_inv are now simp rules by default.
  4113 INCOMPATIBILITY.  Simplifier proof that require deletion of l_inv
  4114 and/or r_inv will now also require deletion of these lemmas.
  4115 
  4116 * Renamed the following theorems, INCOMPATIBILITY:
  4117 
  4118 UpperD ~> Upper_memD
  4119 LowerD ~> Lower_memD
  4120 least_carrier ~> least_closed
  4121 greatest_carrier ~> greatest_closed
  4122 greatest_Lower_above ~> greatest_Lower_below
  4123 one_zero ~> carrier_one_zero
  4124 one_not_zero ~> carrier_one_not_zero  (collision with assumption)
  4125 
  4126 
  4127 *** HOL-Nominal ***
  4128 
  4129 * Nominal datatypes can now contain type-variables.
  4130 
  4131 * Commands 'nominal_inductive' and 'equivariance' work with local
  4132 theory targets.
  4133 
  4134 * Nominal primrec can now works with local theory targets and its
  4135 specification syntax now conforms to the general format as seen in
  4136 'inductive' etc.
  4137 
  4138 * Method "perm_simp" honours the standard simplifier attributes
  4139 (no_asm), (no_asm_use) etc.
  4140 
  4141 * The new predicate #* is defined like freshness, except that on the
  4142 left hand side can be a set or list of atoms.
  4143 
  4144 * Experimental command 'nominal_inductive2' derives strong induction
  4145 principles for inductive definitions.  In contrast to
  4146 'nominal_inductive', which can only deal with a fixed number of
  4147 binders, it can deal with arbitrary expressions standing for sets of
  4148 atoms to be avoided.  The only inductive definition we have at the
  4149 moment that needs this generalisation is the typing rule for Lets in
  4150 the algorithm W:
  4151 
  4152  Gamma |- t1 : T1   (x,close Gamma T1)::Gamma |- t2 : T2   x#Gamma
  4153  -----------------------------------------------------------------
  4154          Gamma |- Let x be t1 in t2 : T2
  4155 
  4156 In this rule one wants to avoid all the binders that are introduced by
  4157 "close Gamma T1".  We are looking for other examples where this
  4158 feature might be useful.  Please let us know.
  4159 
  4160 
  4161 *** HOLCF ***
  4162 
  4163 * Reimplemented the simplification procedure for proving continuity
  4164 subgoals.  The new simproc is extensible; users can declare additional
  4165 continuity introduction rules with the attribute [cont2cont].
  4166 
  4167 * The continuity simproc now uses a different introduction rule for
  4168 solving continuity subgoals on terms with lambda abstractions.  In
  4169 some rare cases the new simproc may fail to solve subgoals that the
  4170 old one could solve, and "simp add: cont2cont_LAM" may be necessary.
  4171 Potential INCOMPATIBILITY.
  4172 
  4173 * Command 'fixrec': specification syntax now conforms to the general
  4174 format as seen in 'inductive' etc.  See src/HOLCF/ex/Fixrec_ex.thy for
  4175 examples.  INCOMPATIBILITY.
  4176 
  4177 
  4178 *** ZF ***
  4179 
  4180 * Proof of Zorn's Lemma for partial orders.
  4181 
  4182 
  4183 *** ML ***
  4184 
  4185 * Multithreading for Poly/ML 5.1/5.2 is no longer supported, only for
  4186 Poly/ML 5.2.1 or later.  Important note: the TimeLimit facility
  4187 depends on multithreading, so timouts will not work before Poly/ML
  4188 5.2.1!
  4189 
  4190 * High-level support for concurrent ML programming, see
  4191 src/Pure/Cuncurrent.  The data-oriented model of "future values" is
  4192 particularly convenient to organize independent functional
  4193 computations.  The concept of "synchronized variables" provides a
  4194 higher-order interface for components with shared state, avoiding the
  4195 delicate details of mutexes and condition variables.  (Requires
  4196 Poly/ML 5.2.1 or later.)
  4197 
  4198 * ML bindings produced via Isar commands are stored within the Isar
  4199 context (theory or proof).  Consequently, commands like 'use' and 'ML'
  4200 become thread-safe and work with undo as expected (concerning
  4201 top-level bindings, not side-effects on global references).
  4202 INCOMPATIBILITY, need to provide proper Isar context when invoking the
  4203 compiler at runtime; really global bindings need to be given outside a
  4204 theory.  (Requires Poly/ML 5.2 or later.)
  4205 
  4206 * Command 'ML_prf' is analogous to 'ML' but works within a proof
  4207 context.  Top-level ML bindings are stored within the proof context in
  4208 a purely sequential fashion, disregarding the nested proof structure.
  4209 ML bindings introduced by 'ML_prf' are discarded at the end of the
  4210 proof.  (Requires Poly/ML 5.2 or later.)
  4211 
  4212 * Simplified ML attribute and method setup, cf. functions Attrib.setup
  4213 and Method.setup, as well as Isar commands 'attribute_setup' and
  4214 'method_setup'.  INCOMPATIBILITY for 'method_setup', need to simplify
  4215 existing code accordingly, or use plain 'setup' together with old
  4216 Method.add_method.
  4217 
  4218 * Simplified ML oracle interface Thm.add_oracle promotes 'a -> cterm
  4219 to 'a -> thm, while results are always tagged with an authentic oracle
  4220 name.  The Isar command 'oracle' is now polymorphic, no argument type
  4221 is specified.  INCOMPATIBILITY, need to simplify existing oracle code
  4222 accordingly.  Note that extra performance may be gained by producing
  4223 the cterm carefully, avoiding slow Thm.cterm_of.
  4224 
  4225 * Simplified interface for defining document antiquotations via
  4226 ThyOutput.antiquotation, ThyOutput.output, and optionally
  4227 ThyOutput.maybe_pretty_source.  INCOMPATIBILITY, need to simplify user
  4228 antiquotations accordingly, see src/Pure/Thy/thy_output.ML for common
  4229 examples.
  4230 
  4231 * More systematic treatment of long names, abstract name bindings, and
  4232 name space operations.  Basic operations on qualified names have been
  4233 move from structure NameSpace to Long_Name, e.g. Long_Name.base_name,
  4234 Long_Name.append.  Old type bstring has been mostly replaced by
  4235 abstract type binding (see structure Binding), which supports precise
  4236 qualification by packages and local theory targets, as well as proper
  4237 tracking of source positions.  INCOMPATIBILITY, need to wrap old
  4238 bstring values into Binding.name, or better pass through abstract
  4239 bindings everywhere.  See further src/Pure/General/long_name.ML,
  4240 src/Pure/General/binding.ML and src/Pure/General/name_space.ML
  4241 
  4242 * Result facts (from PureThy.note_thms, ProofContext.note_thms,
  4243 LocalTheory.note etc.) now refer to the *full* internal name, not the
  4244 bstring as before.  INCOMPATIBILITY, not detected by ML type-checking!
  4245 
  4246 * Disposed old type and term read functions (Sign.read_def_typ,
  4247 Sign.read_typ, Sign.read_def_terms, Sign.read_term,
  4248 Thm.read_def_cterms, Thm.read_cterm etc.).  INCOMPATIBILITY, should
  4249 use regular Syntax.read_typ, Syntax.read_term, Syntax.read_typ_global,
  4250 Syntax.read_term_global etc.; see also OldGoals.read_term as last
  4251 resort for legacy applications.
  4252 
  4253 * Disposed old declarations, tactics, tactic combinators that refer to
  4254 the simpset or claset of an implicit theory (such as Addsimps,
  4255 Simp_tac, SIMPSET).  INCOMPATIBILITY, should use @{simpset} etc. in
  4256 embedded ML text, or local_simpset_of with a proper context passed as
  4257 explicit runtime argument.
  4258 
  4259 * Rules and tactics that read instantiations (read_instantiate,
  4260 res_inst_tac, thin_tac, subgoal_tac etc.) now demand a proper proof
  4261 context, which is required for parsing and type-checking.  Moreover,
  4262 the variables are specified as plain indexnames, not string encodings
  4263 thereof.  INCOMPATIBILITY.
  4264 
  4265 * Generic Toplevel.add_hook interface allows to analyze the result of
  4266 transactions.  E.g. see src/Pure/ProofGeneral/proof_general_pgip.ML
  4267 for theorem dependency output of transactions resulting in a new
  4268 theory state.
  4269 
  4270 * ML antiquotations: block-structured compilation context indicated by
  4271 \<lbrace> ... \<rbrace>; additional antiquotation forms:
  4272 
  4273   @{binding name}                         - basic name binding
  4274   @{let ?pat = term}                      - term abbreviation (HO matching)
  4275   @{note name = fact}                     - fact abbreviation
  4276   @{thm fact}                             - singleton fact (with attributes)
  4277   @{thms fact}                            - general fact (with attributes)
  4278   @{lemma prop by method}                 - singleton goal
  4279   @{lemma prop by meth1 meth2}            - singleton goal
  4280   @{lemma prop1 ... propN by method}      - general goal
  4281   @{lemma prop1 ... propN by meth1 meth2} - general goal
  4282   @{lemma (open) ...}                     - open derivation
  4283 
  4284 
  4285 *** System ***
  4286 
  4287 * The Isabelle "emacs" tool provides a specific interface to invoke
  4288 Proof General / Emacs, with more explicit failure if that is not
  4289 installed (the old isabelle-interface script silently falls back on
  4290 isabelle-process).  The PROOFGENERAL_HOME setting determines the
  4291 installation location of the Proof General distribution.
  4292 
  4293 * Isabelle/lib/classes/Pure.jar provides basic support to integrate
  4294 the Isabelle process into a JVM/Scala application.  See
  4295 Isabelle/lib/jedit/plugin for a minimal example.  (The obsolete Java
  4296 process wrapper has been discontinued.)
  4297 
  4298 * Added homegrown Isabelle font with unicode layout, see lib/fonts.
  4299 
  4300 * Various status messages (with exact source position information) are
  4301 emitted, if proper markup print mode is enabled.  This allows
  4302 user-interface components to provide detailed feedback on internal
  4303 prover operations.
  4304 
  4305 
  4306 
  4307 New in Isabelle2008 (June 2008)
  4308 -------------------------------
  4309 
  4310 *** General ***
  4311 
  4312 * The Isabelle/Isar Reference Manual (isar-ref) has been reorganized
  4313 and updated, with formally checked references as hyperlinks.
  4314 
  4315 * Theory loader: use_thy (and similar operations) no longer set the
  4316 implicit ML context, which was occasionally hard to predict and in
  4317 conflict with concurrency.  INCOMPATIBILITY, use ML within Isar which
  4318 provides a proper context already.
  4319 
  4320 * Theory loader: old-style ML proof scripts being *attached* to a thy
  4321 file are no longer supported.  INCOMPATIBILITY, regular 'uses' and
  4322 'use' within a theory file will do the job.
  4323 
  4324 * Name space merge now observes canonical order, i.e. the second space
  4325 is inserted into the first one, while existing entries in the first
  4326 space take precedence.  INCOMPATIBILITY in rare situations, may try to
  4327 swap theory imports.
  4328 
  4329 * Syntax: symbol \<chi> is now considered a letter.  Potential
  4330 INCOMPATIBILITY in identifier syntax etc.
  4331 
  4332 * Outer syntax: string tokens no longer admit escaped white space,
  4333 which was an accidental (undocumented) feature.  INCOMPATIBILITY, use
  4334 white space without escapes.
  4335 
  4336 * Outer syntax: string tokens may contain arbitrary character codes
  4337 specified via 3 decimal digits (as in SML).  E.g. "foo\095bar" for
  4338 "foo_bar".
  4339 
  4340 
  4341 *** Pure ***
  4342 
  4343 * Context-dependent token translations.  Default setup reverts locally
  4344 fixed variables, and adds hilite markup for undeclared frees.
  4345 
  4346 * Unused theorems can be found using the new command 'unused_thms'.
  4347 There are three ways of invoking it:
  4348 
  4349 (1) unused_thms
  4350      Only finds unused theorems in the current theory.
  4351 
  4352 (2) unused_thms thy_1 ... thy_n -
  4353      Finds unused theorems in the current theory and all of its ancestors,
  4354      excluding the theories thy_1 ... thy_n and all of their ancestors.
  4355 
  4356 (3) unused_thms thy_1 ... thy_n - thy'_1 ... thy'_m
  4357      Finds unused theorems in the theories thy'_1 ... thy'_m and all of
  4358      their ancestors, excluding the theories thy_1 ... thy_n and all of
  4359      their ancestors.
  4360 
  4361 In order to increase the readability of the list produced by
  4362 unused_thms, theorems that have been created by a particular instance
  4363 of a theory command such as 'inductive' or 'function' are considered
  4364 to belong to the same "group", meaning that if at least one theorem in
  4365 this group is used, the other theorems in the same group are no longer
  4366 reported as unused.  Moreover, if all theorems in the group are
  4367 unused, only one theorem in the group is displayed.
  4368 
  4369 Note that proof objects have to be switched on in order for
  4370 unused_thms to work properly (i.e. !proofs must be >= 1, which is
  4371 usually the case when using Proof General with the default settings).
  4372 
  4373 * Authentic naming of facts disallows ad-hoc overwriting of previous
  4374 theorems within the same name space.  INCOMPATIBILITY, need to remove
  4375 duplicate fact bindings, or even accidental fact duplications.  Note
  4376 that tools may maintain dynamically scoped facts systematically, using
  4377 PureThy.add_thms_dynamic.
  4378 
  4379 * Command 'hide' now allows to hide from "fact" name space as well.
  4380 
  4381 * Eliminated destructive theorem database, simpset, claset, and
  4382 clasimpset.  Potential INCOMPATIBILITY, really need to observe linear
  4383 update of theories within ML code.
  4384 
  4385 * Eliminated theory ProtoPure and CPure, leaving just one Pure theory.
  4386 INCOMPATIBILITY, object-logics depending on former Pure require
  4387 additional setup PureThy.old_appl_syntax_setup; object-logics
  4388 depending on former CPure need to refer to Pure.
  4389 
  4390 * Commands 'use' and 'ML' are now purely functional, operating on
  4391 theory/local_theory.  Removed former 'ML_setup' (on theory), use 'ML'
  4392 instead.  Added 'ML_val' as mere diagnostic replacement for 'ML'.
  4393 INCOMPATIBILITY.
  4394 
  4395 * Command 'setup': discontinued implicit version with ML reference.
  4396 
  4397 * Instantiation target allows for simultaneous specification of class
  4398 instance operations together with an instantiation proof.
  4399 Type-checking phase allows to refer to class operations uniformly.
  4400 See src/HOL/Complex/Complex.thy for an Isar example and
  4401 src/HOL/Library/Eval.thy for an ML example.
  4402 
  4403 * Indexing of literal facts: be more serious about including only
  4404 facts from the visible specification/proof context, but not the
  4405 background context (locale etc.).  Affects `prop` notation and method
  4406 "fact".  INCOMPATIBILITY: need to name facts explicitly in rare
  4407 situations.
  4408 
  4409 * Method "cases", "induct", "coinduct": removed obsolete/undocumented
  4410 "(open)" option, which used to expose internal bound variables to the
  4411 proof text.
  4412 
  4413 * Isar statements: removed obsolete case "rule_context".
  4414 INCOMPATIBILITY, better use explicit fixes/assumes.
  4415 
  4416 * Locale proofs: default proof step now includes 'unfold_locales';
  4417 hence 'proof' without argument may be used to unfold locale
  4418 predicates.
  4419 
  4420 
  4421 *** Document preparation ***
  4422 
  4423 * Simplified pdfsetup.sty: color/hyperref is used unconditionally for
  4424 both pdf and dvi (hyperlinks usually work in xdvi as well); removed
  4425 obsolete thumbpdf setup (contemporary PDF viewers do this on the
  4426 spot); renamed link color from "darkblue" to "linkcolor" (default
  4427 value unchanged, can be redefined via \definecolor); no longer sets
  4428 "a4paper" option (unnecessary or even intrusive).
  4429 
  4430 * Antiquotation @{lemma A method} proves proposition A by the given
  4431 method (either a method name or a method name plus (optional) method
  4432 arguments in parentheses) and prints A just like @{prop A}.
  4433 
  4434 
  4435 *** HOL ***
  4436 
  4437 * New primrec package.  Specification syntax conforms in style to
  4438 definition/function/....  No separate induction rule is provided.  The
  4439 "primrec" command distinguishes old-style and new-style specifications
  4440 by syntax.  The former primrec package is now named OldPrimrecPackage.
  4441 When adjusting theories, beware: constants stemming from new-style
  4442 primrec specifications have authentic syntax.
  4443 
  4444 * Metis prover is now an order of magnitude faster, and also works
  4445 with multithreading.
  4446 
  4447 * Metis: the maximum number of clauses that can be produced from a
  4448 theorem is now given by the attribute max_clauses.  Theorems that
  4449 exceed this number are ignored, with a warning printed.
  4450 
  4451 * Sledgehammer no longer produces structured proofs by default. To
  4452 enable, declare [[sledgehammer_full = true]].  Attributes
  4453 reconstruction_modulus, reconstruction_sorts renamed
  4454 sledgehammer_modulus, sledgehammer_sorts.  INCOMPATIBILITY.
  4455 
  4456 * Method "induct_scheme" derives user-specified induction rules
  4457 from well-founded induction and completeness of patterns. This factors
  4458 out some operations that are done internally by the function package
  4459 and makes them available separately.  See
  4460 src/HOL/ex/Induction_Scheme.thy for examples.
  4461 
  4462 * More flexible generation of measure functions for termination
  4463 proofs: Measure functions can be declared by proving a rule of the
  4464 form "is_measure f" and giving it the [measure_function] attribute.
  4465 The "is_measure" predicate is logically meaningless (always true), and
  4466 just guides the heuristic.  To find suitable measure functions, the
  4467 termination prover sets up the goal "is_measure ?f" of the appropriate
  4468 type and generates all solutions by prolog-style backwards proof using
  4469 the declared rules.
  4470 
  4471 This setup also deals with rules like 
  4472 
  4473   "is_measure f ==> is_measure (list_size f)"
  4474 
  4475 which accommodates nested datatypes that recurse through lists.
  4476 Similar rules are predeclared for products and option types.
  4477 
  4478 * Turned the type of sets "'a set" into an abbreviation for "'a => bool"
  4479 
  4480   INCOMPATIBILITIES:
  4481 
  4482   - Definitions of overloaded constants on sets have to be replaced by
  4483     definitions on => and bool.
  4484 
  4485   - Some definitions of overloaded operators on sets can now be proved
  4486     using the definitions of the operators on => and bool.  Therefore,
  4487     the following theorems have been renamed:
  4488 
  4489       subset_def   -> subset_eq
  4490       psubset_def  -> psubset_eq
  4491       set_diff_def -> set_diff_eq
  4492       Compl_def    -> Compl_eq
  4493       Sup_set_def  -> Sup_set_eq
  4494       Inf_set_def  -> Inf_set_eq
  4495       sup_set_def  -> sup_set_eq
  4496       inf_set_def  -> inf_set_eq
  4497 
  4498   - Due to the incompleteness of the HO unification algorithm, some
  4499     rules such as subst may require manual instantiation, if some of
  4500     the unknowns in the rule is a set.
  4501 
  4502   - Higher order unification and forward proofs:
  4503     The proof pattern
  4504 
  4505       have "P (S::'a set)" <...>
  4506       then have "EX S. P S" ..
  4507 
  4508     no longer works (due to the incompleteness of the HO unification
  4509     algorithm) and must be replaced by the pattern
  4510 
  4511       have "EX S. P S"
  4512       proof
  4513         show "P S" <...>
  4514       qed
  4515 
  4516   - Calculational reasoning with subst (or similar rules):
  4517     The proof pattern
  4518 
  4519       have "P (S::'a set)" <...>
  4520       also have "S = T" <...>
  4521       finally have "P T" .
  4522 
  4523     no longer works (for similar reasons as the previous example) and
  4524     must be replaced by something like
  4525 
  4526       have "P (S::'a set)" <...>
  4527       moreover have "S = T" <...>
  4528       ultimately have "P T" by simp
  4529 
  4530   - Tactics or packages written in ML code:
  4531     Code performing pattern matching on types via
  4532 
  4533       Type ("set", [T]) => ...
  4534 
  4535     must be rewritten. Moreover, functions like strip_type or
  4536     binder_types no longer return the right value when applied to a
  4537     type of the form
  4538 
  4539       T1 => ... => Tn => U => bool
  4540 
  4541     rather than
  4542 
  4543       T1 => ... => Tn => U set
  4544 
  4545 * Merged theories Wellfounded_Recursion, Accessible_Part and
  4546 Wellfounded_Relations to theory Wellfounded.
  4547 
  4548 * Explicit class "eq" for executable equality.  INCOMPATIBILITY.
  4549 
  4550 * Class finite no longer treats UNIV as class parameter.  Use class
  4551 enum from theory Library/Enum instead to achieve a similar effect.
  4552 INCOMPATIBILITY.
  4553 
  4554 * Theory List: rule list_induct2 now has explicitly named cases "Nil"
  4555 and "Cons".  INCOMPATIBILITY.
  4556 
  4557 * HOL (and FOL): renamed variables in rules imp_elim and swap.
  4558 Potential INCOMPATIBILITY.
  4559 
  4560 * Theory Product_Type: duplicated lemmas split_Pair_apply and
  4561 injective_fst_snd removed, use split_eta and prod_eqI instead.
  4562 Renamed upd_fst to apfst and upd_snd to apsnd.  INCOMPATIBILITY.
  4563 
  4564 * Theory Nat: removed redundant lemmas that merely duplicate lemmas of
  4565 the same name in theory Orderings:
  4566 
  4567   less_trans
  4568   less_linear
  4569   le_imp_less_or_eq
  4570   le_less_trans
  4571   less_le_trans
  4572   less_not_sym
  4573   less_asym
  4574 
  4575 Renamed less_imp_le to less_imp_le_nat, and less_irrefl to
  4576 less_irrefl_nat.  Potential INCOMPATIBILITY due to more general types
  4577 and different variable names.
  4578 
  4579 * Library/Option_ord.thy: Canonical order on option type.
  4580 
  4581 * Library/RBT.thy: Red-black trees, an efficient implementation of
  4582 finite maps.
  4583 
  4584 * Library/Countable.thy: Type class for countable types.
  4585 
  4586 * Theory Int: The representation of numerals has changed.  The infix
  4587 operator BIT and the bit datatype with constructors B0 and B1 have
  4588 disappeared.  INCOMPATIBILITY, use "Int.Bit0 x" and "Int.Bit1 y" in
  4589 place of "x BIT bit.B0" and "y BIT bit.B1", respectively.  Theorems
  4590 involving BIT, B0, or B1 have been renamed with "Bit0" or "Bit1"
  4591 accordingly.
  4592 
  4593 * Theory Nat: definition of <= and < on natural numbers no longer
  4594 depend on well-founded relations.  INCOMPATIBILITY.  Definitions
  4595 le_def and less_def have disappeared.  Consider lemmas not_less
  4596 [symmetric, where ?'a = nat] and less_eq [symmetric] instead.
  4597 
  4598 * Theory Finite_Set: locales ACf, ACe, ACIf, ACIfSL and ACIfSLlin
  4599 (whose purpose mainly is for various fold_set functionals) have been
  4600 abandoned in favor of the existing algebraic classes
  4601 ab_semigroup_mult, comm_monoid_mult, ab_semigroup_idem_mult,
  4602 lower_semilattice (resp. upper_semilattice) and linorder.
  4603 INCOMPATIBILITY.
  4604 
  4605 * Theory Transitive_Closure: induct and cases rules now declare proper
  4606 case_names ("base" and "step").  INCOMPATIBILITY.
  4607 
  4608 * Theorem Inductive.lfp_ordinal_induct generalized to complete
  4609 lattices.  The form set-specific version is available as
  4610 Inductive.lfp_ordinal_induct_set.
  4611 
  4612 * Renamed theorems "power.simps" to "power_int.simps".
  4613 INCOMPATIBILITY.
  4614 
  4615 * Class semiring_div provides basic abstract properties of semirings
  4616 with division and modulo operations.  Subsumes former class dvd_mod.
  4617 
  4618 * Merged theories IntDef, Numeral and IntArith into unified theory
  4619 Int.  INCOMPATIBILITY.
  4620 
  4621 * Theory Library/Code_Index: type "index" now represents natural
  4622 numbers rather than integers.  INCOMPATIBILITY.
  4623 
  4624 * New class "uminus" with operation "uminus" (split of from class
  4625 "minus" which now only has operation "minus", binary).
  4626 INCOMPATIBILITY.
  4627 
  4628 * Constants "card", "internal_split", "option_map" now with authentic
  4629 syntax.  INCOMPATIBILITY.
  4630 
  4631 * Definitions subset_def, psubset_def, set_diff_def, Compl_def,
  4632 le_bool_def, less_bool_def, le_fun_def, less_fun_def, inf_bool_def,
  4633 sup_bool_def, Inf_bool_def, Sup_bool_def, inf_fun_def, sup_fun_def,
  4634 Inf_fun_def, Sup_fun_def, inf_set_def, sup_set_def, Inf_set_def,
  4635 Sup_set_def, le_def, less_def, option_map_def now with object
  4636 equality.  INCOMPATIBILITY.
  4637 
  4638 * Records. Removed K_record, and replaced it by pure lambda term
  4639 %x. c. The simplifier setup is now more robust against eta expansion.
  4640 INCOMPATIBILITY: in cases explicitly referring to K_record.
  4641 
  4642 * Library/Multiset: {#a, b, c#} abbreviates {#a#} + {#b#} + {#c#}.
  4643 
  4644 * Library/ListVector: new theory of arithmetic vector operations.
  4645 
  4646 * Library/Order_Relation: new theory of various orderings as sets of
  4647 pairs.  Defines preorders, partial orders, linear orders and
  4648 well-orders on sets and on types.
  4649 
  4650 
  4651 *** ZF ***
  4652 
  4653 * Renamed some theories to allow to loading both ZF and HOL in the
  4654 same session:
  4655 
  4656   Datatype  -> Datatype_ZF
  4657   Inductive -> Inductive_ZF
  4658   Int       -> Int_ZF
  4659   IntDiv    -> IntDiv_ZF
  4660   Nat       -> Nat_ZF
  4661   List      -> List_ZF
  4662   Main      -> Main_ZF
  4663 
  4664 INCOMPATIBILITY: ZF theories that import individual theories below
  4665 Main might need to be adapted.  Regular theory Main is still
  4666 available, as trivial extension of Main_ZF.
  4667 
  4668 
  4669 *** ML ***
  4670 
  4671 * ML within Isar: antiquotation @{const name} or @{const
  4672 name(typargs)} produces statically-checked Const term.
  4673 
  4674 * Functor NamedThmsFun: data is available to the user as dynamic fact
  4675 (of the same name).  Removed obsolete print command.
  4676 
  4677 * Removed obsolete "use_legacy_bindings" function.
  4678 
  4679 * The ``print mode'' is now a thread-local value derived from a global
  4680 template (the former print_mode reference), thus access becomes
  4681 non-critical.  The global print_mode reference is for session
  4682 management only; user-code should use print_mode_value,
  4683 print_mode_active, PrintMode.setmp etc.  INCOMPATIBILITY.
  4684 
  4685 * Functions system/system_out provide a robust way to invoke external
  4686 shell commands, with propagation of interrupts (requires Poly/ML
  4687 5.2.1).  Do not use OS.Process.system etc. from the basis library!
  4688 
  4689 
  4690 *** System ***
  4691 
  4692 * Default settings: PROOFGENERAL_OPTIONS no longer impose xemacs ---
  4693 in accordance with Proof General 3.7, which prefers GNU emacs.
  4694 
  4695 * isatool tty runs Isabelle process with plain tty interaction;
  4696 optional line editor may be specified via ISABELLE_LINE_EDITOR
  4697 setting, the default settings attempt to locate "ledit" and "rlwrap".
  4698 
  4699 * isatool browser now works with Cygwin as well, using general
  4700 "javapath" function defined in Isabelle process environment.
  4701 
  4702 * YXML notation provides a simple and efficient alternative to
  4703 standard XML transfer syntax.  See src/Pure/General/yxml.ML and
  4704 isatool yxml as described in the Isabelle system manual.
  4705 
  4706 * JVM class isabelle.IsabelleProcess (located in Isabelle/lib/classes)
  4707 provides general wrapper for managing an Isabelle process in a robust
  4708 fashion, with ``cooked'' output from stdin/stderr.
  4709 
  4710 * Rudimentary Isabelle plugin for jEdit (see Isabelle/lib/jedit),
  4711 based on Isabelle/JVM process wrapper (see Isabelle/lib/classes).
  4712 
  4713 * Removed obsolete THIS_IS_ISABELLE_BUILD feature.  NB: the documented
  4714 way of changing the user's settings is via
  4715 ISABELLE_HOME_USER/etc/settings, which is a fully featured bash
  4716 script.
  4717 
  4718 * Multithreading.max_threads := 0 refers to the number of actual CPU
  4719 cores of the underlying machine, which is a good starting point for
  4720 optimal performance tuning.  The corresponding usedir option -M allows
  4721 "max" as an alias for "0".  WARNING: does not work on certain versions
  4722 of Mac OS (with Poly/ML 5.1).
  4723 
  4724 * isabelle-process: non-ML sessions are run with "nice", to reduce the
  4725 adverse effect of Isabelle flooding interactive front-ends (notably
  4726 ProofGeneral / XEmacs).
  4727 
  4728 
  4729 
  4730 New in Isabelle2007 (November 2007)
  4731 -----------------------------------
  4732 
  4733 *** General ***
  4734 
  4735 * More uniform information about legacy features, notably a
  4736 warning/error of "Legacy feature: ...", depending on the state of the
  4737 tolerate_legacy_features flag (default true). FUTURE INCOMPATIBILITY:
  4738 legacy features will disappear eventually.
  4739 
  4740 * Theory syntax: the header format ``theory A = B + C:'' has been
  4741 discontinued in favour of ``theory A imports B C begin''.  Use isatool
  4742 fixheaders to convert existing theory files.  INCOMPATIBILITY.
  4743 
  4744 * Theory syntax: the old non-Isar theory file format has been
  4745 discontinued altogether.  Note that ML proof scripts may still be used
  4746 with Isar theories; migration is usually quite simple with the ML
  4747 function use_legacy_bindings.  INCOMPATIBILITY.
  4748 
  4749 * Theory syntax: some popular names (e.g. 'class', 'declaration',
  4750 'fun', 'help', 'if') are now keywords.  INCOMPATIBILITY, use double
  4751 quotes.
  4752 
  4753 * Theory loader: be more serious about observing the static theory
  4754 header specifications (including optional directories), but not the
  4755 accidental file locations of previously successful loads.  The strict
  4756 update policy of former update_thy is now already performed by
  4757 use_thy, so the former has been removed; use_thys updates several
  4758 theories simultaneously, just as 'imports' within a theory header
  4759 specification, but without merging the results.  Potential
  4760 INCOMPATIBILITY: may need to refine theory headers and commands
  4761 ROOT.ML which depend on load order.
  4762 
  4763 * Theory loader: optional support for content-based file
  4764 identification, instead of the traditional scheme of full physical
  4765 path plus date stamp; configured by the ISABELLE_FILE_IDENT setting
  4766 (cf. the system manual).  The new scheme allows to work with
  4767 non-finished theories in persistent session images, such that source
  4768 files may be moved later on without requiring reloads.
  4769 
  4770 * Theory loader: old-style ML proof scripts being *attached* to a thy
  4771 file (with the same base name as the theory) are considered a legacy
  4772 feature, which will disappear eventually. Even now, the theory loader
  4773 no longer maintains dependencies on such files.
  4774 
  4775 * Syntax: the scope for resolving ambiguities via type-inference is
  4776 now limited to individual terms, instead of whole simultaneous
  4777 specifications as before. This greatly reduces the complexity of the
  4778 syntax module and improves flexibility by separating parsing and
  4779 type-checking. INCOMPATIBILITY: additional type-constraints (explicit
  4780 'fixes' etc.) are required in rare situations.
  4781 
  4782 * Syntax: constants introduced by new-style packages ('definition',
  4783 'abbreviation' etc.) are passed through the syntax module in
  4784 ``authentic mode''. This means that associated mixfix annotations
  4785 really stick to such constants, independently of potential name space
  4786 ambiguities introduced later on. INCOMPATIBILITY: constants in parse
  4787 trees are represented slightly differently, may need to adapt syntax
  4788 translations accordingly. Use CONST marker in 'translations' and
  4789 @{const_syntax} antiquotation in 'parse_translation' etc.
  4790 
  4791 * Legacy goal package: reduced interface to the bare minimum required
  4792 to keep existing proof scripts running.  Most other user-level
  4793 functions are now part of the OldGoals structure, which is *not* open
  4794 by default (consider isatool expandshort before open OldGoals).
  4795 Removed top_sg, prin, printyp, pprint_term/typ altogether, because
  4796 these tend to cause confusion about the actual goal (!) context being
  4797 used here, which is not necessarily the same as the_context().
  4798 
  4799 * Command 'find_theorems': supports "*" wild-card in "name:"
  4800 criterion; "with_dups" option.  Certain ProofGeneral versions might
  4801 support a specific search form (see ProofGeneral/CHANGES).
  4802 
  4803 * The ``prems limit'' option (cf. ProofContext.prems_limit) is now -1
  4804 by default, which means that "prems" (and also "fixed variables") are
  4805 suppressed from proof state output.  Note that the ProofGeneral
  4806 settings mechanism allows to change and save options persistently, but
  4807 older versions of Isabelle will fail to start up if a negative prems
  4808 limit is imposed.
  4809 
  4810 * Local theory targets may be specified by non-nested blocks of
  4811 ``context/locale/class ... begin'' followed by ``end''.  The body may
  4812 contain definitions, theorems etc., including any derived mechanism
  4813 that has been implemented on top of these primitives.  This concept
  4814 generalizes the existing ``theorem (in ...)'' towards more versatility
  4815 and scalability.
  4816 
  4817 * Proof General interface: proper undo of final 'end' command;
  4818 discontinued Isabelle/classic mode (ML proof scripts).
  4819 
  4820 
  4821 *** Document preparation ***
  4822 
  4823 * Added antiquotation @{theory name} which prints the given name,
  4824 after checking that it refers to a valid ancestor theory in the
  4825 current context.
  4826 
  4827 * Added antiquotations @{ML_type text} and @{ML_struct text} which
  4828 check the given source text as ML type/structure, printing verbatim.
  4829 
  4830 * Added antiquotation @{abbrev "c args"} which prints the abbreviation
  4831 "c args == rhs" given in the current context.  (Any number of
  4832 arguments may be given on the LHS.)
  4833 
  4834 
  4835 *** Pure ***
  4836 
  4837 * The 'class' package offers a combination of axclass and locale to
  4838 achieve Haskell-like type classes in Isabelle.  Definitions and
  4839 theorems within a class context produce both relative results (with
  4840 implicit parameters according to the locale context), and polymorphic
  4841 constants with qualified polymorphism (according to the class
  4842 context).  Within the body context of a 'class' target, a separate
  4843 syntax layer ("user space type system") takes care of converting
  4844 between global polymorphic consts and internal locale representation.
  4845 See src/HOL/ex/Classpackage.thy for examples (as well as main HOL).
  4846 "isatool doc classes" provides a tutorial.
  4847 
  4848 * Generic code generator framework allows to generate executable
  4849 code for ML and Haskell (including Isabelle classes).  A short usage
  4850 sketch:
  4851 
  4852     internal compilation:
  4853         export_code <list of constants (term syntax)> in SML
  4854     writing SML code to a file:
  4855         export_code <list of constants (term syntax)> in SML <filename>
  4856     writing OCaml code to a file:
  4857         export_code <list of constants (term syntax)> in OCaml <filename>
  4858     writing Haskell code to a bunch of files:
  4859         export_code <list of constants (term syntax)> in Haskell <filename>
  4860 
  4861     evaluating closed propositions to True/False using code generation:
  4862         method ``eval''
  4863 
  4864 Reasonable default setup of framework in HOL.
  4865 
  4866 Theorem attributs for selecting and transforming function equations theorems:
  4867 
  4868     [code fun]:        select a theorem as function equation for a specific constant
  4869     [code fun del]:    deselect a theorem as function equation for a specific constant
  4870     [code inline]:     select an equation theorem for unfolding (inlining) in place
  4871     [code inline del]: deselect an equation theorem for unfolding (inlining) in place
  4872 
  4873 User-defined serializations (target in {SML, OCaml, Haskell}):
  4874 
  4875     code_const <and-list of constants (term syntax)>
  4876       {(target) <and-list of const target syntax>}+
  4877 
  4878     code_type <and-list of type constructors>
  4879       {(target) <and-list of type target syntax>}+
  4880 
  4881     code_instance <and-list of instances>
  4882       {(target)}+
  4883         where instance ::= <type constructor> :: <class>
  4884 
  4885     code_class <and_list of classes>
  4886       {(target) <and-list of class target syntax>}+
  4887         where class target syntax ::= <class name> {where {<classop> == <target syntax>}+}?
  4888 
  4889 code_instance and code_class only are effective to target Haskell.
  4890 
  4891 For example usage see src/HOL/ex/Codegenerator.thy and
  4892 src/HOL/ex/Codegenerator_Pretty.thy.  A separate tutorial on code
  4893 generation from Isabelle/HOL theories is available via "isatool doc
  4894 codegen".
  4895 
  4896 * Code generator: consts in 'consts_code' Isar commands are now
  4897 referred to by usual term syntax (including optional type
  4898 annotations).
  4899 
  4900 * Command 'no_translations' removes translation rules from theory
  4901 syntax.
  4902 
  4903 * Overloaded definitions are now actually checked for acyclic
  4904 dependencies.  The overloading scheme is slightly more general than
  4905 that of Haskell98, although Isabelle does not demand an exact
  4906 correspondence to type class and instance declarations.
  4907 INCOMPATIBILITY, use ``defs (unchecked overloaded)'' to admit more
  4908 exotic versions of overloading -- at the discretion of the user!
  4909 
  4910 Polymorphic constants are represented via type arguments, i.e. the
  4911 instantiation that matches an instance against the most general
  4912 declaration given in the signature.  For example, with the declaration
  4913 c :: 'a => 'a => 'a, an instance c :: nat => nat => nat is represented
  4914 as c(nat).  Overloading is essentially simultaneous structural
  4915 recursion over such type arguments.  Incomplete specification patterns
  4916 impose global constraints on all occurrences, e.g. c('a * 'a) on the
  4917 LHS means that more general c('a * 'b) will be disallowed on any RHS.
  4918 Command 'print_theory' outputs the normalized system of recursive
  4919 equations, see section "definitions".
  4920 
  4921 * Configuration options are maintained within the theory or proof
  4922 context (with name and type bool/int/string), providing a very simple
  4923 interface to a poor-man's version of general context data.  Tools may
  4924 declare options in ML (e.g. using Attrib.config_int) and then refer to
  4925 these values using Config.get etc.  Users may change options via an
  4926 associated attribute of the same name.  This form of context
  4927 declaration works particularly well with commands 'declare' or
  4928 'using', for example ``declare [[foo = 42]]''.  Thus it has become
  4929 very easy to avoid global references, which would not observe Isar
  4930 toplevel undo/redo and fail to work with multithreading.
  4931 
  4932 Various global ML references of Pure and HOL have been turned into
  4933 configuration options:
  4934 
  4935   Unify.search_bound		unify_search_bound
  4936   Unify.trace_bound		unify_trace_bound
  4937   Unify.trace_simp		unify_trace_simp
  4938   Unify.trace_types		unify_trace_types
  4939   Simplifier.simp_depth_limit	simp_depth_limit
  4940   Blast.depth_limit		blast_depth_limit
  4941   DatatypeProp.dtK		datatype_distinctness_limit
  4942   fast_arith_neq_limit  	fast_arith_neq_limit
  4943   fast_arith_split_limit	fast_arith_split_limit
  4944 
  4945 * Named collections of theorems may be easily installed as context
  4946 data using the functor NamedThmsFun (see also
  4947 src/Pure/Tools/named_thms.ML).  The user may add or delete facts via
  4948 attributes; there is also a toplevel print command.  This facility is
  4949 just a common case of general context data, which is the preferred way
  4950 for anything more complex than just a list of facts in canonical
  4951 order.
  4952 
  4953 * Isar: command 'declaration' augments a local theory by generic
  4954 declaration functions written in ML.  This enables arbitrary content
  4955 being added to the context, depending on a morphism that tells the
  4956 difference of the original declaration context wrt. the application
  4957 context encountered later on.
  4958 
  4959 * Isar: proper interfaces for simplification procedures.  Command
  4960 'simproc_setup' declares named simprocs (with match patterns, and body
  4961 text in ML).  Attribute "simproc" adds/deletes simprocs in the current
  4962 context.  ML antiquotation @{simproc name} retrieves named simprocs.
  4963 
  4964 * Isar: an extra pair of brackets around attribute declarations
  4965 abbreviates a theorem reference involving an internal dummy fact,
  4966 which will be ignored later --- only the effect of the attribute on
  4967 the background context will persist.  This form of in-place
  4968 declarations is particularly useful with commands like 'declare' and
  4969 'using', for example ``have A using [[simproc a]] by simp''.
  4970 
  4971 * Isar: method "assumption" (and implicit closing of subproofs) now
  4972 takes simple non-atomic goal assumptions into account: after applying
  4973 an assumption as a rule the resulting subgoals are solved by atomic
  4974 assumption steps.  This is particularly useful to finish 'obtain'
  4975 goals, such as "!!x. (!!x. P x ==> thesis) ==> P x ==> thesis",
  4976 without referring to the original premise "!!x. P x ==> thesis" in the
  4977 Isar proof context.  POTENTIAL INCOMPATIBILITY: method "assumption" is
  4978 more permissive.
  4979 
  4980 * Isar: implicit use of prems from the Isar proof context is
  4981 considered a legacy feature.  Common applications like ``have A .''
  4982 may be replaced by ``have A by fact'' or ``note `A`''.  In general,
  4983 referencing facts explicitly here improves readability and
  4984 maintainability of proof texts.
  4985 
  4986 * Isar: improper proof element 'guess' is like 'obtain', but derives
  4987 the obtained context from the course of reasoning!  For example:
  4988 
  4989   assume "EX x y. A x & B y"   -- "any previous fact"
  4990   then guess x and y by clarify
  4991 
  4992 This technique is potentially adventurous, depending on the facts and
  4993 proof tools being involved here.
  4994 
  4995 * Isar: known facts from the proof context may be specified as literal
  4996 propositions, using ASCII back-quote syntax.  This works wherever
  4997 named facts used to be allowed so far, in proof commands, proof
  4998 methods, attributes etc.  Literal facts are retrieved from the context
  4999 according to unification of type and term parameters.  For example,
  5000 provided that "A" and "A ==> B" and "!!x. P x ==> Q x" are known
  5001 theorems in the current context, then these are valid literal facts:
  5002 `A` and `A ==> B` and `!!x. P x ==> Q x" as well as `P a ==> Q a` etc.
  5003 
  5004 There is also a proof method "fact" which does the same composition
  5005 for explicit goal states, e.g. the following proof texts coincide with
  5006 certain special cases of literal facts:
  5007 
  5008   have "A" by fact                 ==  note `A`
  5009   have "A ==> B" by fact           ==  note `A ==> B`
  5010   have "!!x. P x ==> Q x" by fact  ==  note `!!x. P x ==> Q x`
  5011   have "P a ==> Q a" by fact       ==  note `P a ==> Q a`
  5012 
  5013 * Isar: ":" (colon) is no longer a symbolic identifier character in
  5014 outer syntax.  Thus symbolic identifiers may be used without
  5015 additional white space in declarations like this: ``assume *: A''.
  5016 
  5017 * Isar: 'print_facts' prints all local facts of the current context,
  5018 both named and unnamed ones.
  5019 
  5020 * Isar: 'def' now admits simultaneous definitions, e.g.:
  5021 
  5022   def x == "t" and y == "u"
  5023 
  5024 * Isar: added command 'unfolding', which is structurally similar to
  5025 'using', but affects both the goal state and facts by unfolding given
  5026 rewrite rules.  Thus many occurrences of the 'unfold' method or
  5027 'unfolded' attribute may be replaced by first-class proof text.
  5028 
  5029 * Isar: methods 'unfold' / 'fold', attributes 'unfolded' / 'folded',
  5030 and command 'unfolding' now all support object-level equalities
  5031 (potentially conditional).  The underlying notion of rewrite rule is
  5032 analogous to the 'rule_format' attribute, but *not* that of the
  5033 Simplifier (which is usually more generous).
  5034 
  5035 * Isar: the new attribute [rotated n] (default n = 1) rotates the
  5036 premises of a theorem by n. Useful in conjunction with drule.
  5037 
  5038 * Isar: the goal restriction operator [N] (default N = 1) evaluates a
  5039 method expression within a sandbox consisting of the first N
  5040 sub-goals, which need to exist.  For example, ``simp_all [3]''
  5041 simplifies the first three sub-goals, while (rule foo, simp_all)[]
  5042 simplifies all new goals that emerge from applying rule foo to the
  5043 originally first one.
  5044 
  5045 * Isar: schematic goals are no longer restricted to higher-order
  5046 patterns; e.g. ``lemma "?P(?x)" by (rule TrueI)'' now works as
  5047 expected.
  5048 
  5049 * Isar: the conclusion of a long theorem statement is now either
  5050 'shows' (a simultaneous conjunction, as before), or 'obtains'
  5051 (essentially a disjunction of cases with local parameters and
  5052 assumptions).  The latter allows to express general elimination rules
  5053 adequately; in this notation common elimination rules look like this:
  5054 
  5055   lemma exE:    -- "EX x. P x ==> (!!x. P x ==> thesis) ==> thesis"
  5056     assumes "EX x. P x"
  5057     obtains x where "P x"
  5058 
  5059   lemma conjE:  -- "A & B ==> (A ==> B ==> thesis) ==> thesis"
  5060     assumes "A & B"
  5061     obtains A and B
  5062 
  5063   lemma disjE:  -- "A | B ==> (A ==> thesis) ==> (B ==> thesis) ==> thesis"
  5064     assumes "A | B"
  5065     obtains
  5066       A
  5067     | B
  5068 
  5069 The subsequent classical rules even refer to the formal "thesis"
  5070 explicitly:
  5071 
  5072   lemma classical:     -- "(~ thesis ==> thesis) ==> thesis"
  5073     obtains "~ thesis"
  5074 
  5075   lemma Peirce's_Law:  -- "((thesis ==> something) ==> thesis) ==> thesis"
  5076     obtains "thesis ==> something"
  5077 
  5078 The actual proof of an 'obtains' statement is analogous to that of the
  5079 Isar proof element 'obtain', only that there may be several cases.
  5080 Optional case names may be specified in parentheses; these will be
  5081 available both in the present proof and as annotations in the
  5082 resulting rule, for later use with the 'cases' method (cf. attribute
  5083 case_names).
  5084 
  5085 * Isar: the assumptions of a long theorem statement are available as
  5086 "assms" fact in the proof context.  This is more appropriate than the
  5087 (historical) "prems", which refers to all assumptions of the current
  5088 context, including those from the target locale, proof body etc.
  5089 
  5090 * Isar: 'print_statement' prints theorems from the current theory or
  5091 proof context in long statement form, according to the syntax of a
  5092 top-level lemma.
  5093 
  5094 * Isar: 'obtain' takes an optional case name for the local context
  5095 introduction rule (default "that").
  5096 
  5097 * Isar: removed obsolete 'concl is' patterns.  INCOMPATIBILITY, use
  5098 explicit (is "_ ==> ?foo") in the rare cases where this still happens
  5099 to occur.
  5100 
  5101 * Pure: syntax "CONST name" produces a fully internalized constant
  5102 according to the current context.  This is particularly useful for
  5103 syntax translations that should refer to internal constant
  5104 representations independently of name spaces.
  5105 
  5106 * Pure: syntax constant for foo (binder "FOO ") is called "foo_binder"
  5107 instead of "FOO ". This allows multiple binder declarations to coexist
  5108 in the same context.  INCOMPATIBILITY.
  5109 
  5110 * Isar/locales: 'notation' provides a robust interface to the 'syntax'
  5111 primitive that also works in a locale context (both for constants and
  5112 fixed variables). Type declaration and internal syntactic representation
  5113 of given constants retrieved from the context. Likewise, the
  5114 'no_notation' command allows to remove given syntax annotations from the
  5115 current context.
  5116 
  5117 * Isar/locales: new derived specification elements 'axiomatization',
  5118 'definition', 'abbreviation', which support type-inference, admit
  5119 object-level specifications (equality, equivalence).  See also the
  5120 isar-ref manual.  Examples:
  5121 
  5122   axiomatization
  5123     eq  (infix "===" 50) where
  5124     eq_refl: "x === x" and eq_subst: "x === y ==> P x ==> P y"
  5125 
  5126   definition "f x y = x + y + 1"
  5127   definition g where "g x = f x x"
  5128 
  5129   abbreviation
  5130     neq  (infix "=!=" 50) where
  5131     "x =!= y == ~ (x === y)"
  5132 
  5133 These specifications may be also used in a locale context.  Then the
  5134 constants being introduced depend on certain fixed parameters, and the
  5135 constant name is qualified by the locale base name.  An internal
  5136 abbreviation takes care for convenient input and output, making the
  5137 parameters implicit and using the original short name.  See also
  5138 src/HOL/ex/Abstract_NAT.thy for an example of deriving polymorphic
  5139 entities from a monomorphic theory.
  5140 
  5141 Presently, abbreviations are only available 'in' a target locale, but
  5142 not inherited by general import expressions.  Also note that
  5143 'abbreviation' may be used as a type-safe replacement for 'syntax' +
  5144 'translations' in common applications.  The "no_abbrevs" print mode
  5145 prevents folding of abbreviations in term output.
  5146 
  5147 Concrete syntax is attached to specified constants in internal form,
  5148 independently of name spaces.  The parse tree representation is
  5149 slightly different -- use 'notation' instead of raw 'syntax', and
  5150 'translations' with explicit "CONST" markup to accommodate this.
  5151 
  5152 * Pure/Isar: unified syntax for new-style specification mechanisms
  5153 (e.g.  'definition', 'abbreviation', or 'inductive' in HOL) admits
  5154 full type inference and dummy patterns ("_").  For example:
  5155 
  5156   definition "K x _ = x"
  5157 
  5158   inductive conj for A B
  5159   where "A ==> B ==> conj A B"
  5160 
  5161 * Pure: command 'print_abbrevs' prints all constant abbreviations of
  5162 the current context.  Print mode "no_abbrevs" prevents inversion of
  5163 abbreviations on output.
  5164 
  5165 * Isar/locales: improved parameter handling: use of locales "var" and
  5166 "struct" no longer necessary; - parameter renamings are no longer
  5167 required to be injective.  For example, this allows to define
  5168 endomorphisms as locale endom = homom mult mult h.
  5169 
  5170 * Isar/locales: changed the way locales with predicates are defined.
  5171 Instead of accumulating the specification, the imported expression is
  5172 now an interpretation.  INCOMPATIBILITY: different normal form of
  5173 locale expressions.  In particular, in interpretations of locales with
  5174 predicates, goals repesenting already interpreted fragments are not
  5175 removed automatically.  Use methods `intro_locales' and
  5176 `unfold_locales'; see below.
  5177 
  5178 * Isar/locales: new methods `intro_locales' and `unfold_locales'
  5179 provide backward reasoning on locales predicates.  The methods are
  5180 aware of interpretations and discharge corresponding goals.
  5181 `intro_locales' is less aggressive then `unfold_locales' and does not
  5182 unfold predicates to assumptions.
  5183 
  5184 * Isar/locales: the order in which locale fragments are accumulated
  5185 has changed.  This enables to override declarations from fragments due
  5186 to interpretations -- for example, unwanted simp rules.
  5187 
  5188 * Isar/locales: interpretation in theories and proof contexts has been
  5189 extended.  One may now specify (and prove) equations, which are
  5190 unfolded in interpreted theorems.  This is useful for replacing
  5191 defined concepts (constants depending on locale parameters) by
  5192 concepts already existing in the target context.  Example:
  5193 
  5194   interpretation partial_order ["op <= :: [int, int] => bool"]
  5195     where "partial_order.less (op <=) (x::int) y = (x < y)"
  5196 
  5197 Typically, the constant `partial_order.less' is created by a
  5198 definition specification element in the context of locale
  5199 partial_order.
  5200 
  5201 * Method "induct": improved internal context management to support
  5202 local fixes and defines on-the-fly. Thus explicit meta-level
  5203 connectives !!  and ==> are rarely required anymore in inductive goals
  5204 (using object-logic connectives for this purpose has been long
  5205 obsolete anyway). Common proof patterns are explained in
  5206 src/HOL/Induct/Common_Patterns.thy, see also
  5207 src/HOL/Isar_examples/Puzzle.thy and src/HOL/Lambda for realistic
  5208 examples.
  5209 
  5210 * Method "induct": improved handling of simultaneous goals. Instead of
  5211 introducing object-level conjunction, the statement is now split into
  5212 several conclusions, while the corresponding symbolic cases are nested
  5213 accordingly. INCOMPATIBILITY, proofs need to be structured explicitly,
  5214 see src/HOL/Induct/Common_Patterns.thy, for example.
  5215 
  5216 * Method "induct": mutual induction rules are now specified as a list
  5217 of rule sharing the same induction cases. HOL packages usually provide
  5218 foo_bar.inducts for mutually defined items foo and bar (e.g. inductive
  5219 predicates/sets or datatypes). INCOMPATIBILITY, users need to specify
  5220 mutual induction rules differently, i.e. like this:
  5221 
  5222   (induct rule: foo_bar.inducts)
  5223   (induct set: foo bar)
  5224   (induct pred: foo bar)
  5225   (induct type: foo bar)
  5226 
  5227 The ML function ProjectRule.projections turns old-style rules into the
  5228 new format.
  5229 
  5230 * Method "coinduct": dual of induction, see
  5231 src/HOL/Library/Coinductive_List.thy for various examples.
  5232 
  5233 * Method "cases", "induct", "coinduct": the ``(open)'' option is
  5234 considered a legacy feature.
  5235 
  5236 * Attribute "symmetric" produces result with standardized schematic
  5237 variables (index 0).  Potential INCOMPATIBILITY.
  5238 
  5239 * Simplifier: by default the simplifier trace only shows top level
  5240 rewrites now. That is, trace_simp_depth_limit is set to 1 by
  5241 default. Thus there is less danger of being flooded by the trace. The
  5242 trace indicates where parts have been suppressed.
  5243   
  5244 * Provers/classical: removed obsolete classical version of elim_format
  5245 attribute; classical elim/dest rules are now treated uniformly when
  5246 manipulating the claset.
  5247 
  5248 * Provers/classical: stricter checks to ensure that supplied intro,
  5249 dest and elim rules are well-formed; dest and elim rules must have at
  5250 least one premise.
  5251 
  5252 * Provers/classical: attributes dest/elim/intro take an optional
  5253 weight argument for the rule (just as the Pure versions).  Weights are
  5254 ignored by automated tools, but determine the search order of single
  5255 rule steps.
  5256 
  5257 * Syntax: input syntax now supports dummy variable binding "%_. b",
  5258 where the body does not mention the bound variable.  Note that dummy
  5259 patterns implicitly depend on their context of bounds, which makes
  5260 "{_. _}" match any set comprehension as expected.  Potential
  5261 INCOMPATIBILITY -- parse translations need to cope with syntactic
  5262 constant "_idtdummy" in the binding position.
  5263 
  5264 * Syntax: removed obsolete syntactic constant "_K" and its associated
  5265 parse translation.  INCOMPATIBILITY -- use dummy abstraction instead,
  5266 for example "A -> B" => "Pi A (%_. B)".
  5267 
  5268 * Pure: 'class_deps' command visualizes the subclass relation, using
  5269 the graph browser tool.
  5270 
  5271 * Pure: 'print_theory' now suppresses certain internal declarations by
  5272 default; use '!' option for full details.
  5273 
  5274 
  5275 *** HOL ***
  5276 
  5277 * Method "metis" proves goals by applying the Metis general-purpose
  5278 resolution prover (see also http://gilith.com/software/metis/).
  5279 Examples are in the directory MetisExamples.  WARNING: the
  5280 Isabelle/HOL-Metis integration does not yet work properly with
  5281 multi-threading.
  5282   
  5283 * Command 'sledgehammer' invokes external automatic theorem provers as
  5284 background processes.  It generates calls to the "metis" method if
  5285 successful. These can be pasted into the proof.  Users do not have to
  5286 wait for the automatic provers to return.  WARNING: does not really
  5287 work with multi-threading.
  5288 
  5289 * New "auto_quickcheck" feature tests outermost goal statements for
  5290 potential counter-examples.  Controlled by ML references
  5291 auto_quickcheck (default true) and auto_quickcheck_time_limit (default
  5292 5000 milliseconds).  Fails silently if statements is outside of
  5293 executable fragment, or any other codgenerator problem occurs.
  5294 
  5295 * New constant "undefined" with axiom "undefined x = undefined".
  5296 
  5297 * Added class "HOL.eq", allowing for code generation with polymorphic
  5298 equality.
  5299 
  5300 * Some renaming of class constants due to canonical name prefixing in
  5301 the new 'class' package:
  5302 
  5303     HOL.abs ~> HOL.abs_class.abs
  5304     HOL.divide ~> HOL.divide_class.divide
  5305     0 ~> HOL.zero_class.zero
  5306     1 ~> HOL.one_class.one
  5307     op + ~> HOL.plus_class.plus
  5308     op - ~> HOL.minus_class.minus
  5309     uminus ~> HOL.minus_class.uminus
  5310     op * ~> HOL.times_class.times
  5311     op < ~> HOL.ord_class.less
  5312     op <= > HOL.ord_class.less_eq
  5313     Nat.power ~> Power.power_class.power
  5314     Nat.size ~> Nat.size_class.size
  5315     Numeral.number_of ~> Numeral.number_class.number_of
  5316     FixedPoint.Inf ~> Lattices.complete_lattice_class.Inf
  5317     FixedPoint.Sup ~> Lattices.complete_lattice_class.Sup
  5318     Orderings.min ~> Orderings.ord_class.min
  5319     Orderings.max ~> Orderings.ord_class.max
  5320     Divides.op div ~> Divides.div_class.div
  5321     Divides.op mod ~> Divides.div_class.mod
  5322     Divides.op dvd ~> Divides.div_class.dvd
  5323 
  5324 INCOMPATIBILITY.  Adaptions may be required in the following cases:
  5325 
  5326 a) User-defined constants using any of the names "plus", "minus",
  5327 "times", "less" or "less_eq". The standard syntax translations for
  5328 "+", "-" and "*" may go wrong.  INCOMPATIBILITY: use more specific
  5329 names.
  5330 
  5331 b) Variables named "plus", "minus", "times", "less", "less_eq"
  5332 INCOMPATIBILITY: use more specific names.
  5333 
  5334 c) Permutative equations (e.g. "a + b = b + a")
  5335 Since the change of names also changes the order of terms, permutative
  5336 rewrite rules may get applied in a different order. Experience shows
  5337 that this is rarely the case (only two adaptions in the whole Isabelle
  5338 distribution).  INCOMPATIBILITY: rewrite proofs
  5339 
  5340 d) ML code directly refering to constant names
  5341 This in general only affects hand-written proof tactics, simprocs and
  5342 so on.  INCOMPATIBILITY: grep your sourcecode and replace names.
  5343 Consider using @{const_name} antiquotation.
  5344 
  5345 * New class "default" with associated constant "default".
  5346 
  5347 * Function "sgn" is now overloaded and available on int, real, complex
  5348 (and other numeric types), using class "sgn".  Two possible defs of
  5349 sgn are given as equational assumptions in the classes sgn_if and
  5350 sgn_div_norm; ordered_idom now also inherits from sgn_if.
  5351 INCOMPATIBILITY.
  5352 
  5353 * Locale "partial_order" now unified with class "order" (cf. theory
  5354 Orderings), added parameter "less".  INCOMPATIBILITY.
  5355 
  5356 * Renamings in classes "order" and "linorder": facts "refl", "trans" and
  5357 "cases" to "order_refl", "order_trans" and "linorder_cases", to avoid
  5358 clashes with HOL "refl" and "trans".  INCOMPATIBILITY.
  5359 
  5360 * Classes "order" and "linorder": potential INCOMPATIBILITY due to
  5361 changed order of proof goals in instance proofs.
  5362 
  5363 * The transitivity reasoner for partial and linear orders is set up
  5364 for classes "order" and "linorder".  Instances of the reasoner are available
  5365 in all contexts importing or interpreting the corresponding locales.
  5366 Method "order" invokes the reasoner separately; the reasoner
  5367 is also integrated with the Simplifier as a solver.  Diagnostic
  5368 command 'print_orders' shows the available instances of the reasoner
  5369 in the current context.
  5370 
  5371 * Localized monotonicity predicate in theory "Orderings"; integrated
  5372 lemmas max_of_mono and min_of_mono with this predicate.
  5373 INCOMPATIBILITY.
  5374 
  5375 * Formulation of theorem "dense" changed slightly due to integration
  5376 with new class dense_linear_order.
  5377 
  5378 * Uniform lattice theory development in HOL.
  5379 
  5380     constants "meet" and "join" now named "inf" and "sup"
  5381     constant "Meet" now named "Inf"
  5382 
  5383     classes "meet_semilorder" and "join_semilorder" now named
  5384       "lower_semilattice" and "upper_semilattice"
  5385     class "lorder" now named "lattice"
  5386     class "comp_lat" now named "complete_lattice"
  5387 
  5388     Instantiation of lattice classes allows explicit definitions
  5389     for "inf" and "sup" operations (or "Inf" and "Sup" for complete lattices).
  5390 
  5391   INCOMPATIBILITY.  Theorem renames:
  5392 
  5393     meet_left_le            ~> inf_le1
  5394     meet_right_le           ~> inf_le2
  5395     join_left_le            ~> sup_ge1
  5396     join_right_le           ~> sup_ge2
  5397     meet_join_le            ~> inf_sup_ord
  5398     le_meetI                ~> le_infI
  5399     join_leI                ~> le_supI
  5400     le_meet                 ~> le_inf_iff
  5401     le_join                 ~> ge_sup_conv
  5402     meet_idempotent         ~> inf_idem
  5403     join_idempotent         ~> sup_idem
  5404     meet_comm               ~> inf_commute
  5405     join_comm               ~> sup_commute
  5406     meet_leI1               ~> le_infI1
  5407     meet_leI2               ~> le_infI2
  5408     le_joinI1               ~> le_supI1
  5409     le_joinI2               ~> le_supI2
  5410     meet_assoc              ~> inf_assoc
  5411     join_assoc              ~> sup_assoc
  5412     meet_left_comm          ~> inf_left_commute
  5413     meet_left_idempotent    ~> inf_left_idem
  5414     join_left_comm          ~> sup_left_commute
  5415     join_left_idempotent    ~> sup_left_idem
  5416     meet_aci                ~> inf_aci
  5417     join_aci                ~> sup_aci
  5418     le_def_meet             ~> le_iff_inf
  5419     le_def_join             ~> le_iff_sup
  5420     join_absorp2            ~> sup_absorb2
  5421     join_absorp1            ~> sup_absorb1
  5422     meet_absorp1            ~> inf_absorb1
  5423     meet_absorp2            ~> inf_absorb2
  5424     meet_join_absorp        ~> inf_sup_absorb
  5425     join_meet_absorp        ~> sup_inf_absorb
  5426     distrib_join_le         ~> distrib_sup_le
  5427     distrib_meet_le         ~> distrib_inf_le
  5428 
  5429     add_meet_distrib_left   ~> add_inf_distrib_left
  5430     add_join_distrib_left   ~> add_sup_distrib_left
  5431     is_join_neg_meet        ~> is_join_neg_inf
  5432     is_meet_neg_join        ~> is_meet_neg_sup
  5433     add_meet_distrib_right  ~> add_inf_distrib_right
  5434     add_join_distrib_right  ~> add_sup_distrib_right
  5435     add_meet_join_distribs  ~> add_sup_inf_distribs
  5436     join_eq_neg_meet        ~> sup_eq_neg_inf
  5437     meet_eq_neg_join        ~> inf_eq_neg_sup
  5438     add_eq_meet_join        ~> add_eq_inf_sup
  5439     meet_0_imp_0            ~> inf_0_imp_0
  5440     join_0_imp_0            ~> sup_0_imp_0
  5441     meet_0_eq_0             ~> inf_0_eq_0
  5442     join_0_eq_0             ~> sup_0_eq_0
  5443     neg_meet_eq_join        ~> neg_inf_eq_sup
  5444     neg_join_eq_meet        ~> neg_sup_eq_inf
  5445     join_eq_if              ~> sup_eq_if
  5446 
  5447     mono_meet               ~> mono_inf
  5448     mono_join               ~> mono_sup
  5449     meet_bool_eq            ~> inf_bool_eq
  5450     join_bool_eq            ~> sup_bool_eq
  5451     meet_fun_eq             ~> inf_fun_eq
  5452     join_fun_eq             ~> sup_fun_eq
  5453     meet_set_eq             ~> inf_set_eq
  5454     join_set_eq             ~> sup_set_eq
  5455     meet1_iff               ~> inf1_iff
  5456     meet2_iff               ~> inf2_iff
  5457     meet1I                  ~> inf1I
  5458     meet2I                  ~> inf2I
  5459     meet1D1                 ~> inf1D1
  5460     meet2D1                 ~> inf2D1
  5461     meet1D2                 ~> inf1D2
  5462     meet2D2                 ~> inf2D2
  5463     meet1E                  ~> inf1E
  5464     meet2E                  ~> inf2E
  5465     join1_iff               ~> sup1_iff
  5466     join2_iff               ~> sup2_iff
  5467     join1I1                 ~> sup1I1
  5468     join2I1                 ~> sup2I1
  5469     join1I1                 ~> sup1I1
  5470     join2I2                 ~> sup1I2
  5471     join1CI                 ~> sup1CI
  5472     join2CI                 ~> sup2CI
  5473     join1E                  ~> sup1E
  5474     join2E                  ~> sup2E
  5475 
  5476     is_meet_Meet            ~> is_meet_Inf
  5477     Meet_bool_def           ~> Inf_bool_def
  5478     Meet_fun_def            ~> Inf_fun_def
  5479     Meet_greatest           ~> Inf_greatest
  5480     Meet_lower              ~> Inf_lower
  5481     Meet_set_def            ~> Inf_set_def
  5482 
  5483     Sup_def                 ~> Sup_Inf
  5484     Sup_bool_eq             ~> Sup_bool_def
  5485     Sup_fun_eq              ~> Sup_fun_def
  5486     Sup_set_eq              ~> Sup_set_def
  5487 
  5488     listsp_meetI            ~> listsp_infI
  5489     listsp_meet_eq          ~> listsp_inf_eq
  5490 
  5491     meet_min                ~> inf_min
  5492     join_max                ~> sup_max
  5493 
  5494 * Added syntactic class "size"; overloaded constant "size" now has
  5495 type "'a::size ==> bool"
  5496 
  5497 * Internal reorganisation of `size' of datatypes: size theorems
  5498 "foo.size" are no longer subsumed by "foo.simps" (but are still
  5499 simplification rules by default!); theorems "prod.size" now named
  5500 "*.size".
  5501 
  5502 * Class "div" now inherits from class "times" rather than "type".
  5503 INCOMPATIBILITY.
  5504 
  5505 * HOL/Finite_Set: "name-space" locales Lattice, Distrib_lattice,
  5506 Linorder etc.  have disappeared; operations defined in terms of
  5507 fold_set now are named Inf_fin, Sup_fin.  INCOMPATIBILITY.
  5508 
  5509 * HOL/Nat: neq0_conv no longer declared as iff.  INCOMPATIBILITY.
  5510 
  5511 * HOL-Word: New extensive library and type for generic, fixed size
  5512 machine words, with arithmetic, bit-wise, shifting and rotating
  5513 operations, reflection into int, nat, and bool lists, automation for
  5514 linear arithmetic (by automatic reflection into nat or int), including
  5515 lemmas on overflow and monotonicity.  Instantiated to all appropriate
  5516 arithmetic type classes, supporting automatic simplification of
  5517 numerals on all operations.
  5518 
  5519 * Library/Boolean_Algebra: locales for abstract boolean algebras.
  5520 
  5521 * Library/Numeral_Type: numbers as types, e.g. TYPE(32).
  5522 
  5523 * Code generator library theories:
  5524   - Code_Integer represents HOL integers by big integer literals in target
  5525     languages.
  5526   - Code_Char represents HOL characters by character literals in target
  5527     languages.
  5528   - Code_Char_chr like Code_Char, but also offers treatment of character
  5529     codes; includes Code_Integer.
  5530   - Executable_Set allows to generate code for finite sets using lists.
  5531   - Executable_Rat implements rational numbers as triples (sign, enumerator,
  5532     denominator).
  5533   - Executable_Real implements a subset of real numbers, namly those
  5534     representable by rational numbers.
  5535   - Efficient_Nat implements natural numbers by integers, which in general will
  5536     result in higher efficency; pattern matching with 0/Suc is eliminated;
  5537     includes Code_Integer.
  5538   - Code_Index provides an additional datatype index which is mapped to
  5539     target-language built-in integers.
  5540   - Code_Message provides an additional datatype message_string which is isomorphic to
  5541     strings; messages are mapped to target-language strings.
  5542 
  5543 * New package for inductive predicates
  5544 
  5545   An n-ary predicate p with m parameters z_1, ..., z_m can now be defined via
  5546 
  5547     inductive
  5548       p :: "U_1 => ... => U_m => T_1 => ... => T_n => bool"
  5549       for z_1 :: U_1 and ... and z_n :: U_m
  5550     where
  5551       rule_1: "... ==> p z_1 ... z_m t_1_1 ... t_1_n"
  5552     | ...
  5553 
  5554   with full support for type-inference, rather than
  5555 
  5556     consts s :: "U_1 => ... => U_m => (T_1 * ... * T_n) set"
  5557 
  5558     abbreviation p :: "U_1 => ... => U_m => T_1 => ... => T_n => bool"
  5559     where "p z_1 ... z_m x_1 ... x_n == (x_1, ..., x_n) : s z_1 ... z_m"
  5560 
  5561     inductive "s z_1 ... z_m"
  5562     intros
  5563       rule_1: "... ==> (t_1_1, ..., t_1_n) : s z_1 ... z_m"
  5564       ...
  5565 
  5566   For backward compatibility, there is a wrapper allowing inductive
  5567   sets to be defined with the new package via
  5568 
  5569     inductive_set
  5570       s :: "U_1 => ... => U_m => (T_1 * ... * T_n) set"
  5571       for z_1 :: U_1 and ... and z_n :: U_m
  5572     where
  5573       rule_1: "... ==> (t_1_1, ..., t_1_n) : s z_1 ... z_m"
  5574     | ...
  5575 
  5576   or
  5577 
  5578     inductive_set
  5579       s :: "U_1 => ... => U_m => (T_1 * ... * T_n) set"
  5580       and p :: "U_1 => ... => U_m => T_1 => ... => T_n => bool"
  5581       for z_1 :: U_1 and ... and z_n :: U_m
  5582     where
  5583       "p z_1 ... z_m x_1 ... x_n == (x_1, ..., x_n) : s z_1 ... z_m"
  5584     | rule_1: "... ==> p z_1 ... z_m t_1_1 ... t_1_n"
  5585     | ...
  5586 
  5587   if the additional syntax "p ..." is required.
  5588 
  5589   Numerous examples can be found in the subdirectories src/HOL/Auth,
  5590   src/HOL/Bali, src/HOL/Induct, and src/HOL/MicroJava.
  5591 
  5592   INCOMPATIBILITIES:
  5593 
  5594   - Since declaration and definition of inductive sets or predicates
  5595     is no longer separated, abbreviations involving the newly
  5596     introduced sets or predicates must be specified together with the
  5597     introduction rules after the 'where' keyword (see above), rather
  5598     than before the actual inductive definition.
  5599 
  5600   - The variables in induction and elimination rules are now
  5601     quantified in the order of their occurrence in the introduction
  5602     rules, rather than in alphabetical order. Since this may break
  5603     some proofs, these proofs either have to be repaired, e.g. by
  5604     reordering the variables a_i_1 ... a_i_{k_i} in Isar 'case'
  5605     statements of the form
  5606 
  5607       case (rule_i a_i_1 ... a_i_{k_i})
  5608 
  5609     or the old order of quantification has to be restored by explicitly adding
  5610     meta-level quantifiers in the introduction rules, i.e.
  5611 
  5612       | rule_i: "!!a_i_1 ... a_i_{k_i}. ... ==> p z_1 ... z_m t_i_1 ... t_i_n"
  5613 
  5614   - The format of the elimination rules is now
  5615 
  5616       p z_1 ... z_m x_1 ... x_n ==>
  5617         (!!a_1_1 ... a_1_{k_1}. x_1 = t_1_1 ==> ... ==> x_n = t_1_n ==> ... ==> P)
  5618         ==> ... ==> P
  5619 
  5620     for predicates and
  5621 
  5622       (x_1, ..., x_n) : s z_1 ... z_m ==>
  5623         (!!a_1_1 ... a_1_{k_1}. x_1 = t_1_1 ==> ... ==> x_n = t_1_n ==> ... ==> P)
  5624         ==> ... ==> P
  5625 
  5626     for sets rather than
  5627 
  5628       x : s z_1 ... z_m ==>
  5629         (!!a_1_1 ... a_1_{k_1}. x = (t_1_1, ..., t_1_n) ==> ... ==> P)
  5630         ==> ... ==> P
  5631 
  5632     This may require terms in goals to be expanded to n-tuples
  5633     (e.g. using case_tac or simplification with the split_paired_all
  5634     rule) before the above elimination rule is applicable.
  5635 
  5636   - The elimination or case analysis rules for (mutually) inductive
  5637     sets or predicates are now called "p_1.cases" ... "p_k.cases". The
  5638     list of rules "p_1_..._p_k.elims" is no longer available.
  5639 
  5640 * New package "function"/"fun" for general recursive functions,
  5641 supporting mutual and nested recursion, definitions in local contexts,
  5642 more general pattern matching and partiality. See HOL/ex/Fundefs.thy
  5643 for small examples, and the separate tutorial on the function
  5644 package. The old recdef "package" is still available as before, but
  5645 users are encouraged to use the new package.
  5646 
  5647 * Method "lexicographic_order" automatically synthesizes termination
  5648 relations as lexicographic combinations of size measures. 
  5649 
  5650 * Case-expressions allow arbitrary constructor-patterns (including
  5651 "_") and take their order into account, like in functional
  5652 programming.  Internally, this is translated into nested
  5653 case-expressions; missing cases are added and mapped to the predefined
  5654 constant "undefined". In complicated cases printing may no longer show
  5655 the original input but the internal form. Lambda-abstractions allow
  5656 the same form of pattern matching: "% pat1 => e1 | ..." is an
  5657 abbreviation for "%x. case x of pat1 => e1 | ..." where x is a new
  5658 variable.
  5659 
  5660 * IntDef: The constant "int :: nat => int" has been removed; now "int"
  5661 is an abbreviation for "of_nat :: nat => int". The simplification
  5662 rules for "of_nat" have been changed to work like "int" did
  5663 previously.  Potential INCOMPATIBILITY:
  5664   - "of_nat (Suc m)" simplifies to "1 + of_nat m" instead of "of_nat m + 1"
  5665   - of_nat_diff and of_nat_mult are no longer default simp rules
  5666 
  5667 * Method "algebra" solves polynomial equations over (semi)rings using
  5668 Groebner bases. The (semi)ring structure is defined by locales and the
  5669 tool setup depends on that generic context. Installing the method for
  5670 a specific type involves instantiating the locale and possibly adding
  5671 declarations for computation on the coefficients.  The method is
  5672 already instantiated for natural numbers and for the axiomatic class
  5673 of idoms with numerals.  See also the paper by Chaieb and Wenzel at
  5674 CALCULEMUS 2007 for the general principles underlying this
  5675 architecture of context-aware proof-tools.
  5676 
  5677 * Method "ferrack" implements quantifier elimination over
  5678 special-purpose dense linear orders using locales (analogous to
  5679 "algebra"). The method is already installed for class
  5680 {ordered_field,recpower,number_ring} which subsumes real, hyperreal,
  5681 rat, etc.
  5682 
  5683 * Former constant "List.op @" now named "List.append".  Use ML
  5684 antiquotations @{const_name List.append} or @{term " ... @ ... "} to
  5685 circumvent possible incompatibilities when working on ML level.
  5686 
  5687 * primrec: missing cases mapped to "undefined" instead of "arbitrary".
  5688 
  5689 * New function listsum :: 'a list => 'a for arbitrary monoids.
  5690 Special syntax: "SUM x <- xs. f x" (and latex variants)
  5691 
  5692 * New syntax for Haskell-like list comprehension (input only), eg.
  5693 [(x,y). x <- xs, y <- ys, x ~= y], see also src/HOL/List.thy.
  5694 
  5695 * The special syntax for function "filter" has changed from [x :
  5696 xs. P] to [x <- xs. P] to avoid an ambiguity caused by list
  5697 comprehension syntax, and for uniformity.  INCOMPATIBILITY.
  5698 
  5699 * [a..b] is now defined for arbitrary linear orders.  It used to be
  5700 defined on nat only, as an abbreviation for [a..<Suc b]
  5701 INCOMPATIBILITY.
  5702 
  5703 * Renamed lemma "set_take_whileD"  to "set_takeWhileD".
  5704 
  5705 * New functions "sorted" and "sort" in src/HOL/List.thy.
  5706 
  5707 * New lemma collection field_simps (an extension of ring_simps) for
  5708 manipulating (in)equations involving division. Multiplies with all
  5709 denominators that can be proved to be non-zero (in equations) or
  5710 positive/negative (in inequations).
  5711 
  5712 * Lemma collections ring_eq_simps, group_eq_simps and ring_distrib
  5713 have been improved and renamed to ring_simps, group_simps and
  5714 ring_distribs.  Removed lemmas field_xyz in theory Ring_and_Field
  5715 because they were subsumed by lemmas xyz.  INCOMPATIBILITY.
  5716 
  5717 * Theory Library/Commutative_Ring: switched from recdef to function
  5718 package; constants add, mul, pow now curried.  Infix syntax for
  5719 algebraic operations.
  5720 
  5721 * Dropped redundant lemma def_imp_eq in favor of meta_eq_to_obj_eq.
  5722 INCOMPATIBILITY.
  5723 
  5724 * Dropped redundant lemma if_def2 in favor of if_bool_eq_conj.
  5725 INCOMPATIBILITY.
  5726 
  5727 * HOL/records: generalised field-update to take a function on the
  5728 field rather than the new value: r(|A := x|) is translated to A_update
  5729 (K x) r The K-combinator that is internally used is called K_record.
  5730 INCOMPATIBILITY: Usage of the plain update functions has to be
  5731 adapted.
  5732  
  5733 * Class "semiring_0" now contains annihilation axioms x * 0 = 0 and 0
  5734 * x = 0, which are required for a semiring.  Richer structures do not
  5735 inherit from semiring_0 anymore, because this property is a theorem
  5736 there, not an axiom.  INCOMPATIBILITY: In instances of semiring_0,
  5737 there is more to prove, but this is mostly trivial.
  5738 
  5739 * Class "recpower" is generalized to arbitrary monoids, not just
  5740 commutative semirings.  INCOMPATIBILITY: may need to incorporate
  5741 commutativity or semiring properties additionally.
  5742 
  5743 * Constant "List.list_all2" in List.thy now uses authentic syntax.
  5744 INCOMPATIBILITY: translations containing list_all2 may go wrong,
  5745 better use 'abbreviation'.
  5746 
  5747 * Renamed constant "List.op mem" to "List.member".  INCOMPATIBILITY.
  5748 
  5749 * Numeral syntax: type 'bin' which was a mere type copy of 'int' has
  5750 been abandoned in favour of plain 'int'.  INCOMPATIBILITY --
  5751 significant changes for setting up numeral syntax for types:
  5752   - New constants Numeral.pred and Numeral.succ instead
  5753       of former Numeral.bin_pred and Numeral.bin_succ.
  5754   - Use integer operations instead of bin_add, bin_mult and so on.
  5755   - Numeral simplification theorems named Numeral.numeral_simps instead of Bin_simps.
  5756   - ML structure Bin_Simprocs now named Int_Numeral_Base_Simprocs.
  5757 
  5758 See src/HOL/Integ/IntArith.thy for an example setup.
  5759 
  5760 * Command 'normal_form' computes the normal form of a term that may
  5761 contain free variables.  For example ``normal_form "rev [a, b, c]"''
  5762 produces ``[b, c, a]'' (without proof).  This command is suitable for
  5763 heavy-duty computations because the functions are compiled to ML
  5764 first.  Correspondingly, a method "normalization" is provided.  See
  5765 further src/HOL/ex/NormalForm.thy and src/Tools/nbe.ML.
  5766 
  5767 * Alternative iff syntax "A <-> B" for equality on bool (with priority
  5768 25 like -->); output depends on the "iff" print_mode, the default is
  5769 "A = B" (with priority 50).
  5770 
  5771 * Relations less (<) and less_eq (<=) are also available on type bool.
  5772 Modified syntax to disallow nesting without explicit parentheses,
  5773 e.g. "(x < y) < z" or "x < (y < z)", but NOT "x < y < z".  Potential
  5774 INCOMPATIBILITY.
  5775 
  5776 * "LEAST x:A. P" expands to "LEAST x. x:A & P" (input only).
  5777 
  5778 * Relation composition operator "op O" now has precedence 75 and binds
  5779 stronger than union and intersection. INCOMPATIBILITY.
  5780 
  5781 * The old set interval syntax "{m..n(}" (and relatives) has been
  5782 removed.  Use "{m..<n}" (and relatives) instead.
  5783 
  5784 * In the context of the assumption "~(s = t)" the Simplifier rewrites
  5785 "t = s" to False (by simproc "neq").  INCOMPATIBILITY, consider using
  5786 ``declare [[simproc del: neq]]''.
  5787 
  5788 * Simplifier: "m dvd n" where m and n are numbers is evaluated to
  5789 True/False.
  5790 
  5791 * Theorem Cons_eq_map_conv no longer declared as "simp".
  5792 
  5793 * Theorem setsum_mult renamed to setsum_right_distrib.
  5794 
  5795 * Prefer ex1I over ex_ex1I in single-step reasoning, e.g. by the
  5796 ``rule'' method.
  5797 
  5798 * Reimplemented methods "sat" and "satx", with several improvements:
  5799 goals no longer need to be stated as "<prems> ==> False", equivalences
  5800 (i.e. "=" on type bool) are handled, variable names of the form
  5801 "lit_<n>" are no longer reserved, significant speedup.
  5802 
  5803 * Methods "sat" and "satx" can now replay MiniSat proof traces.
  5804 zChaff is still supported as well.
  5805 
  5806 * 'inductive' and 'datatype': provide projections of mutual rules,
  5807 bundled as foo_bar.inducts;
  5808 
  5809 * Library: moved theories Parity, GCD, Binomial, Infinite_Set to
  5810 Library.
  5811 
  5812 * Library: moved theory Accessible_Part to main HOL.
  5813 
  5814 * Library: added theory Coinductive_List of potentially infinite lists
  5815 as greatest fixed-point.
  5816 
  5817 * Library: added theory AssocList which implements (finite) maps as
  5818 association lists.
  5819 
  5820 * Method "evaluation" solves goals (i.e. a boolean expression)
  5821 efficiently by compiling it to ML.  The goal is "proved" (via an
  5822 oracle) if it evaluates to True.
  5823 
  5824 * Linear arithmetic now splits certain operators (e.g. min, max, abs)
  5825 also when invoked by the simplifier.  This results in the Simplifier
  5826 being more powerful on arithmetic goals.  INCOMPATIBILITY.
  5827 Configuration option fast_arith_split_limit=0 recovers the old
  5828 behavior.
  5829 
  5830 * Support for hex (0x20) and binary (0b1001) numerals.
  5831 
  5832 * New method: reify eqs (t), where eqs are equations for an
  5833 interpretation I :: 'a list => 'b => 'c and t::'c is an optional
  5834 parameter, computes a term s::'b and a list xs::'a list and proves the
  5835 theorem I xs s = t. This is also known as reification or quoting. The
  5836 resulting theorem is applied to the subgoal to substitute t with I xs
  5837 s.  If t is omitted, the subgoal itself is reified.
  5838 
  5839 * New method: reflection corr_thm eqs (t). The parameters eqs and (t)
  5840 are as explained above. corr_thm is a theorem for I vs (f t) = I vs t,
  5841 where f is supposed to be a computable function (in the sense of code
  5842 generattion). The method uses reify to compute s and xs as above then
  5843 applies corr_thm and uses normalization by evaluation to "prove" f s =
  5844 r and finally gets the theorem t = r, which is again applied to the
  5845 subgoal. An Example is available in src/HOL/ex/ReflectionEx.thy.
  5846 
  5847 * Reflection: Automatic reification now handels binding, an example is
  5848 available in src/HOL/ex/ReflectionEx.thy
  5849 
  5850 * HOL-Statespace: ``State Spaces: The Locale Way'' introduces a
  5851 command 'statespace' that is similar to 'record', but introduces an
  5852 abstract specification based on the locale infrastructure instead of
  5853 HOL types.  This leads to extra flexibility in composing state spaces,
  5854 in particular multiple inheritance and renaming of components.
  5855 
  5856 
  5857 *** HOL-Complex ***
  5858 
  5859 * Hyperreal: Functions root and sqrt are now defined on negative real
  5860 inputs so that root n (- x) = - root n x and sqrt (- x) = - sqrt x.
  5861 Nonnegativity side conditions have been removed from many lemmas, so
  5862 that more subgoals may now be solved by simplification; potential
  5863 INCOMPATIBILITY.
  5864 
  5865 * Real: new type classes formalize real normed vector spaces and
  5866 algebras, using new overloaded constants scaleR :: real => 'a => 'a
  5867 and norm :: 'a => real.
  5868 
  5869 * Real: constant of_real :: real => 'a::real_algebra_1 injects from
  5870 reals into other types. The overloaded constant Reals :: 'a set is now
  5871 defined as range of_real; potential INCOMPATIBILITY.
  5872 
  5873 * Real: proper support for ML code generation, including 'quickcheck'.
  5874 Reals are implemented as arbitrary precision rationals.
  5875 
  5876 * Hyperreal: Several constants that previously worked only for the
  5877 reals have been generalized, so they now work over arbitrary vector
  5878 spaces. Type annotations may need to be added in some cases; potential
  5879 INCOMPATIBILITY.
  5880 
  5881   Infinitesimal  :: ('a::real_normed_vector) star set
  5882   HFinite        :: ('a::real_normed_vector) star set
  5883   HInfinite      :: ('a::real_normed_vector) star set
  5884   approx         :: ('a::real_normed_vector) star => 'a star => bool
  5885   monad          :: ('a::real_normed_vector) star => 'a star set
  5886   galaxy         :: ('a::real_normed_vector) star => 'a star set
  5887   (NS)LIMSEQ     :: [nat => 'a::real_normed_vector, 'a] => bool
  5888   (NS)convergent :: (nat => 'a::real_normed_vector) => bool
  5889   (NS)Bseq       :: (nat => 'a::real_normed_vector) => bool
  5890   (NS)Cauchy     :: (nat => 'a::real_normed_vector) => bool
  5891   (NS)LIM        :: ['a::real_normed_vector => 'b::real_normed_vector, 'a, 'b] => bool
  5892   is(NS)Cont     :: ['a::real_normed_vector => 'b::real_normed_vector, 'a] => bool
  5893   deriv          :: ['a::real_normed_field => 'a, 'a, 'a] => bool
  5894   sgn            :: 'a::real_normed_vector => 'a
  5895   exp            :: 'a::{recpower,real_normed_field,banach} => 'a
  5896 
  5897 * Complex: Some complex-specific constants are now abbreviations for
  5898 overloaded ones: complex_of_real = of_real, cmod = norm, hcmod =
  5899 hnorm.  Other constants have been entirely removed in favor of the
  5900 polymorphic versions (INCOMPATIBILITY):
  5901 
  5902   approx        <-- capprox
  5903   HFinite       <-- CFinite
  5904   HInfinite     <-- CInfinite
  5905   Infinitesimal <-- CInfinitesimal
  5906   monad         <-- cmonad
  5907   galaxy        <-- cgalaxy
  5908   (NS)LIM       <-- (NS)CLIM, (NS)CRLIM
  5909   is(NS)Cont    <-- is(NS)Contc, is(NS)contCR
  5910   (ns)deriv     <-- (ns)cderiv
  5911 
  5912 
  5913 *** HOL-Algebra ***
  5914 
  5915 * Formalisation of ideals and the quotient construction over rings.
  5916 
  5917 * Order and lattice theory no longer based on records.
  5918 INCOMPATIBILITY.
  5919 
  5920 * Renamed lemmas least_carrier -> least_closed and greatest_carrier ->
  5921 greatest_closed.  INCOMPATIBILITY.
  5922 
  5923 * Method algebra is now set up via an attribute.  For examples see
  5924 Ring.thy.  INCOMPATIBILITY: the method is now weaker on combinations
  5925 of algebraic structures.
  5926 
  5927 * Renamed theory CRing to Ring.
  5928 
  5929 
  5930 *** HOL-Nominal ***
  5931 
  5932 * Substantial, yet incomplete support for nominal datatypes (binding
  5933 structures) based on HOL-Nominal logic.  See src/HOL/Nominal and
  5934 src/HOL/Nominal/Examples.  Prospective users should consult
  5935 http://isabelle.in.tum.de/nominal/
  5936 
  5937 
  5938 *** ML ***
  5939 
  5940 * ML basics: just one true type int, which coincides with IntInf.int
  5941 (even on SML/NJ).
  5942 
  5943 * ML within Isar: antiquotations allow to embed statically-checked
  5944 formal entities in the source, referring to the context available at
  5945 compile-time.  For example:
  5946 
  5947 ML {* @{sort "{zero,one}"} *}
  5948 ML {* @{typ "'a => 'b"} *}
  5949 ML {* @{term "%x. x"} *}
  5950 ML {* @{prop "x == y"} *}
  5951 ML {* @{ctyp "'a => 'b"} *}
  5952 ML {* @{cterm "%x. x"} *}
  5953 ML {* @{cprop "x == y"} *}
  5954 ML {* @{thm asm_rl} *}
  5955 ML {* @{thms asm_rl} *}
  5956 ML {* @{type_name c} *}
  5957 ML {* @{type_syntax c} *}
  5958 ML {* @{const_name c} *}
  5959 ML {* @{const_syntax c} *}
  5960 ML {* @{context} *}
  5961 ML {* @{theory} *}
  5962 ML {* @{theory Pure} *}
  5963 ML {* @{theory_ref} *}
  5964 ML {* @{theory_ref Pure} *}
  5965 ML {* @{simpset} *}
  5966 ML {* @{claset} *}
  5967 ML {* @{clasimpset} *}
  5968 
  5969 The same works for sources being ``used'' within an Isar context.
  5970 
  5971 * ML in Isar: improved error reporting; extra verbosity with
  5972 ML_Context.trace enabled.
  5973 
  5974 * Pure/General/table.ML: the join operations now works via exceptions
  5975 DUP/SAME instead of type option. This is simpler in simple cases, and
  5976 admits slightly more efficient complex applications.
  5977 
  5978 * Pure: 'advanced' translation functions (parse_translation etc.) now
  5979 use Context.generic instead of just theory.
  5980 
  5981 * Pure: datatype Context.generic joins theory/Proof.context and
  5982 provides some facilities for code that works in either kind of
  5983 context, notably GenericDataFun for uniform theory and proof data.
  5984 
  5985 * Pure: simplified internal attribute type, which is now always
  5986 Context.generic * thm -> Context.generic * thm. Global (theory) vs.
  5987 local (Proof.context) attributes have been discontinued, while
  5988 minimizing code duplication. Thm.rule_attribute and
  5989 Thm.declaration_attribute build canonical attributes; see also structure
  5990 Context for further operations on Context.generic, notably
  5991 GenericDataFun. INCOMPATIBILITY, need to adapt attribute type
  5992 declarations and definitions.
  5993 
  5994 * Context data interfaces (Theory/Proof/GenericDataFun): removed
  5995 name/print, uninitialized data defaults to ad-hoc copy of empty value,
  5996 init only required for impure data. INCOMPATIBILITY: empty really need
  5997 to be empty (no dependencies on theory content!)
  5998 
  5999 * Pure/kernel: consts certification ignores sort constraints given in
  6000 signature declarations. (This information is not relevant to the
  6001 logic, but only for type inference.) SIGNIFICANT INTERNAL CHANGE,
  6002 potential INCOMPATIBILITY.
  6003 
  6004 * Pure: axiomatic type classes are now purely definitional, with
  6005 explicit proofs of class axioms and super class relations performed
  6006 internally. See Pure/axclass.ML for the main internal interfaces --
  6007 notably AxClass.define_class supercedes AxClass.add_axclass, and
  6008 AxClass.axiomatize_class/classrel/arity supersede
  6009 Sign.add_classes/classrel/arities.
  6010 
  6011 * Pure/Isar: Args/Attrib parsers operate on Context.generic --
  6012 global/local versions on theory vs. Proof.context have been
  6013 discontinued; Attrib.syntax and Method.syntax have been adapted
  6014 accordingly.  INCOMPATIBILITY, need to adapt parser expressions for
  6015 attributes, methods, etc.
  6016 
  6017 * Pure: several functions of signature "... -> theory -> theory * ..."
  6018 have been reoriented to "... -> theory -> ... * theory" in order to
  6019 allow natural usage in combination with the ||>, ||>>, |-> and
  6020 fold_map combinators.
  6021 
  6022 * Pure: official theorem names (closed derivations) and additional
  6023 comments (tags) are now strictly separate.  Name hints -- which are
  6024 maintained as tags -- may be attached any time without affecting the
  6025 derivation.
  6026 
  6027 * Pure: primitive rule lift_rule now takes goal cterm instead of an
  6028 actual goal state (thm).  Use Thm.lift_rule (Thm.cprem_of st i) to
  6029 achieve the old behaviour.
  6030 
  6031 * Pure: the "Goal" constant is now called "prop", supporting a
  6032 slightly more general idea of ``protecting'' meta-level rule
  6033 statements.
  6034 
  6035 * Pure: Logic.(un)varify only works in a global context, which is now
  6036 enforced instead of silently assumed.  INCOMPATIBILITY, may use
  6037 Logic.legacy_(un)varify as temporary workaround.
  6038 
  6039 * Pure: structure Name provides scalable operations for generating
  6040 internal variable names, notably Name.variants etc.  This replaces
  6041 some popular functions from term.ML:
  6042 
  6043   Term.variant		->  Name.variant
  6044   Term.variantlist	->  Name.variant_list
  6045   Term.invent_names	->  Name.invent_list
  6046 
  6047 Note that low-level renaming rarely occurs in new code -- operations
  6048 from structure Variable are used instead (see below).
  6049 
  6050 * Pure: structure Variable provides fundamental operations for proper
  6051 treatment of fixed/schematic variables in a context.  For example,
  6052 Variable.import introduces fixes for schematics of given facts and
  6053 Variable.export reverses the effect (up to renaming) -- this replaces
  6054 various freeze_thaw operations.
  6055 
  6056 * Pure: structure Goal provides simple interfaces for
  6057 init/conclude/finish and tactical prove operations (replacing former
  6058 Tactic.prove).  Goal.prove is the canonical way to prove results
  6059 within a given context; Goal.prove_global is a degraded version for
  6060 theory level goals, including a global Drule.standard.  Note that
  6061 OldGoals.prove_goalw_cterm has long been obsolete, since it is
  6062 ill-behaved in a local proof context (e.g. with local fixes/assumes or
  6063 in a locale context).
  6064 
  6065 * Pure/Syntax: generic interfaces for parsing (Syntax.parse_term etc.)
  6066 and type checking (Syntax.check_term etc.), with common combinations
  6067 (Syntax.read_term etc.). These supersede former Sign.read_term etc.
  6068 which are considered legacy and await removal.
  6069 
  6070 * Pure/Syntax: generic interfaces for type unchecking
  6071 (Syntax.uncheck_terms etc.) and unparsing (Syntax.unparse_term etc.),
  6072 with common combinations (Syntax.pretty_term, Syntax.string_of_term
  6073 etc.).  Former Sign.pretty_term, Sign.string_of_term etc. are still
  6074 available for convenience, but refer to the very same operations using
  6075 a mere theory instead of a full context.
  6076 
  6077 * Isar: simplified treatment of user-level errors, using exception
  6078 ERROR of string uniformly.  Function error now merely raises ERROR,
  6079 without any side effect on output channels.  The Isar toplevel takes
  6080 care of proper display of ERROR exceptions.  ML code may use plain
  6081 handle/can/try; cat_error may be used to concatenate errors like this:
  6082 
  6083   ... handle ERROR msg => cat_error msg "..."
  6084 
  6085 Toplevel ML code (run directly or through the Isar toplevel) may be
  6086 embedded into the Isar toplevel with exception display/debug like
  6087 this:
  6088 
  6089   Isar.toplevel (fn () => ...)
  6090 
  6091 INCOMPATIBILITY, removed special transform_error facilities, removed
  6092 obsolete variants of user-level exceptions (ERROR_MESSAGE,
  6093 Context.PROOF, ProofContext.CONTEXT, Proof.STATE, ProofHistory.FAIL)
  6094 -- use plain ERROR instead.
  6095 
  6096 * Isar: theory setup now has type (theory -> theory), instead of a
  6097 list.  INCOMPATIBILITY, may use #> to compose setup functions.
  6098 
  6099 * Isar: ML toplevel pretty printer for type Proof.context, subject to
  6100 ProofContext.debug/verbose flags.
  6101 
  6102 * Isar: Toplevel.theory_to_proof admits transactions that modify the
  6103 theory before entering a proof state.  Transactions now always see a
  6104 quasi-functional intermediate checkpoint, both in interactive and
  6105 batch mode.
  6106 
  6107 * Isar: simplified interfaces for outer syntax.  Renamed
  6108 OuterSyntax.add_keywords to OuterSyntax.keywords.  Removed
  6109 OuterSyntax.add_parsers -- this functionality is now included in
  6110 OuterSyntax.command etc.  INCOMPATIBILITY.
  6111 
  6112 * Simplifier: the simpset of a running simplification process now
  6113 contains a proof context (cf. Simplifier.the_context), which is the
  6114 very context that the initial simpset has been retrieved from (by
  6115 simpset_of/local_simpset_of).  Consequently, all plug-in components
  6116 (solver, looper etc.) may depend on arbitrary proof data.
  6117 
  6118 * Simplifier.inherit_context inherits the proof context (plus the
  6119 local bounds) of the current simplification process; any simproc
  6120 etc. that calls the Simplifier recursively should do this!  Removed
  6121 former Simplifier.inherit_bounds, which is already included here --
  6122 INCOMPATIBILITY.  Tools based on low-level rewriting may even have to
  6123 specify an explicit context using Simplifier.context/theory_context.
  6124 
  6125 * Simplifier/Classical Reasoner: more abstract interfaces
  6126 change_simpset/claset for modifying the simpset/claset reference of a
  6127 theory; raw versions simpset/claset_ref etc. have been discontinued --
  6128 INCOMPATIBILITY.
  6129 
  6130 * Provers: more generic wrt. syntax of object-logics, avoid hardwired
  6131 "Trueprop" etc.
  6132 
  6133 
  6134 *** System ***
  6135 
  6136 * settings: the default heap location within ISABELLE_HOME_USER now
  6137 includes ISABELLE_IDENTIFIER.  This simplifies use of multiple
  6138 Isabelle installations.
  6139 
  6140 * isabelle-process: option -S (secure mode) disables some critical
  6141 operations, notably runtime compilation and evaluation of ML source
  6142 code.
  6143 
  6144 * Basic Isabelle mode for jEdit, see Isabelle/lib/jedit/.
  6145 
  6146 * Support for parallel execution, using native multicore support of
  6147 Poly/ML 5.1.  The theory loader exploits parallelism when processing
  6148 independent theories, according to the given theory header
  6149 specifications. The maximum number of worker threads is specified via
  6150 usedir option -M or the "max-threads" setting in Proof General. A
  6151 speedup factor of 1.5--3.5 can be expected on a 4-core machine, and up
  6152 to 6 on a 8-core machine.  User-code needs to observe certain
  6153 guidelines for thread-safe programming, see appendix A in the Isar
  6154 Implementation manual.
  6155 
  6156 
  6157 
  6158 New in Isabelle2005 (October 2005)
  6159 ----------------------------------
  6160 
  6161 *** General ***
  6162 
  6163 * Theory headers: the new header syntax for Isar theories is
  6164 
  6165   theory <name>
  6166   imports <theory1> ... <theoryN>
  6167   uses <file1> ... <fileM>
  6168   begin
  6169 
  6170 where the 'uses' part is optional.  The previous syntax
  6171 
  6172   theory <name> = <theory1> + ... + <theoryN>:
  6173 
  6174 will disappear in the next release.  Use isatool fixheaders to convert
  6175 existing theory files.  Note that there is no change in ancient
  6176 non-Isar theories now, but these will disappear soon.
  6177 
  6178 * Theory loader: parent theories can now also be referred to via
  6179 relative and absolute paths.
  6180 
  6181 * Command 'find_theorems' searches for a list of criteria instead of a
  6182 list of constants. Known criteria are: intro, elim, dest, name:string,
  6183 simp:term, and any term. Criteria can be preceded by '-' to select
  6184 theorems that do not match. Intro, elim, dest select theorems that
  6185 match the current goal, name:s selects theorems whose fully qualified
  6186 name contain s, and simp:term selects all simplification rules whose
  6187 lhs match term.  Any other term is interpreted as pattern and selects
  6188 all theorems matching the pattern. Available in ProofGeneral under
  6189 'ProofGeneral -> Find Theorems' or C-c C-f.  Example:
  6190 
  6191   C-c C-f (100) "(_::nat) + _ + _" intro -name: "HOL."
  6192 
  6193 prints the last 100 theorems matching the pattern "(_::nat) + _ + _",
  6194 matching the current goal as introduction rule and not having "HOL."
  6195 in their name (i.e. not being defined in theory HOL).
  6196 
  6197 * Command 'thms_containing' has been discontinued in favour of
  6198 'find_theorems'; INCOMPATIBILITY.
  6199 
  6200 * Communication with Proof General is now 8bit clean, which means that
  6201 Unicode text in UTF-8 encoding may be used within theory texts (both
  6202 formal and informal parts).  Cf. option -U of the Isabelle Proof
  6203 General interface.  Here are some simple examples (cf. src/HOL/ex):
  6204 
  6205   http://isabelle.in.tum.de/library/HOL/ex/Hebrew.html
  6206   http://isabelle.in.tum.de/library/HOL/ex/Chinese.html
  6207 
  6208 * Improved efficiency of the Simplifier and, to a lesser degree, the
  6209 Classical Reasoner.  Typical big applications run around 2 times
  6210 faster.
  6211 
  6212 
  6213 *** Document preparation ***
  6214 
  6215 * Commands 'display_drafts' and 'print_drafts' perform simple output
  6216 of raw sources.  Only those symbols that do not require additional
  6217 LaTeX packages (depending on comments in isabellesym.sty) are
  6218 displayed properly, everything else is left verbatim.  isatool display
  6219 and isatool print are used as front ends (these are subject to the
  6220 DVI/PDF_VIEWER and PRINT_COMMAND settings, respectively).
  6221 
  6222 * Command tags control specific markup of certain regions of text,
  6223 notably folding and hiding.  Predefined tags include "theory" (for
  6224 theory begin and end), "proof" for proof commands, and "ML" for
  6225 commands involving ML code; the additional tags "visible" and
  6226 "invisible" are unused by default.  Users may give explicit tag
  6227 specifications in the text, e.g. ''by %invisible (auto)''.  The
  6228 interpretation of tags is determined by the LaTeX job during document
  6229 preparation: see option -V of isatool usedir, or options -n and -t of
  6230 isatool document, or even the LaTeX macros \isakeeptag, \isafoldtag,
  6231 \isadroptag.
  6232 
  6233 Several document versions may be produced at the same time via isatool
  6234 usedir (the generated index.html will link all of them).  Typical
  6235 specifications include ''-V document=theory,proof,ML'' to present
  6236 theory/proof/ML parts faithfully, ''-V outline=/proof,/ML'' to fold
  6237 proof and ML command