NEWS
author blanchet
Fri Jul 20 22:19:46 2012 +0200 (2012-07-20)
changeset 48402 327ebf1c42a8
parent 48371 3a5a5a992519
child 48428 ffa0618cc4d4
permissions -rw-r--r--
added MaSh to news
     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 * Updated and extended "isar-ref" manual, reduced remaining material
    13 in old "ref" manual.
    14 
    15 
    16 *** Pure ***
    17 
    18 * Command "export_code": relative file names are interpreted
    19 relatively to master directory of current theory rather than
    20 the rather arbitrary current working directory.
    21 INCOMPATIBILITY.
    22 
    23 * Discontinued obsolete attribute "COMP".  Potential INCOMPATIBILITY,
    24 use regular rule composition via "OF" / "THEN", or explicit proof
    25 structure instead.  Note that Isabelle/ML provides a variety of
    26 operators like COMP, INCR_COMP, COMP_INCR, which need to be applied
    27 with some care where this is really required.
    28 
    29 
    30 *** HOL ***
    31 
    32 * Simproc "finite_Collect" rewrites set comprehensions into pointfree
    33 expressions.
    34 
    35 * Quickcheck:
    36 
    37   - added an optimisation for equality premises.
    38     It is switched on by default, and can be switched off by setting
    39     the configuration quickcheck_optimise_equality to false.    
    40 
    41 * The SMT solver Z3 has now by default a restricted set of directly
    42 supported features. For the full set of features (div/mod, nonlinear
    43 arithmetic, datatypes/records) with potential proof reconstruction
    44 failures, enable the configuration option "z3_with_extensions".
    45 Minor INCOMPATIBILITY.
    46 
    47 * Sledgehammer:
    48 
    49   - Added MaSh relevance filter based on machine-learning; see the
    50     Sledgehammer manual for details.
    51   - Rationalized type encodings ("type_enc" option).
    52   - Renamed options:
    53       max_relevant ~> max_facts
    54       relevance_thresholds ~> fact_thresholds
    55 
    56 
    57 *** Document preparation ***
    58 
    59 * Default for \<euro> is now based on eurosym package, instead of
    60 slightly exotic babel/greek.
    61 
    62 
    63 *** System ***
    64 
    65 * Discontinued support for Poly/ML 5.2.1, which was the last version
    66 without exception positions and advanced ML compiler/toplevel
    67 configuration.
    68 
    69 
    70 
    71 New in Isabelle2012 (May 2012)
    72 ------------------------------
    73 
    74 *** General ***
    75 
    76 * Prover IDE (PIDE) improvements:
    77 
    78   - more robust Sledgehammer integration (as before the sledgehammer
    79     command-line needs to be typed into the source buffer)
    80   - markup for bound variables
    81   - markup for types of term variables (displayed as tooltips)
    82   - support for user-defined Isar commands within the running session
    83   - improved support for Unicode outside original 16bit range
    84     e.g. glyph for \<A> (thanks to jEdit 4.5.1)
    85 
    86 * Forward declaration of outer syntax keywords within the theory
    87 header -- minor INCOMPATIBILITY for user-defined commands.  Allow new
    88 commands to be used in the same theory where defined.
    89 
    90 * Auxiliary contexts indicate block structure for specifications with
    91 additional parameters and assumptions.  Such unnamed contexts may be
    92 nested within other targets, like 'theory', 'locale', 'class',
    93 'instantiation' etc.  Results from the local context are generalized
    94 accordingly and applied to the enclosing target context.  Example:
    95 
    96   context
    97     fixes x y z :: 'a
    98     assumes xy: "x = y" and yz: "y = z"
    99   begin
   100 
   101   lemma my_trans: "x = z" using xy yz by simp
   102 
   103   end
   104 
   105   thm my_trans
   106 
   107 The most basic application is to factor-out context elements of
   108 several fixes/assumes/shows theorem statements, e.g. see
   109 ~~/src/HOL/Isar_Examples/Group_Context.thy
   110 
   111 Any other local theory specification element works within the "context
   112 ... begin ... end" block as well.
   113 
   114 * Bundled declarations associate attributed fact expressions with a
   115 given name in the context.  These may be later included in other
   116 contexts.  This allows to manage context extensions casually, without
   117 the logical dependencies of locales and locale interpretation.  See
   118 commands 'bundle', 'include', 'including' etc. in the isar-ref manual.
   119 
   120 * Commands 'lemmas' and 'theorems' allow local variables using 'for'
   121 declaration, and results are standardized before being stored.  Thus
   122 old-style "standard" after instantiation or composition of facts
   123 becomes obsolete.  Minor INCOMPATIBILITY, due to potential change of
   124 indices of schematic variables.
   125 
   126 * Rule attributes in local theory declarations (e.g. locale or class)
   127 are now statically evaluated: the resulting theorem is stored instead
   128 of the original expression.  INCOMPATIBILITY in rare situations, where
   129 the historic accident of dynamic re-evaluation in interpretations
   130 etc. was exploited.
   131 
   132 * New tutorial "Programming and Proving in Isabelle/HOL"
   133 ("prog-prove").  It completely supersedes "A Tutorial Introduction to
   134 Structured Isar Proofs" ("isar-overview"), which has been removed.  It
   135 also supersedes "Isabelle/HOL, A Proof Assistant for Higher-Order
   136 Logic" as the recommended beginners tutorial, but does not cover all
   137 of the material of that old tutorial.
   138 
   139 * Updated and extended reference manuals: "isar-ref",
   140 "implementation", "system"; reduced remaining material in old "ref"
   141 manual.
   142 
   143 
   144 *** Pure ***
   145 
   146 * Command 'definition' no longer exports the foundational "raw_def"
   147 into the user context.  Minor INCOMPATIBILITY, may use the regular
   148 "def" result with attribute "abs_def" to imitate the old version.
   149 
   150 * Attribute "abs_def" turns an equation of the form "f x y == t" into
   151 "f == %x y. t", which ensures that "simp" or "unfold" steps always
   152 expand it.  This also works for object-logic equality.  (Formerly
   153 undocumented feature.)
   154 
   155 * Sort constraints are now propagated in simultaneous statements, just
   156 like type constraints.  INCOMPATIBILITY in rare situations, where
   157 distinct sorts used to be assigned accidentally.  For example:
   158 
   159   lemma "P (x::'a::foo)" and "Q (y::'a::bar)"  -- "now illegal"
   160 
   161   lemma "P (x::'a)" and "Q (y::'a::bar)"
   162     -- "now uniform 'a::bar instead of default sort for first occurrence (!)"
   163 
   164 * Rule composition via attribute "OF" (or ML functions OF/MRS) is more
   165 tolerant against multiple unifiers, as long as the final result is
   166 unique.  (As before, rules are composed in canonical right-to-left
   167 order to accommodate newly introduced premises.)
   168 
   169 * Renamed some inner syntax categories:
   170 
   171     num ~> num_token
   172     xnum ~> xnum_token
   173     xstr ~> str_token
   174 
   175 Minor INCOMPATIBILITY.  Note that in practice "num_const" or
   176 "num_position" etc. are mainly used instead (which also include
   177 position information via constraints).
   178 
   179 * Simplified configuration options for syntax ambiguity: see
   180 "syntax_ambiguity_warning" and "syntax_ambiguity_limit" in isar-ref
   181 manual.  Minor INCOMPATIBILITY.
   182 
   183 * Discontinued configuration option "syntax_positions": atomic terms
   184 in parse trees are always annotated by position constraints.
   185 
   186 * Old code generator for SML and its commands 'code_module',
   187 'code_library', 'consts_code', 'types_code' have been discontinued.
   188 Use commands of the generic code generator instead.  INCOMPATIBILITY.
   189 
   190 * Redundant attribute "code_inline" has been discontinued. Use
   191 "code_unfold" instead.  INCOMPATIBILITY.
   192 
   193 * Dropped attribute "code_unfold_post" in favor of the its dual
   194 "code_abbrev", which yields a common pattern in definitions like
   195 
   196   definition [code_abbrev]: "f = t"
   197 
   198 INCOMPATIBILITY.
   199 
   200 * Obsolete 'types' command has been discontinued.  Use 'type_synonym'
   201 instead.  INCOMPATIBILITY.
   202 
   203 * Discontinued old "prems" fact, which used to refer to the accidental
   204 collection of foundational premises in the context (already marked as
   205 legacy since Isabelle2011).
   206 
   207 
   208 *** HOL ***
   209 
   210 * Type 'a set is now a proper type constructor (just as before
   211 Isabelle2008).  Definitions mem_def and Collect_def have disappeared.
   212 Non-trivial INCOMPATIBILITY.  For developments keeping predicates and
   213 sets separate, it is often sufficient to rephrase some set S that has
   214 been accidentally used as predicates by "%x. x : S", and some
   215 predicate P that has been accidentally used as set by "{x. P x}".
   216 Corresponding proofs in a first step should be pruned from any
   217 tinkering with former theorems mem_def and Collect_def as far as
   218 possible.
   219 
   220 For developments which deliberately mix predicates and sets, a
   221 planning step is necessary to determine what should become a predicate
   222 and what a set.  It can be helpful to carry out that step in
   223 Isabelle2011-1 before jumping right into the current release.
   224 
   225 * Code generation by default implements sets as container type rather
   226 than predicates.  INCOMPATIBILITY.
   227 
   228 * New type synonym 'a rel = ('a * 'a) set
   229 
   230 * The representation of numerals has changed.  Datatype "num"
   231 represents strictly positive binary numerals, along with functions
   232 "numeral :: num => 'a" and "neg_numeral :: num => 'a" to represent
   233 positive and negated numeric literals, respectively.  See also
   234 definitions in ~~/src/HOL/Num.thy.  Potential INCOMPATIBILITY, some
   235 user theories may require adaptations as follows:
   236 
   237   - Theorems with number_ring or number_semiring constraints: These
   238     classes are gone; use comm_ring_1 or comm_semiring_1 instead.
   239 
   240   - Theories defining numeric types: Remove number, number_semiring,
   241     and number_ring instances. Defer all theorems about numerals until
   242     after classes one and semigroup_add have been instantiated.
   243 
   244   - Numeral-only simp rules: Replace each rule having a "number_of v"
   245     pattern with two copies, one for numeral and one for neg_numeral.
   246 
   247   - Theorems about subclasses of semiring_1 or ring_1: These classes
   248     automatically support numerals now, so more simp rules and
   249     simprocs may now apply within the proof.
   250 
   251   - Definitions and theorems using old constructors Pls/Min/Bit0/Bit1:
   252     Redefine using other integer operations.
   253 
   254 * Transfer: New package intended to generalize the existing
   255 "descending" method and related theorem attributes from the Quotient
   256 package.  (Not all functionality is implemented yet, but future
   257 development will focus on Transfer as an eventual replacement for the
   258 corresponding parts of the Quotient package.)
   259 
   260   - transfer_rule attribute: Maintains a collection of transfer rules,
   261     which relate constants at two different types. Transfer rules may
   262     relate different type instances of the same polymorphic constant,
   263     or they may relate an operation on a raw type to a corresponding
   264     operation on an abstract type (quotient or subtype). For example:
   265 
   266     ((A ===> B) ===> list_all2 A ===> list_all2 B) map map
   267     (cr_int ===> cr_int ===> cr_int) (%(x,y) (u,v). (x+u, y+v)) plus_int
   268 
   269   - transfer method: Replaces a subgoal on abstract types with an
   270     equivalent subgoal on the corresponding raw types. Constants are
   271     replaced with corresponding ones according to the transfer rules.
   272     Goals are generalized over all free variables by default; this is
   273     necessary for variables whose types change, but can be overridden
   274     for specific variables with e.g. "transfer fixing: x y z".  The
   275     variant transfer' method allows replacing a subgoal with one that
   276     is logically stronger (rather than equivalent).
   277 
   278   - relator_eq attribute: Collects identity laws for relators of
   279     various type constructors, e.g. "list_all2 (op =) = (op =)".  The
   280     transfer method uses these lemmas to infer transfer rules for
   281     non-polymorphic constants on the fly.
   282 
   283   - transfer_prover method: Assists with proving a transfer rule for a
   284     new constant, provided the constant is defined in terms of other
   285     constants that already have transfer rules. It should be applied
   286     after unfolding the constant definitions.
   287 
   288   - HOL/ex/Transfer_Int_Nat.thy: Example theory demonstrating transfer
   289     from type nat to type int.
   290 
   291 * Lifting: New package intended to generalize the quotient_definition
   292 facility of the Quotient package; designed to work with Transfer.
   293 
   294   - lift_definition command: Defines operations on an abstract type in
   295     terms of a corresponding operation on a representation
   296     type.  Example syntax:
   297 
   298     lift_definition dlist_insert :: "'a => 'a dlist => 'a dlist"
   299       is List.insert
   300 
   301     Users must discharge a respectfulness proof obligation when each
   302     constant is defined. (For a type copy, i.e. a typedef with UNIV,
   303     the proof is discharged automatically.) The obligation is
   304     presented in a user-friendly, readable form; a respectfulness
   305     theorem in the standard format and a transfer rule are generated
   306     by the package.
   307 
   308   - Integration with code_abstype: For typedefs (e.g. subtypes
   309     corresponding to a datatype invariant, such as dlist),
   310     lift_definition generates a code certificate theorem and sets up
   311     code generation for each constant.
   312 
   313   - setup_lifting command: Sets up the Lifting package to work with a
   314     user-defined type. The user must provide either a quotient theorem
   315     or a type_definition theorem.  The package configures transfer
   316     rules for equality and quantifiers on the type, and sets up the
   317     lift_definition command to work with the type.
   318 
   319   - Usage examples: See Quotient_Examples/Lift_DList.thy,
   320     Quotient_Examples/Lift_RBT.thy, Quotient_Examples/Lift_FSet.thy,
   321     Word/Word.thy and Library/Float.thy.
   322 
   323 * Quotient package:
   324 
   325   - The 'quotient_type' command now supports a 'morphisms' option with
   326     rep and abs functions, similar to typedef.
   327 
   328   - 'quotient_type' sets up new types to work with the Lifting and
   329     Transfer packages, as with 'setup_lifting'.
   330 
   331   - The 'quotient_definition' command now requires the user to prove a
   332     respectfulness property at the point where the constant is
   333     defined, similar to lift_definition; INCOMPATIBILITY.
   334 
   335   - Renamed predicate 'Quotient' to 'Quotient3', and renamed theorems
   336     accordingly, INCOMPATIBILITY.
   337 
   338 * New diagnostic command 'find_unused_assms' to find potentially
   339 superfluous assumptions in theorems using Quickcheck.
   340 
   341 * Quickcheck:
   342 
   343   - Quickcheck returns variable assignments as counterexamples, which
   344     allows to reveal the underspecification of functions under test.
   345     For example, refuting "hd xs = x", it presents the variable
   346     assignment xs = [] and x = a1 as a counterexample, assuming that
   347     any property is false whenever "hd []" occurs in it.
   348 
   349     These counterexample are marked as potentially spurious, as
   350     Quickcheck also returns "xs = []" as a counterexample to the
   351     obvious theorem "hd xs = hd xs".
   352 
   353     After finding a potentially spurious counterexample, Quickcheck
   354     continues searching for genuine ones.
   355 
   356     By default, Quickcheck shows potentially spurious and genuine
   357     counterexamples. The option "genuine_only" sets quickcheck to only
   358     show genuine counterexamples.
   359 
   360   - The command 'quickcheck_generator' creates random and exhaustive
   361     value generators for a given type and operations.
   362 
   363     It generates values by using the operations as if they were
   364     constructors of that type.
   365 
   366   - Support for multisets.
   367 
   368   - Added "use_subtype" options.
   369 
   370   - Added "quickcheck_locale" configuration to specify how to process
   371     conjectures in a locale context.
   372 
   373 * Nitpick: Fixed infinite loop caused by the 'peephole_optim' option
   374 and affecting 'rat' and 'real'.
   375 
   376 * Sledgehammer:
   377   - Integrated more tightly with SPASS, as described in the ITP 2012
   378     paper "More SPASS with Isabelle".
   379   - Made it try "smt" as a fallback if "metis" fails or times out.
   380   - Added support for the following provers: Alt-Ergo (via Why3 and
   381     TFF1), iProver, iProver-Eq.
   382   - Sped up the minimizer.
   383   - Added "lam_trans", "uncurry_aliases", and "minimize" options.
   384   - Renamed "slicing" ("no_slicing") option to "slice" ("dont_slice").
   385   - Renamed "sound" option to "strict".
   386 
   387 * Metis: Added possibility to specify lambda translations scheme as a
   388 parenthesized argument (e.g., "by (metis (lifting) ...)").
   389 
   390 * SMT: Renamed "smt_fixed" option to "smt_read_only_certificates".
   391 
   392 * Command 'try0': Renamed from 'try_methods'. INCOMPATIBILITY.
   393 
   394 * New "case_product" attribute to generate a case rule doing multiple
   395 case distinctions at the same time.  E.g.
   396 
   397   list.exhaust [case_product nat.exhaust]
   398 
   399 produces a rule which can be used to perform case distinction on both
   400 a list and a nat.
   401 
   402 * New "eventually_elim" method as a generalized variant of the
   403 eventually_elim* rules.  Supports structured proofs.
   404 
   405 * Typedef with implicit set definition is considered legacy.  Use
   406 "typedef (open)" form instead, which will eventually become the
   407 default.
   408 
   409 * Record: code generation can be switched off manually with
   410 
   411   declare [[record_coden = false]]  -- "default true"
   412 
   413 * Datatype: type parameters allow explicit sort constraints.
   414 
   415 * Concrete syntax for case expressions includes constraints for source
   416 positions, and thus produces Prover IDE markup for its bindings.
   417 INCOMPATIBILITY for old-style syntax translations that augment the
   418 pattern notation; e.g. see src/HOL/HOLCF/One.thy for translations of
   419 one_case.
   420 
   421 * Clarified attribute "mono_set": pure declaration without modifying
   422 the result of the fact expression.
   423 
   424 * More default pred/set conversions on a couple of relation operations
   425 and predicates.  Added powers of predicate relations.  Consolidation
   426 of some relation theorems:
   427 
   428   converse_def ~> converse_unfold
   429   rel_comp_def ~> relcomp_unfold
   430   symp_def ~> (modified, use symp_def and sym_def instead)
   431   transp_def ~> transp_trans
   432   Domain_def ~> Domain_unfold
   433   Range_def ~> Domain_converse [symmetric]
   434 
   435 Generalized theorems INF_INT_eq, INF_INT_eq2, SUP_UN_eq, SUP_UN_eq2.
   436 
   437 See theory "Relation" for examples for making use of pred/set
   438 conversions by means of attributes "to_set" and "to_pred".
   439 
   440 INCOMPATIBILITY.
   441 
   442 * Renamed facts about the power operation on relations, i.e., relpow
   443 to match the constant's name:
   444 
   445   rel_pow_1 ~> relpow_1
   446   rel_pow_0_I ~> relpow_0_I
   447   rel_pow_Suc_I ~> relpow_Suc_I
   448   rel_pow_Suc_I2 ~> relpow_Suc_I2
   449   rel_pow_0_E ~> relpow_0_E
   450   rel_pow_Suc_E ~> relpow_Suc_E
   451   rel_pow_E ~> relpow_E
   452   rel_pow_Suc_D2 ~> relpow_Suc_D2
   453   rel_pow_Suc_E2 ~> relpow_Suc_E2
   454   rel_pow_Suc_D2' ~> relpow_Suc_D2'
   455   rel_pow_E2 ~> relpow_E2
   456   rel_pow_add ~> relpow_add
   457   rel_pow_commute ~> relpow
   458   rel_pow_empty ~> relpow_empty:
   459   rtrancl_imp_UN_rel_pow ~> rtrancl_imp_UN_relpow
   460   rel_pow_imp_rtrancl ~> relpow_imp_rtrancl
   461   rtrancl_is_UN_rel_pow ~> rtrancl_is_UN_relpow
   462   rtrancl_imp_rel_pow ~> rtrancl_imp_relpow
   463   rel_pow_fun_conv ~> relpow_fun_conv
   464   rel_pow_finite_bounded1 ~> relpow_finite_bounded1
   465   rel_pow_finite_bounded ~> relpow_finite_bounded
   466   rtrancl_finite_eq_rel_pow ~> rtrancl_finite_eq_relpow
   467   trancl_finite_eq_rel_pow ~> trancl_finite_eq_relpow
   468   single_valued_rel_pow ~> single_valued_relpow
   469 
   470 INCOMPATIBILITY.
   471 
   472 * Theory Relation: Consolidated constant name for relation composition
   473 and corresponding theorem names:
   474 
   475   - Renamed constant rel_comp to relcomp.
   476 
   477   - Dropped abbreviation pred_comp. Use relcompp instead.
   478 
   479   - Renamed theorems:
   480 
   481     rel_compI ~> relcompI
   482     rel_compEpair ~> relcompEpair
   483     rel_compE ~> relcompE
   484     pred_comp_rel_comp_eq ~> relcompp_relcomp_eq
   485     rel_comp_empty1 ~> relcomp_empty1
   486     rel_comp_mono ~> relcomp_mono
   487     rel_comp_subset_Sigma ~> relcomp_subset_Sigma
   488     rel_comp_distrib ~> relcomp_distrib
   489     rel_comp_distrib2 ~> relcomp_distrib2
   490     rel_comp_UNION_distrib ~> relcomp_UNION_distrib
   491     rel_comp_UNION_distrib2 ~> relcomp_UNION_distrib2
   492     single_valued_rel_comp ~> single_valued_relcomp
   493     rel_comp_def ~> relcomp_unfold
   494     converse_rel_comp ~> converse_relcomp
   495     pred_compI ~> relcomppI
   496     pred_compE ~> relcomppE
   497     pred_comp_bot1 ~> relcompp_bot1
   498     pred_comp_bot2 ~> relcompp_bot2
   499     transp_pred_comp_less_eq ~> transp_relcompp_less_eq
   500     pred_comp_mono ~> relcompp_mono
   501     pred_comp_distrib ~> relcompp_distrib
   502     pred_comp_distrib2 ~> relcompp_distrib2
   503     converse_pred_comp ~> converse_relcompp
   504 
   505     finite_rel_comp ~> finite_relcomp
   506 
   507     set_rel_comp ~> set_relcomp
   508 
   509 INCOMPATIBILITY.
   510 
   511 * Theory Divides: Discontinued redundant theorems about div and mod.
   512 INCOMPATIBILITY, use the corresponding generic theorems instead.
   513 
   514   DIVISION_BY_ZERO ~> div_by_0, mod_by_0
   515   zdiv_self ~> div_self
   516   zmod_self ~> mod_self
   517   zdiv_zero ~> div_0
   518   zmod_zero ~> mod_0
   519   zdiv_zmod_equality ~> div_mod_equality2
   520   zdiv_zmod_equality2 ~> div_mod_equality
   521   zmod_zdiv_trivial ~> mod_div_trivial
   522   zdiv_zminus_zminus ~> div_minus_minus
   523   zmod_zminus_zminus ~> mod_minus_minus
   524   zdiv_zminus2 ~> div_minus_right
   525   zmod_zminus2 ~> mod_minus_right
   526   zdiv_minus1_right ~> div_minus1_right
   527   zmod_minus1_right ~> mod_minus1_right
   528   zdvd_mult_div_cancel ~> dvd_mult_div_cancel
   529   zmod_zmult1_eq ~> mod_mult_right_eq
   530   zpower_zmod ~> power_mod
   531   zdvd_zmod ~> dvd_mod
   532   zdvd_zmod_imp_zdvd ~> dvd_mod_imp_dvd
   533   mod_mult_distrib ~> mult_mod_left
   534   mod_mult_distrib2 ~> mult_mod_right
   535 
   536 * Removed redundant theorems nat_mult_2 and nat_mult_2_right; use
   537 generic mult_2 and mult_2_right instead. INCOMPATIBILITY.
   538 
   539 * Finite_Set.fold now qualified.  INCOMPATIBILITY.
   540 
   541 * Consolidated theorem names concerning fold combinators:
   542 
   543   inf_INFI_fold_inf ~> inf_INF_fold_inf
   544   sup_SUPR_fold_sup ~> sup_SUP_fold_sup
   545   INFI_fold_inf ~> INF_fold_inf
   546   SUPR_fold_sup ~> SUP_fold_sup
   547   union_set ~> union_set_fold
   548   minus_set ~> minus_set_fold
   549   INFI_set_fold ~> INF_set_fold
   550   SUPR_set_fold ~> SUP_set_fold
   551   INF_code ~> INF_set_foldr
   552   SUP_code ~> SUP_set_foldr
   553   foldr.simps ~> foldr.simps (in point-free formulation)
   554   foldr_fold_rev ~> foldr_conv_fold
   555   foldl_fold ~> foldl_conv_fold
   556   foldr_foldr ~> foldr_conv_foldl
   557   foldl_foldr ~> foldl_conv_foldr
   558   fold_set_remdups ~> fold_set_fold_remdups
   559   fold_set ~> fold_set_fold
   560   fold1_set ~> fold1_set_fold
   561 
   562 INCOMPATIBILITY.
   563 
   564 * Dropped rarely useful theorems concerning fold combinators:
   565 foldl_apply, foldl_fun_comm, foldl_rev, fold_weak_invariant,
   566 rev_foldl_cons, fold_set_remdups, fold_set, fold_set1,
   567 concat_conv_foldl, foldl_weak_invariant, foldl_invariant,
   568 foldr_invariant, foldl_absorb0, foldl_foldr1_lemma, foldl_foldr1,
   569 listsum_conv_fold, listsum_foldl, sort_foldl_insort, foldl_assoc,
   570 foldr_conv_foldl, start_le_sum, elem_le_sum, sum_eq_0_conv.
   571 INCOMPATIBILITY.  For the common phrases "%xs. List.foldr plus xs 0"
   572 and "List.foldl plus 0", prefer "List.listsum".  Otherwise it can be
   573 useful to boil down "List.foldr" and "List.foldl" to "List.fold" by
   574 unfolding "foldr_conv_fold" and "foldl_conv_fold".
   575 
   576 * Dropped lemmas minus_set_foldr, union_set_foldr, union_coset_foldr,
   577 inter_coset_foldr, Inf_fin_set_foldr, Sup_fin_set_foldr,
   578 Min_fin_set_foldr, Max_fin_set_foldr, Inf_set_foldr, Sup_set_foldr,
   579 INF_set_foldr, SUP_set_foldr.  INCOMPATIBILITY.  Prefer corresponding
   580 lemmas over fold rather than foldr, or make use of lemmas
   581 fold_conv_foldr and fold_rev.
   582 
   583 * Congruence rules Option.map_cong and Option.bind_cong for recursion
   584 through option types.
   585 
   586 * "Transitive_Closure.ntrancl": bounded transitive closure on
   587 relations.
   588 
   589 * Constant "Set.not_member" now qualified.  INCOMPATIBILITY.
   590 
   591 * Theory Int: Discontinued many legacy theorems specific to type int.
   592 INCOMPATIBILITY, use the corresponding generic theorems instead.
   593 
   594   zminus_zminus ~> minus_minus
   595   zminus_0 ~> minus_zero
   596   zminus_zadd_distrib ~> minus_add_distrib
   597   zadd_commute ~> add_commute
   598   zadd_assoc ~> add_assoc
   599   zadd_left_commute ~> add_left_commute
   600   zadd_ac ~> add_ac
   601   zmult_ac ~> mult_ac
   602   zadd_0 ~> add_0_left
   603   zadd_0_right ~> add_0_right
   604   zadd_zminus_inverse2 ~> left_minus
   605   zmult_zminus ~> mult_minus_left
   606   zmult_commute ~> mult_commute
   607   zmult_assoc ~> mult_assoc
   608   zadd_zmult_distrib ~> left_distrib
   609   zadd_zmult_distrib2 ~> right_distrib
   610   zdiff_zmult_distrib ~> left_diff_distrib
   611   zdiff_zmult_distrib2 ~> right_diff_distrib
   612   zmult_1 ~> mult_1_left
   613   zmult_1_right ~> mult_1_right
   614   zle_refl ~> order_refl
   615   zle_trans ~> order_trans
   616   zle_antisym ~> order_antisym
   617   zle_linear ~> linorder_linear
   618   zless_linear ~> linorder_less_linear
   619   zadd_left_mono ~> add_left_mono
   620   zadd_strict_right_mono ~> add_strict_right_mono
   621   zadd_zless_mono ~> add_less_le_mono
   622   int_0_less_1 ~> zero_less_one
   623   int_0_neq_1 ~> zero_neq_one
   624   zless_le ~> less_le
   625   zpower_zadd_distrib ~> power_add
   626   zero_less_zpower_abs_iff ~> zero_less_power_abs_iff
   627   zero_le_zpower_abs ~> zero_le_power_abs
   628 
   629 * Theory Deriv: Renamed
   630 
   631   DERIV_nonneg_imp_nonincreasing ~> DERIV_nonneg_imp_nondecreasing
   632 
   633 * Theory Library/Multiset: Improved code generation of multisets.
   634 
   635 * Theory HOL/Library/Set_Algebras: Addition and multiplication on sets
   636 are expressed via type classes again. The special syntax
   637 \<oplus>/\<otimes> has been replaced by plain +/*. Removed constant
   638 setsum_set, which is now subsumed by Big_Operators.setsum.
   639 INCOMPATIBILITY.
   640 
   641 * Theory HOL/Library/Diagonalize has been removed. INCOMPATIBILITY,
   642 use theory HOL/Library/Nat_Bijection instead.
   643 
   644 * Theory HOL/Library/RBT_Impl: Backing implementation of red-black
   645 trees is now inside a type class context.  Names of affected
   646 operations and lemmas have been prefixed by rbt_.  INCOMPATIBILITY for
   647 theories working directly with raw red-black trees, adapt the names as
   648 follows:
   649 
   650   Operations:
   651   bulkload -> rbt_bulkload
   652   del_from_left -> rbt_del_from_left
   653   del_from_right -> rbt_del_from_right
   654   del -> rbt_del
   655   delete -> rbt_delete
   656   ins -> rbt_ins
   657   insert -> rbt_insert
   658   insertw -> rbt_insert_with
   659   insert_with_key -> rbt_insert_with_key
   660   map_entry -> rbt_map_entry
   661   lookup -> rbt_lookup
   662   sorted -> rbt_sorted
   663   tree_greater -> rbt_greater
   664   tree_less -> rbt_less
   665   tree_less_symbol -> rbt_less_symbol
   666   union -> rbt_union
   667   union_with -> rbt_union_with
   668   union_with_key -> rbt_union_with_key
   669 
   670   Lemmas:
   671   balance_left_sorted -> balance_left_rbt_sorted
   672   balance_left_tree_greater -> balance_left_rbt_greater
   673   balance_left_tree_less -> balance_left_rbt_less
   674   balance_right_sorted -> balance_right_rbt_sorted
   675   balance_right_tree_greater -> balance_right_rbt_greater
   676   balance_right_tree_less -> balance_right_rbt_less
   677   balance_sorted -> balance_rbt_sorted
   678   balance_tree_greater -> balance_rbt_greater
   679   balance_tree_less -> balance_rbt_less
   680   bulkload_is_rbt -> rbt_bulkload_is_rbt
   681   combine_sorted -> combine_rbt_sorted
   682   combine_tree_greater -> combine_rbt_greater
   683   combine_tree_less -> combine_rbt_less
   684   delete_in_tree -> rbt_delete_in_tree
   685   delete_is_rbt -> rbt_delete_is_rbt
   686   del_from_left_tree_greater -> rbt_del_from_left_rbt_greater
   687   del_from_left_tree_less -> rbt_del_from_left_rbt_less
   688   del_from_right_tree_greater -> rbt_del_from_right_rbt_greater
   689   del_from_right_tree_less -> rbt_del_from_right_rbt_less
   690   del_in_tree -> rbt_del_in_tree
   691   del_inv1_inv2 -> rbt_del_inv1_inv2
   692   del_sorted -> rbt_del_rbt_sorted
   693   del_tree_greater -> rbt_del_rbt_greater
   694   del_tree_less -> rbt_del_rbt_less
   695   dom_lookup_Branch -> dom_rbt_lookup_Branch
   696   entries_lookup -> entries_rbt_lookup
   697   finite_dom_lookup -> finite_dom_rbt_lookup
   698   insert_sorted -> rbt_insert_rbt_sorted
   699   insertw_is_rbt -> rbt_insertw_is_rbt
   700   insertwk_is_rbt -> rbt_insertwk_is_rbt
   701   insertwk_sorted -> rbt_insertwk_rbt_sorted
   702   insertw_sorted -> rbt_insertw_rbt_sorted
   703   ins_sorted -> ins_rbt_sorted
   704   ins_tree_greater -> ins_rbt_greater
   705   ins_tree_less -> ins_rbt_less
   706   is_rbt_sorted -> is_rbt_rbt_sorted
   707   lookup_balance -> rbt_lookup_balance
   708   lookup_bulkload -> rbt_lookup_rbt_bulkload
   709   lookup_delete -> rbt_lookup_rbt_delete
   710   lookup_Empty -> rbt_lookup_Empty
   711   lookup_from_in_tree -> rbt_lookup_from_in_tree
   712   lookup_in_tree -> rbt_lookup_in_tree
   713   lookup_ins -> rbt_lookup_ins
   714   lookup_insert -> rbt_lookup_rbt_insert
   715   lookup_insertw -> rbt_lookup_rbt_insertw
   716   lookup_insertwk -> rbt_lookup_rbt_insertwk
   717   lookup_keys -> rbt_lookup_keys
   718   lookup_map -> rbt_lookup_map
   719   lookup_map_entry -> rbt_lookup_rbt_map_entry
   720   lookup_tree_greater -> rbt_lookup_rbt_greater
   721   lookup_tree_less -> rbt_lookup_rbt_less
   722   lookup_union -> rbt_lookup_rbt_union
   723   map_entry_color_of -> rbt_map_entry_color_of
   724   map_entry_inv1 -> rbt_map_entry_inv1
   725   map_entry_inv2 -> rbt_map_entry_inv2
   726   map_entry_is_rbt -> rbt_map_entry_is_rbt
   727   map_entry_sorted -> rbt_map_entry_rbt_sorted
   728   map_entry_tree_greater -> rbt_map_entry_rbt_greater
   729   map_entry_tree_less -> rbt_map_entry_rbt_less
   730   map_tree_greater -> map_rbt_greater
   731   map_tree_less -> map_rbt_less
   732   map_sorted -> map_rbt_sorted
   733   paint_sorted -> paint_rbt_sorted
   734   paint_lookup -> paint_rbt_lookup
   735   paint_tree_greater -> paint_rbt_greater
   736   paint_tree_less -> paint_rbt_less
   737   sorted_entries -> rbt_sorted_entries
   738   tree_greater_eq_trans -> rbt_greater_eq_trans
   739   tree_greater_nit -> rbt_greater_nit
   740   tree_greater_prop -> rbt_greater_prop
   741   tree_greater_simps -> rbt_greater_simps
   742   tree_greater_trans -> rbt_greater_trans
   743   tree_less_eq_trans -> rbt_less_eq_trans
   744   tree_less_nit -> rbt_less_nit
   745   tree_less_prop -> rbt_less_prop
   746   tree_less_simps -> rbt_less_simps
   747   tree_less_trans -> rbt_less_trans
   748   tree_ord_props -> rbt_ord_props
   749   union_Branch -> rbt_union_Branch
   750   union_is_rbt -> rbt_union_is_rbt
   751   unionw_is_rbt -> rbt_unionw_is_rbt
   752   unionwk_is_rbt -> rbt_unionwk_is_rbt
   753   unionwk_sorted -> rbt_unionwk_rbt_sorted
   754 
   755 * Theory HOL/Library/Float: Floating point numbers are now defined as
   756 a subset of the real numbers.  All operations are defined using the
   757 lifing-framework and proofs use the transfer method.  INCOMPATIBILITY.
   758 
   759   Changed Operations:
   760   float_abs -> abs
   761   float_nprt -> nprt
   762   float_pprt -> pprt
   763   pow2 -> use powr
   764   round_down -> float_round_down
   765   round_up -> float_round_up
   766   scale -> exponent
   767 
   768   Removed Operations:
   769   ceiling_fl, lb_mult, lb_mod, ub_mult, ub_mod
   770 
   771   Renamed Lemmas:
   772   abs_float_def -> Float.compute_float_abs
   773   bitlen_ge0 -> bitlen_nonneg
   774   bitlen.simps -> Float.compute_bitlen
   775   float_components -> Float_mantissa_exponent
   776   float_divl.simps -> Float.compute_float_divl
   777   float_divr.simps -> Float.compute_float_divr
   778   float_eq_odd -> mult_powr_eq_mult_powr_iff
   779   float_power -> real_of_float_power
   780   lapprox_posrat_def -> Float.compute_lapprox_posrat
   781   lapprox_rat.simps -> Float.compute_lapprox_rat
   782   le_float_def' -> Float.compute_float_le
   783   le_float_def -> less_eq_float.rep_eq
   784   less_float_def' -> Float.compute_float_less
   785   less_float_def -> less_float.rep_eq
   786   normfloat_def -> Float.compute_normfloat
   787   normfloat_imp_odd_or_zero -> mantissa_not_dvd and mantissa_noteq_0
   788   normfloat -> normfloat_def
   789   normfloat_unique -> use normfloat_def
   790   number_of_float_Float -> Float.compute_float_numeral, Float.compute_float_neg_numeral
   791   one_float_def -> Float.compute_float_one
   792   plus_float_def -> Float.compute_float_plus
   793   rapprox_posrat_def -> Float.compute_rapprox_posrat
   794   rapprox_rat.simps -> Float.compute_rapprox_rat
   795   real_of_float_0 -> zero_float.rep_eq
   796   real_of_float_1 -> one_float.rep_eq
   797   real_of_float_abs -> abs_float.rep_eq
   798   real_of_float_add -> plus_float.rep_eq
   799   real_of_float_minus -> uminus_float.rep_eq
   800   real_of_float_mult -> times_float.rep_eq
   801   real_of_float_simp -> Float.rep_eq
   802   real_of_float_sub -> minus_float.rep_eq
   803   round_down.simps -> Float.compute_float_round_down
   804   round_up.simps -> Float.compute_float_round_up
   805   times_float_def -> Float.compute_float_times
   806   uminus_float_def -> Float.compute_float_uminus
   807   zero_float_def -> Float.compute_float_zero
   808 
   809   Lemmas not necessary anymore, use the transfer method:
   810   bitlen_B0, bitlen_B1, bitlen_ge1, bitlen_Min, bitlen_Pls, float_divl,
   811   float_divr, float_le_simp, float_less1_mantissa_bound,
   812   float_less_simp, float_less_zero, float_le_zero,
   813   float_pos_less1_e_neg, float_pos_m_pos, float_split, float_split2,
   814   floor_pos_exp, lapprox_posrat, lapprox_posrat_bottom, lapprox_rat,
   815   lapprox_rat_bottom, normalized_float, rapprox_posrat,
   816   rapprox_posrat_le1, rapprox_rat, real_of_float_ge0_exp,
   817   real_of_float_neg_exp, real_of_float_nge0_exp, round_down floor_fl,
   818   round_up, zero_le_float, zero_less_float
   819 
   820 * New theory HOL/Library/DAList provides an abstract type for
   821 association lists with distinct keys.
   822 
   823 * Session HOL/IMP: Added new theory of abstract interpretation of
   824 annotated commands.
   825 
   826 * Session HOL-Import: Re-implementation from scratch is faster,
   827 simpler, and more scalable.  Requires a proof bundle, which is
   828 available as an external component.  Discontinued old (and mostly
   829 dead) Importer for HOL4 and HOL Light.  INCOMPATIBILITY.
   830 
   831 * Session HOL-Word: Discontinued many redundant theorems specific to
   832 type 'a word. INCOMPATIBILITY, use the corresponding generic theorems
   833 instead.
   834 
   835   word_sub_alt ~> word_sub_wi
   836   word_add_alt ~> word_add_def
   837   word_mult_alt ~> word_mult_def
   838   word_minus_alt ~> word_minus_def
   839   word_0_alt ~> word_0_wi
   840   word_1_alt ~> word_1_wi
   841   word_add_0 ~> add_0_left
   842   word_add_0_right ~> add_0_right
   843   word_mult_1 ~> mult_1_left
   844   word_mult_1_right ~> mult_1_right
   845   word_add_commute ~> add_commute
   846   word_add_assoc ~> add_assoc
   847   word_add_left_commute ~> add_left_commute
   848   word_mult_commute ~> mult_commute
   849   word_mult_assoc ~> mult_assoc
   850   word_mult_left_commute ~> mult_left_commute
   851   word_left_distrib ~> left_distrib
   852   word_right_distrib ~> right_distrib
   853   word_left_minus ~> left_minus
   854   word_diff_0_right ~> diff_0_right
   855   word_diff_self ~> diff_self
   856   word_sub_def ~> diff_minus
   857   word_diff_minus ~> diff_minus
   858   word_add_ac ~> add_ac
   859   word_mult_ac ~> mult_ac
   860   word_plus_ac0 ~> add_0_left add_0_right add_ac
   861   word_times_ac1 ~> mult_1_left mult_1_right mult_ac
   862   word_order_trans ~> order_trans
   863   word_order_refl ~> order_refl
   864   word_order_antisym ~> order_antisym
   865   word_order_linear ~> linorder_linear
   866   lenw1_zero_neq_one ~> zero_neq_one
   867   word_number_of_eq ~> number_of_eq
   868   word_of_int_add_hom ~> wi_hom_add
   869   word_of_int_sub_hom ~> wi_hom_sub
   870   word_of_int_mult_hom ~> wi_hom_mult
   871   word_of_int_minus_hom ~> wi_hom_neg
   872   word_of_int_succ_hom ~> wi_hom_succ
   873   word_of_int_pred_hom ~> wi_hom_pred
   874   word_of_int_0_hom ~> word_0_wi
   875   word_of_int_1_hom ~> word_1_wi
   876 
   877 * Session HOL-Word: New proof method "word_bitwise" for splitting
   878 machine word equalities and inequalities into logical circuits,
   879 defined in HOL/Word/WordBitwise.thy.  Supports addition, subtraction,
   880 multiplication, shifting by constants, bitwise operators and numeric
   881 constants.  Requires fixed-length word types, not 'a word.  Solves
   882 many standard word identities outright and converts more into first
   883 order problems amenable to blast or similar.  See also examples in
   884 HOL/Word/Examples/WordExamples.thy.
   885 
   886 * Session HOL-Probability: Introduced the type "'a measure" to
   887 represent measures, this replaces the records 'a algebra and 'a
   888 measure_space.  The locales based on subset_class now have two
   889 locale-parameters the space \<Omega> and the set of measurable sets M.
   890 The product of probability spaces uses now the same constant as the
   891 finite product of sigma-finite measure spaces "PiM :: ('i => 'a)
   892 measure".  Most constants are defined now outside of locales and gain
   893 an additional parameter, like null_sets, almost_eventually or \<mu>'.
   894 Measure space constructions for distributions and densities now got
   895 their own constants distr and density.  Instead of using locales to
   896 describe measure spaces with a finite space, the measure count_space
   897 and point_measure is introduced.  INCOMPATIBILITY.
   898 
   899   Renamed constants:
   900   measure -> emeasure
   901   finite_measure.\<mu>' -> measure
   902   product_algebra_generator -> prod_algebra
   903   product_prob_space.emb -> prod_emb
   904   product_prob_space.infprod_algebra -> PiM
   905 
   906   Removed locales:
   907   completeable_measure_space
   908   finite_measure_space
   909   finite_prob_space
   910   finite_product_finite_prob_space
   911   finite_product_sigma_algebra
   912   finite_sigma_algebra
   913   measure_space
   914   pair_finite_prob_space
   915   pair_finite_sigma_algebra
   916   pair_finite_space
   917   pair_sigma_algebra
   918   product_sigma_algebra
   919 
   920   Removed constants:
   921   conditional_space
   922   distribution -> use distr measure, or distributed predicate
   923   image_space
   924   joint_distribution -> use distr measure, or distributed predicate
   925   pair_measure_generator
   926   product_prob_space.infprod_algebra -> use PiM
   927   subvimage
   928 
   929   Replacement theorems:
   930   finite_additivity_sufficient -> ring_of_sets.countably_additiveI_finite
   931   finite_measure.empty_measure -> measure_empty
   932   finite_measure.finite_continuity_from_above -> finite_measure.finite_Lim_measure_decseq
   933   finite_measure.finite_continuity_from_below -> finite_measure.finite_Lim_measure_incseq
   934   finite_measure.finite_measure_countably_subadditive -> finite_measure.finite_measure_subadditive_countably
   935   finite_measure.finite_measure_eq -> finite_measure.emeasure_eq_measure
   936   finite_measure.finite_measure -> finite_measure.emeasure_finite
   937   finite_measure.finite_measure_finite_singleton -> finite_measure.finite_measure_eq_setsum_singleton
   938   finite_measure.positive_measure' -> measure_nonneg
   939   finite_measure.real_measure -> finite_measure.emeasure_real
   940   finite_product_prob_space.finite_measure_times -> finite_product_prob_space.finite_measure_PiM_emb
   941   finite_product_sigma_algebra.in_P -> sets_PiM_I_finite
   942   finite_product_sigma_algebra.P_empty -> space_PiM_empty, sets_PiM_empty
   943   information_space.conditional_entropy_eq -> information_space.conditional_entropy_simple_distributed
   944   information_space.conditional_entropy_positive -> information_space.conditional_entropy_nonneg_simple
   945   information_space.conditional_mutual_information_eq_mutual_information -> information_space.conditional_mutual_information_eq_mutual_information_simple
   946   information_space.conditional_mutual_information_generic_positive -> information_space.conditional_mutual_information_nonneg_simple
   947   information_space.conditional_mutual_information_positive -> information_space.conditional_mutual_information_nonneg_simple
   948   information_space.entropy_commute -> information_space.entropy_commute_simple
   949   information_space.entropy_eq -> information_space.entropy_simple_distributed
   950   information_space.entropy_generic_eq -> information_space.entropy_simple_distributed
   951   information_space.entropy_positive -> information_space.entropy_nonneg_simple
   952   information_space.entropy_uniform_max -> information_space.entropy_uniform
   953   information_space.KL_eq_0_imp -> information_space.KL_eq_0_iff_eq
   954   information_space.KL_eq_0 -> information_space.KL_same_eq_0
   955   information_space.KL_ge_0 -> information_space.KL_nonneg
   956   information_space.mutual_information_eq -> information_space.mutual_information_simple_distributed
   957   information_space.mutual_information_positive -> information_space.mutual_information_nonneg_simple
   958   Int_stable_cuboids -> Int_stable_atLeastAtMost
   959   Int_stable_product_algebra_generator -> positive_integral
   960   measure_preserving -> equality "distr M N f = N" "f : measurable M N"
   961   measure_space.additive -> emeasure_additive
   962   measure_space.AE_iff_null_set -> AE_iff_null
   963   measure_space.almost_everywhere_def -> eventually_ae_filter
   964   measure_space.almost_everywhere_vimage -> AE_distrD
   965   measure_space.continuity_from_above -> INF_emeasure_decseq
   966   measure_space.continuity_from_above_Lim -> Lim_emeasure_decseq
   967   measure_space.continuity_from_below_Lim -> Lim_emeasure_incseq
   968   measure_space.continuity_from_below -> SUP_emeasure_incseq
   969   measure_space_density -> emeasure_density
   970   measure_space.density_is_absolutely_continuous -> absolutely_continuousI_density
   971   measure_space.integrable_vimage -> integrable_distr
   972   measure_space.integral_translated_density -> integral_density
   973   measure_space.integral_vimage -> integral_distr
   974   measure_space.measure_additive -> plus_emeasure
   975   measure_space.measure_compl -> emeasure_compl
   976   measure_space.measure_countable_increasing -> emeasure_countable_increasing
   977   measure_space.measure_countably_subadditive -> emeasure_subadditive_countably
   978   measure_space.measure_decseq -> decseq_emeasure
   979   measure_space.measure_Diff -> emeasure_Diff
   980   measure_space.measure_Diff_null_set -> emeasure_Diff_null_set
   981   measure_space.measure_eq_0 -> emeasure_eq_0
   982   measure_space.measure_finitely_subadditive -> emeasure_subadditive_finite
   983   measure_space.measure_finite_singleton -> emeasure_eq_setsum_singleton
   984   measure_space.measure_incseq -> incseq_emeasure
   985   measure_space.measure_insert -> emeasure_insert
   986   measure_space.measure_mono -> emeasure_mono
   987   measure_space.measure_not_negative -> emeasure_not_MInf
   988   measure_space.measure_preserving_Int_stable -> measure_eqI_generator_eq
   989   measure_space.measure_setsum -> setsum_emeasure
   990   measure_space.measure_setsum_split -> setsum_emeasure_cover
   991   measure_space.measure_space_vimage -> emeasure_distr
   992   measure_space.measure_subadditive_finite -> emeasure_subadditive_finite
   993   measure_space.measure_subadditive -> subadditive
   994   measure_space.measure_top -> emeasure_space
   995   measure_space.measure_UN_eq_0 -> emeasure_UN_eq_0
   996   measure_space.measure_Un_null_set -> emeasure_Un_null_set
   997   measure_space.positive_integral_translated_density -> positive_integral_density
   998   measure_space.positive_integral_vimage -> positive_integral_distr
   999   measure_space.real_continuity_from_above -> Lim_measure_decseq
  1000   measure_space.real_continuity_from_below -> Lim_measure_incseq
  1001   measure_space.real_measure_countably_subadditive -> measure_subadditive_countably
  1002   measure_space.real_measure_Diff -> measure_Diff
  1003   measure_space.real_measure_finite_Union -> measure_finite_Union
  1004   measure_space.real_measure_setsum_singleton -> measure_eq_setsum_singleton
  1005   measure_space.real_measure_subadditive -> measure_subadditive
  1006   measure_space.real_measure_Union -> measure_Union
  1007   measure_space.real_measure_UNION -> measure_UNION
  1008   measure_space.simple_function_vimage -> simple_function_comp
  1009   measure_space.simple_integral_vimage -> simple_integral_distr
  1010   measure_space.simple_integral_vimage -> simple_integral_distr
  1011   measure_unique_Int_stable -> measure_eqI_generator_eq
  1012   measure_unique_Int_stable_vimage -> measure_eqI_generator_eq
  1013   pair_sigma_algebra.measurable_cut_fst -> sets_Pair1
  1014   pair_sigma_algebra.measurable_cut_snd -> sets_Pair2
  1015   pair_sigma_algebra.measurable_pair_image_fst -> measurable_Pair1
  1016   pair_sigma_algebra.measurable_pair_image_snd -> measurable_Pair2
  1017   pair_sigma_algebra.measurable_product_swap -> measurable_pair_swap_iff
  1018   pair_sigma_algebra.pair_sigma_algebra_measurable -> measurable_pair_swap
  1019   pair_sigma_algebra.pair_sigma_algebra_swap_measurable -> measurable_pair_swap'
  1020   pair_sigma_algebra.sets_swap -> sets_pair_swap
  1021   pair_sigma_finite.measure_cut_measurable_fst -> pair_sigma_finite.measurable_emeasure_Pair1
  1022   pair_sigma_finite.measure_cut_measurable_snd -> pair_sigma_finite.measurable_emeasure_Pair2
  1023   pair_sigma_finite.measure_preserving_swap -> pair_sigma_finite.distr_pair_swap
  1024   pair_sigma_finite.pair_measure_alt2 -> pair_sigma_finite.emeasure_pair_measure_alt2
  1025   pair_sigma_finite.pair_measure_alt -> pair_sigma_finite.emeasure_pair_measure_alt
  1026   pair_sigma_finite.pair_measure_times -> pair_sigma_finite.emeasure_pair_measure_Times
  1027   prob_space.indep_distribution_eq_measure -> prob_space.indep_vars_iff_distr_eq_PiM
  1028   prob_space.indep_var_distributionD -> prob_space.indep_var_distribution_eq
  1029   prob_space.measure_space_1 -> prob_space.emeasure_space_1
  1030   prob_space.prob_space_vimage -> prob_space_distr
  1031   prob_space.random_variable_restrict -> measurable_restrict
  1032   prob_space_unique_Int_stable -> measure_eqI_prob_space
  1033   product_algebraE -> prod_algebraE_all
  1034   product_algebra_generator_der -> prod_algebra_eq_finite
  1035   product_algebra_generator_into_space -> prod_algebra_sets_into_space
  1036   product_algebraI -> sets_PiM_I_finite
  1037   product_measure_exists -> product_sigma_finite.sigma_finite
  1038   product_prob_space.finite_index_eq_finite_product -> product_prob_space.sets_PiM_generator
  1039   product_prob_space.finite_measure_infprod_emb_Pi -> product_prob_space.measure_PiM_emb
  1040   product_prob_space.infprod_spec -> product_prob_space.emeasure_PiM_emb_not_empty
  1041   product_prob_space.measurable_component -> measurable_component_singleton
  1042   product_prob_space.measurable_emb -> measurable_prod_emb
  1043   product_prob_space.measurable_into_infprod_algebra -> measurable_PiM_single
  1044   product_prob_space.measurable_singleton_infprod -> measurable_component_singleton
  1045   product_prob_space.measure_emb -> emeasure_prod_emb
  1046   product_prob_space.measure_preserving_restrict -> product_prob_space.distr_restrict
  1047   product_sigma_algebra.product_algebra_into_space -> space_closed
  1048   product_sigma_finite.measure_fold -> product_sigma_finite.distr_merge
  1049   product_sigma_finite.measure_preserving_component_singelton -> product_sigma_finite.distr_singleton
  1050   product_sigma_finite.measure_preserving_merge -> product_sigma_finite.distr_merge
  1051   sequence_space.measure_infprod -> sequence_space.measure_PiM_countable
  1052   sets_product_algebra -> sets_PiM
  1053   sigma_algebra.measurable_sigma -> measurable_measure_of
  1054   sigma_finite_measure.disjoint_sigma_finite -> sigma_finite_disjoint
  1055   sigma_finite_measure.RN_deriv_vimage -> sigma_finite_measure.RN_deriv_distr
  1056   sigma_product_algebra_sigma_eq -> sigma_prod_algebra_sigma_eq
  1057   space_product_algebra -> space_PiM
  1058 
  1059 * Session HOL-TPTP: support to parse and import TPTP problems (all
  1060 languages) into Isabelle/HOL.
  1061 
  1062 
  1063 *** FOL ***
  1064 
  1065 * New "case_product" attribute (see HOL).
  1066 
  1067 
  1068 *** ZF ***
  1069 
  1070 * Greater support for structured proofs involving induction or case
  1071 analysis.
  1072 
  1073 * Much greater use of mathematical symbols.
  1074 
  1075 * Removal of many ML theorem bindings.  INCOMPATIBILITY.
  1076 
  1077 
  1078 *** ML ***
  1079 
  1080 * Antiquotation @{keyword "name"} produces a parser for outer syntax
  1081 from a minor keyword introduced via theory header declaration.
  1082 
  1083 * Antiquotation @{command_spec "name"} produces the
  1084 Outer_Syntax.command_spec from a major keyword introduced via theory
  1085 header declaration; it can be passed to Outer_Syntax.command etc.
  1086 
  1087 * Local_Theory.define no longer hard-wires default theorem name
  1088 "foo_def", but retains the binding as given.  If that is Binding.empty
  1089 / Attrib.empty_binding, the result is not registered as user-level
  1090 fact.  The Local_Theory.define_internal variant allows to specify a
  1091 non-empty name (used for the foundation in the background theory),
  1092 while omitting the fact binding in the user-context.  Potential
  1093 INCOMPATIBILITY for derived definitional packages: need to specify
  1094 naming policy for primitive definitions more explicitly.
  1095 
  1096 * Renamed Thm.capply to Thm.apply, and Thm.cabs to Thm.lambda in
  1097 conformance with similar operations in structure Term and Logic.
  1098 
  1099 * Antiquotation @{attributes [...]} embeds attribute source
  1100 representation into the ML text, which is particularly useful with
  1101 declarations like Local_Theory.note.
  1102 
  1103 * Structure Proof_Context follows standard naming scheme.  Old
  1104 ProofContext has been discontinued.  INCOMPATIBILITY.
  1105 
  1106 * Refined Local_Theory.declaration {syntax, pervasive}, with subtle
  1107 change of semantics: update is applied to auxiliary local theory
  1108 context as well.
  1109 
  1110 * Modernized some old-style infix operations:
  1111 
  1112   addeqcongs    ~> Simplifier.add_eqcong
  1113   deleqcongs    ~> Simplifier.del_eqcong
  1114   addcongs      ~> Simplifier.add_cong
  1115   delcongs      ~> Simplifier.del_cong
  1116   setmksimps    ~> Simplifier.set_mksimps
  1117   setmkcong     ~> Simplifier.set_mkcong
  1118   setmksym      ~> Simplifier.set_mksym
  1119   setmkeqTrue   ~> Simplifier.set_mkeqTrue
  1120   settermless   ~> Simplifier.set_termless
  1121   setsubgoaler  ~> Simplifier.set_subgoaler
  1122   addsplits     ~> Splitter.add_split
  1123   delsplits     ~> Splitter.del_split
  1124 
  1125 
  1126 *** System ***
  1127 
  1128 * USER_HOME settings variable points to cross-platform user home
  1129 directory, which coincides with HOME on POSIX systems only.  Likewise,
  1130 the Isabelle path specification "~" now expands to $USER_HOME, instead
  1131 of former $HOME.  A different default for USER_HOME may be set
  1132 explicitly in shell environment, before Isabelle settings are
  1133 evaluated.  Minor INCOMPATIBILITY: need to adapt Isabelle path where
  1134 the generic user home was intended.
  1135 
  1136 * ISABELLE_HOME_WINDOWS refers to ISABELLE_HOME in windows file name
  1137 notation, which is useful for the jEdit file browser, for example.
  1138 
  1139 * ISABELLE_JDK_HOME settings variable points to JDK with javac and jar
  1140 (not just JRE).
  1141 
  1142 
  1143 
  1144 New in Isabelle2011-1 (October 2011)
  1145 ------------------------------------
  1146 
  1147 *** General ***
  1148 
  1149 * Improved Isabelle/jEdit Prover IDE (PIDE), which can be invoked as
  1150 "isabelle jedit" or "ISABELLE_HOME/Isabelle" on the command line.
  1151 
  1152   - Management of multiple theory files directly from the editor
  1153     buffer store -- bypassing the file-system (no requirement to save
  1154     files for checking).
  1155 
  1156   - Markup of formal entities within the text buffer, with semantic
  1157     highlighting, tooltips and hyperlinks to jump to defining source
  1158     positions.
  1159 
  1160   - Improved text rendering, with sub/superscripts in the source
  1161     buffer (including support for copy/paste wrt. output panel, HTML
  1162     theory output and other non-Isabelle text boxes).
  1163 
  1164   - Refined scheduling of proof checking and printing of results,
  1165     based on interactive editor view.  (Note: jEdit folding and
  1166     narrowing allows to restrict buffer perspectives explicitly.)
  1167 
  1168   - Reduced CPU performance requirements, usable on machines with few
  1169     cores.
  1170 
  1171   - Reduced memory requirements due to pruning of unused document
  1172     versions (garbage collection).
  1173 
  1174 See also ~~/src/Tools/jEdit/README.html for further information,
  1175 including some remaining limitations.
  1176 
  1177 * Theory loader: source files are exclusively located via the master
  1178 directory of each theory node (where the .thy file itself resides).
  1179 The global load path (such as src/HOL/Library) has been discontinued.
  1180 Note that the path element ~~ may be used to reference theories in the
  1181 Isabelle home folder -- for instance, "~~/src/HOL/Library/FuncSet".
  1182 INCOMPATIBILITY.
  1183 
  1184 * Theory loader: source files are identified by content via SHA1
  1185 digests.  Discontinued former path/modtime identification and optional
  1186 ISABELLE_FILE_IDENT plugin scripts.
  1187 
  1188 * Parallelization of nested Isar proofs is subject to
  1189 Goal.parallel_proofs_threshold (default 100).  See also isabelle
  1190 usedir option -Q.
  1191 
  1192 * Name space: former unsynchronized references are now proper
  1193 configuration options, with more conventional names:
  1194 
  1195   long_names   ~> names_long
  1196   short_names  ~> names_short
  1197   unique_names ~> names_unique
  1198 
  1199 Minor INCOMPATIBILITY, need to declare options in context like this:
  1200 
  1201   declare [[names_unique = false]]
  1202 
  1203 * Literal facts `prop` may contain dummy patterns, e.g. `_ = _`.  Note
  1204 that the result needs to be unique, which means fact specifications
  1205 may have to be refined after enriching a proof context.
  1206 
  1207 * Attribute "case_names" has been refined: the assumptions in each case
  1208 can be named now by following the case name with [name1 name2 ...].
  1209 
  1210 * Isabelle/Isar reference manual has been updated and extended:
  1211   - "Synopsis" provides a catalog of main Isar language concepts.
  1212   - Formal references in syntax diagrams, via @{rail} antiquotation.
  1213   - Updated material from classic "ref" manual, notably about
  1214     "Classical Reasoner".
  1215 
  1216 
  1217 *** HOL ***
  1218 
  1219 * Class bot and top require underlying partial order rather than
  1220 preorder: uniqueness of bot and top is guaranteed.  INCOMPATIBILITY.
  1221 
  1222 * Class complete_lattice: generalized a couple of lemmas from sets;
  1223 generalized theorems INF_cong and SUP_cong.  New type classes for
  1224 complete boolean algebras and complete linear orders.  Lemmas
  1225 Inf_less_iff, less_Sup_iff, INF_less_iff, less_SUP_iff now reside in
  1226 class complete_linorder.
  1227 
  1228 Changed proposition of lemmas Inf_bool_def, Sup_bool_def, Inf_fun_def,
  1229 Sup_fun_def, Inf_apply, Sup_apply.
  1230 
  1231 Removed redundant lemmas (the right hand side gives hints how to
  1232 replace them for (metis ...), or (simp only: ...) proofs):
  1233 
  1234   Inf_singleton ~> Inf_insert [where A="{}", unfolded Inf_empty inf_top_right]
  1235   Sup_singleton ~> Sup_insert [where A="{}", unfolded Sup_empty sup_bot_right]
  1236   Inf_binary ~> Inf_insert, Inf_empty, and inf_top_right
  1237   Sup_binary ~> Sup_insert, Sup_empty, and sup_bot_right
  1238   Int_eq_Inter ~> Inf_insert, Inf_empty, and inf_top_right
  1239   Un_eq_Union ~> Sup_insert, Sup_empty, and sup_bot_right
  1240   Inter_def ~> INF_def, image_def
  1241   Union_def ~> SUP_def, image_def
  1242   INT_eq ~> INF_def, and image_def
  1243   UN_eq ~> SUP_def, and image_def
  1244   INF_subset ~> INF_superset_mono [OF _ order_refl]
  1245 
  1246 More consistent and comprehensive names:
  1247 
  1248   INTER_eq_Inter_image ~> INF_def
  1249   UNION_eq_Union_image ~> SUP_def
  1250   INFI_def ~> INF_def
  1251   SUPR_def ~> SUP_def
  1252   INF_leI ~> INF_lower
  1253   INF_leI2 ~> INF_lower2
  1254   le_INFI ~> INF_greatest
  1255   le_SUPI ~> SUP_upper
  1256   le_SUPI2 ~> SUP_upper2
  1257   SUP_leI ~> SUP_least
  1258   INFI_bool_eq ~> INF_bool_eq
  1259   SUPR_bool_eq ~> SUP_bool_eq
  1260   INFI_apply ~> INF_apply
  1261   SUPR_apply ~> SUP_apply
  1262   INTER_def ~> INTER_eq
  1263   UNION_def ~> UNION_eq
  1264 
  1265 INCOMPATIBILITY.
  1266 
  1267 * Renamed theory Complete_Lattice to Complete_Lattices.
  1268 INCOMPATIBILITY.
  1269 
  1270 * Theory Complete_Lattices: lemmas Inf_eq_top_iff, INF_eq_top_iff,
  1271 INF_image, Inf_insert, INF_top, Inf_top_conv, INF_top_conv, SUP_bot,
  1272 Sup_bot_conv, SUP_bot_conv, Sup_eq_top_iff, SUP_eq_top_iff, SUP_image,
  1273 Sup_insert are now declared as [simp].  INCOMPATIBILITY.
  1274 
  1275 * Theory Lattice: lemmas compl_inf_bot, compl_le_comp_iff,
  1276 compl_sup_top, inf_idem, inf_left_idem, inf_sup_absorb, sup_idem,
  1277 sup_inf_absob, sup_left_idem are now declared as [simp].  Minor
  1278 INCOMPATIBILITY.
  1279 
  1280 * Added syntactic classes "inf" and "sup" for the respective
  1281 constants.  INCOMPATIBILITY: Changes in the argument order of the
  1282 (mostly internal) locale predicates for some derived classes.
  1283 
  1284 * Theorem collections ball_simps and bex_simps do not contain theorems
  1285 referring to UNION any longer; these have been moved to collection
  1286 UN_ball_bex_simps.  INCOMPATIBILITY.
  1287 
  1288 * Theory Archimedean_Field: floor now is defined as parameter of a
  1289 separate type class floor_ceiling.
  1290 
  1291 * Theory Finite_Set: more coherent development of fold_set locales:
  1292 
  1293     locale fun_left_comm ~> locale comp_fun_commute
  1294     locale fun_left_comm_idem ~> locale comp_fun_idem
  1295 
  1296 Both use point-free characterization; interpretation proofs may need
  1297 adjustment.  INCOMPATIBILITY.
  1298 
  1299 * Theory Limits: Type "'a net" has been renamed to "'a filter", in
  1300 accordance with standard mathematical terminology. INCOMPATIBILITY.
  1301 
  1302 * Theory Complex_Main: The locale interpretations for the
  1303 bounded_linear and bounded_bilinear locales have been removed, in
  1304 order to reduce the number of duplicate lemmas. Users must use the
  1305 original names for distributivity theorems, potential INCOMPATIBILITY.
  1306 
  1307   divide.add ~> add_divide_distrib
  1308   divide.diff ~> diff_divide_distrib
  1309   divide.setsum ~> setsum_divide_distrib
  1310   mult.add_right ~> right_distrib
  1311   mult.diff_right ~> right_diff_distrib
  1312   mult_right.setsum ~> setsum_right_distrib
  1313   mult_left.diff ~> left_diff_distrib
  1314 
  1315 * Theory Complex_Main: Several redundant theorems have been removed or
  1316 replaced by more general versions. INCOMPATIBILITY.
  1317 
  1318   real_diff_def ~> minus_real_def
  1319   real_divide_def ~> divide_real_def
  1320   real_less_def ~> less_le
  1321   real_abs_def ~> abs_real_def
  1322   real_sgn_def ~> sgn_real_def
  1323   real_mult_commute ~> mult_commute
  1324   real_mult_assoc ~> mult_assoc
  1325   real_mult_1 ~> mult_1_left
  1326   real_add_mult_distrib ~> left_distrib
  1327   real_zero_not_eq_one ~> zero_neq_one
  1328   real_mult_inverse_left ~> left_inverse
  1329   INVERSE_ZERO ~> inverse_zero
  1330   real_le_refl ~> order_refl
  1331   real_le_antisym ~> order_antisym
  1332   real_le_trans ~> order_trans
  1333   real_le_linear ~> linear
  1334   real_le_eq_diff ~> le_iff_diff_le_0
  1335   real_add_left_mono ~> add_left_mono
  1336   real_mult_order ~> mult_pos_pos
  1337   real_mult_less_mono2 ~> mult_strict_left_mono
  1338   real_of_int_real_of_nat ~> real_of_int_of_nat_eq
  1339   real_0_le_divide_iff ~> zero_le_divide_iff
  1340   realpow_two_disj ~> power2_eq_iff
  1341   real_squared_diff_one_factored ~> square_diff_one_factored
  1342   realpow_two_diff ~> square_diff_square_factored
  1343   reals_complete2 ~> complete_real
  1344   real_sum_squared_expand ~> power2_sum
  1345   exp_ln_eq ~> ln_unique
  1346   expi_add ~> exp_add
  1347   expi_zero ~> exp_zero
  1348   lemma_DERIV_subst ~> DERIV_cong
  1349   LIMSEQ_Zfun_iff ~> tendsto_Zfun_iff
  1350   LIMSEQ_const ~> tendsto_const
  1351   LIMSEQ_norm ~> tendsto_norm
  1352   LIMSEQ_add ~> tendsto_add
  1353   LIMSEQ_minus ~> tendsto_minus
  1354   LIMSEQ_minus_cancel ~> tendsto_minus_cancel
  1355   LIMSEQ_diff ~> tendsto_diff
  1356   bounded_linear.LIMSEQ ~> bounded_linear.tendsto
  1357   bounded_bilinear.LIMSEQ ~> bounded_bilinear.tendsto
  1358   LIMSEQ_mult ~> tendsto_mult
  1359   LIMSEQ_inverse ~> tendsto_inverse
  1360   LIMSEQ_divide ~> tendsto_divide
  1361   LIMSEQ_pow ~> tendsto_power
  1362   LIMSEQ_setsum ~> tendsto_setsum
  1363   LIMSEQ_setprod ~> tendsto_setprod
  1364   LIMSEQ_norm_zero ~> tendsto_norm_zero_iff
  1365   LIMSEQ_rabs_zero ~> tendsto_rabs_zero_iff
  1366   LIMSEQ_imp_rabs ~> tendsto_rabs
  1367   LIMSEQ_add_minus ~> tendsto_add [OF _ tendsto_minus]
  1368   LIMSEQ_add_const ~> tendsto_add [OF _ tendsto_const]
  1369   LIMSEQ_diff_const ~> tendsto_diff [OF _ tendsto_const]
  1370   LIMSEQ_Complex ~> tendsto_Complex
  1371   LIM_ident ~> tendsto_ident_at
  1372   LIM_const ~> tendsto_const
  1373   LIM_add ~> tendsto_add
  1374   LIM_add_zero ~> tendsto_add_zero
  1375   LIM_minus ~> tendsto_minus
  1376   LIM_diff ~> tendsto_diff
  1377   LIM_norm ~> tendsto_norm
  1378   LIM_norm_zero ~> tendsto_norm_zero
  1379   LIM_norm_zero_cancel ~> tendsto_norm_zero_cancel
  1380   LIM_norm_zero_iff ~> tendsto_norm_zero_iff
  1381   LIM_rabs ~> tendsto_rabs
  1382   LIM_rabs_zero ~> tendsto_rabs_zero
  1383   LIM_rabs_zero_cancel ~> tendsto_rabs_zero_cancel
  1384   LIM_rabs_zero_iff ~> tendsto_rabs_zero_iff
  1385   LIM_compose ~> tendsto_compose
  1386   LIM_mult ~> tendsto_mult
  1387   LIM_scaleR ~> tendsto_scaleR
  1388   LIM_of_real ~> tendsto_of_real
  1389   LIM_power ~> tendsto_power
  1390   LIM_inverse ~> tendsto_inverse
  1391   LIM_sgn ~> tendsto_sgn
  1392   isCont_LIM_compose ~> isCont_tendsto_compose
  1393   bounded_linear.LIM ~> bounded_linear.tendsto
  1394   bounded_linear.LIM_zero ~> bounded_linear.tendsto_zero
  1395   bounded_bilinear.LIM ~> bounded_bilinear.tendsto
  1396   bounded_bilinear.LIM_prod_zero ~> bounded_bilinear.tendsto_zero
  1397   bounded_bilinear.LIM_left_zero ~> bounded_bilinear.tendsto_left_zero
  1398   bounded_bilinear.LIM_right_zero ~> bounded_bilinear.tendsto_right_zero
  1399   LIM_inverse_fun ~> tendsto_inverse [OF tendsto_ident_at]
  1400 
  1401 * Theory Complex_Main: The definition of infinite series was
  1402 generalized.  Now it is defined on the type class {topological_space,
  1403 comm_monoid_add}.  Hence it is useable also for extended real numbers.
  1404 
  1405 * Theory Complex_Main: The complex exponential function "expi" is now
  1406 a type-constrained abbreviation for "exp :: complex => complex"; thus
  1407 several polymorphic lemmas about "exp" are now applicable to "expi".
  1408 
  1409 * Code generation:
  1410 
  1411   - Theory Library/Code_Char_ord provides native ordering of
  1412     characters in the target language.
  1413 
  1414   - Commands code_module and code_library are legacy, use export_code
  1415     instead.
  1416 
  1417   - Method "evaluation" is legacy, use method "eval" instead.
  1418 
  1419   - Legacy evaluator "SML" is deactivated by default.  May be
  1420     reactivated by the following theory command:
  1421 
  1422       setup {* Value.add_evaluator ("SML", Codegen.eval_term) *}
  1423 
  1424 * Declare ext [intro] by default.  Rare INCOMPATIBILITY.
  1425 
  1426 * New proof method "induction" that gives induction hypotheses the
  1427 name "IH", thus distinguishing them from further hypotheses that come
  1428 from rule induction.  The latter are still called "hyps".  Method
  1429 "induction" is a thin wrapper around "induct" and follows the same
  1430 syntax.
  1431 
  1432 * Method "fastsimp" has been renamed to "fastforce", but "fastsimp" is
  1433 still available as a legacy feature for some time.
  1434 
  1435 * Nitpick:
  1436   - Added "need" and "total_consts" options.
  1437   - Reintroduced "show_skolems" option by popular demand.
  1438   - Renamed attribute: nitpick_def ~> nitpick_unfold.
  1439     INCOMPATIBILITY.
  1440 
  1441 * Sledgehammer:
  1442   - Use quasi-sound (and efficient) translations by default.
  1443   - Added support for the following provers: E-ToFoF, LEO-II,
  1444     Satallax, SNARK, Waldmeister, and Z3 with TPTP syntax.
  1445   - Automatically preplay and minimize proofs before showing them if
  1446     this can be done within reasonable time.
  1447   - sledgehammer available_provers ~> sledgehammer supported_provers.
  1448     INCOMPATIBILITY.
  1449   - Added "preplay_timeout", "slicing", "type_enc", "sound",
  1450     "max_mono_iters", and "max_new_mono_instances" options.
  1451   - Removed "explicit_apply" and "full_types" options as well as "Full
  1452     Types" Proof General menu item. INCOMPATIBILITY.
  1453 
  1454 * Metis:
  1455   - Removed "metisF" -- use "metis" instead. INCOMPATIBILITY.
  1456   - Obsoleted "metisFT" -- use "metis (full_types)" instead.
  1457     INCOMPATIBILITY.
  1458 
  1459 * Command 'try':
  1460   - Renamed 'try_methods' and added "simp:", "intro:", "dest:", and
  1461     "elim:" options. INCOMPATIBILITY.
  1462   - Introduced 'try' that not only runs 'try_methods' but also
  1463     'solve_direct', 'sledgehammer', 'quickcheck', and 'nitpick'.
  1464 
  1465 * Quickcheck:
  1466   - Added "eval" option to evaluate terms for the found counterexample
  1467     (currently only supported by the default (exhaustive) tester).
  1468   - Added post-processing of terms to obtain readable counterexamples
  1469     (currently only supported by the default (exhaustive) tester).
  1470   - New counterexample generator quickcheck[narrowing] enables
  1471     narrowing-based testing.  Requires the Glasgow Haskell compiler
  1472     with its installation location defined in the Isabelle settings
  1473     environment as ISABELLE_GHC.
  1474   - Removed quickcheck tester "SML" based on the SML code generator
  1475     (formly in HOL/Library).
  1476 
  1477 * Function package: discontinued option "tailrec".  INCOMPATIBILITY,
  1478 use 'partial_function' instead.
  1479 
  1480 * Theory Library/Extended_Reals replaces now the positive extended
  1481 reals found in probability theory. This file is extended by
  1482 Multivariate_Analysis/Extended_Real_Limits.
  1483 
  1484 * Theory Library/Old_Recdef: old 'recdef' package has been moved here,
  1485 from where it must be imported explicitly if it is really required.
  1486 INCOMPATIBILITY.
  1487 
  1488 * Theory Library/Wfrec: well-founded recursion combinator "wfrec" has
  1489 been moved here.  INCOMPATIBILITY.
  1490 
  1491 * Theory Library/Saturated provides type of numbers with saturated
  1492 arithmetic.
  1493 
  1494 * Theory Library/Product_Lattice defines a pointwise ordering for the
  1495 product type 'a * 'b, and provides instance proofs for various order
  1496 and lattice type classes.
  1497 
  1498 * Theory Library/Countable now provides the "countable_datatype" proof
  1499 method for proving "countable" class instances for datatypes.
  1500 
  1501 * Theory Library/Cset_Monad allows do notation for computable sets
  1502 (cset) via the generic monad ad-hoc overloading facility.
  1503 
  1504 * Library: Theories of common data structures are split into theories
  1505 for implementation, an invariant-ensuring type, and connection to an
  1506 abstract type. INCOMPATIBILITY.
  1507 
  1508   - RBT is split into RBT and RBT_Mapping.
  1509   - AssocList is split and renamed into AList and AList_Mapping.
  1510   - DList is split into DList_Impl, DList, and DList_Cset.
  1511   - Cset is split into Cset and List_Cset.
  1512 
  1513 * Theory Library/Nat_Infinity has been renamed to
  1514 Library/Extended_Nat, with name changes of the following types and
  1515 constants:
  1516 
  1517   type inat   ~> type enat
  1518   Fin         ~> enat
  1519   Infty       ~> infinity (overloaded)
  1520   iSuc        ~> eSuc
  1521   the_Fin     ~> the_enat
  1522 
  1523 Every theorem name containing "inat", "Fin", "Infty", or "iSuc" has
  1524 been renamed accordingly. INCOMPATIBILITY.
  1525 
  1526 * Session Multivariate_Analysis: The euclidean_space type class now
  1527 fixes a constant "Basis :: 'a set" consisting of the standard
  1528 orthonormal basis for the type. Users now have the option of
  1529 quantifying over this set instead of using the "basis" function, e.g.
  1530 "ALL x:Basis. P x" vs "ALL i<DIM('a). P (basis i)".
  1531 
  1532 * Session Multivariate_Analysis: Type "('a, 'b) cart" has been renamed
  1533 to "('a, 'b) vec" (the syntax "'a ^ 'b" remains unaffected). Constants
  1534 "Cart_nth" and "Cart_lambda" have been respectively renamed to
  1535 "vec_nth" and "vec_lambda"; theorems mentioning those names have
  1536 changed to match. Definition theorems for overloaded constants now use
  1537 the standard "foo_vec_def" naming scheme. A few other theorems have
  1538 been renamed as follows (INCOMPATIBILITY):
  1539 
  1540   Cart_eq          ~> vec_eq_iff
  1541   dist_nth_le_cart ~> dist_vec_nth_le
  1542   tendsto_vector   ~> vec_tendstoI
  1543   Cauchy_vector    ~> vec_CauchyI
  1544 
  1545 * Session Multivariate_Analysis: Several duplicate theorems have been
  1546 removed, and other theorems have been renamed or replaced with more
  1547 general versions. INCOMPATIBILITY.
  1548 
  1549   finite_choice ~> finite_set_choice
  1550   eventually_conjI ~> eventually_conj
  1551   eventually_and ~> eventually_conj_iff
  1552   eventually_false ~> eventually_False
  1553   setsum_norm ~> norm_setsum
  1554   Lim_sequentially ~> LIMSEQ_def
  1555   Lim_ident_at ~> LIM_ident
  1556   Lim_const ~> tendsto_const
  1557   Lim_cmul ~> tendsto_scaleR [OF tendsto_const]
  1558   Lim_neg ~> tendsto_minus
  1559   Lim_add ~> tendsto_add
  1560   Lim_sub ~> tendsto_diff
  1561   Lim_mul ~> tendsto_scaleR
  1562   Lim_vmul ~> tendsto_scaleR [OF _ tendsto_const]
  1563   Lim_null_norm ~> tendsto_norm_zero_iff [symmetric]
  1564   Lim_linear ~> bounded_linear.tendsto
  1565   Lim_component ~> tendsto_euclidean_component
  1566   Lim_component_cart ~> tendsto_vec_nth
  1567   Lim_inner ~> tendsto_inner [OF tendsto_const]
  1568   dot_lsum ~> inner_setsum_left
  1569   dot_rsum ~> inner_setsum_right
  1570   continuous_cmul ~> continuous_scaleR [OF continuous_const]
  1571   continuous_neg ~> continuous_minus
  1572   continuous_sub ~> continuous_diff
  1573   continuous_vmul ~> continuous_scaleR [OF _ continuous_const]
  1574   continuous_mul ~> continuous_scaleR
  1575   continuous_inv ~> continuous_inverse
  1576   continuous_at_within_inv ~> continuous_at_within_inverse
  1577   continuous_at_inv ~> continuous_at_inverse
  1578   continuous_at_norm ~> continuous_norm [OF continuous_at_id]
  1579   continuous_at_infnorm ~> continuous_infnorm [OF continuous_at_id]
  1580   continuous_at_component ~> continuous_component [OF continuous_at_id]
  1581   continuous_on_neg ~> continuous_on_minus
  1582   continuous_on_sub ~> continuous_on_diff
  1583   continuous_on_cmul ~> continuous_on_scaleR [OF continuous_on_const]
  1584   continuous_on_vmul ~> continuous_on_scaleR [OF _ continuous_on_const]
  1585   continuous_on_mul ~> continuous_on_scaleR
  1586   continuous_on_mul_real ~> continuous_on_mult
  1587   continuous_on_inner ~> continuous_on_inner [OF continuous_on_const]
  1588   continuous_on_norm ~> continuous_on_norm [OF continuous_on_id]
  1589   continuous_on_inverse ~> continuous_on_inv
  1590   uniformly_continuous_on_neg ~> uniformly_continuous_on_minus
  1591   uniformly_continuous_on_sub ~> uniformly_continuous_on_diff
  1592   subset_interior ~> interior_mono
  1593   subset_closure ~> closure_mono
  1594   closure_univ ~> closure_UNIV
  1595   real_arch_lt ~> reals_Archimedean2
  1596   real_arch ~> reals_Archimedean3
  1597   real_abs_norm ~> abs_norm_cancel
  1598   real_abs_sub_norm ~> norm_triangle_ineq3
  1599   norm_cauchy_schwarz_abs ~> Cauchy_Schwarz_ineq2
  1600 
  1601 * Session HOL-Probability:
  1602   - Caratheodory's extension lemma is now proved for ring_of_sets.
  1603   - Infinite products of probability measures are now available.
  1604   - Sigma closure is independent, if the generator is independent
  1605   - Use extended reals instead of positive extended
  1606     reals. INCOMPATIBILITY.
  1607 
  1608 * Session HOLCF: Discontinued legacy theorem names, INCOMPATIBILITY.
  1609 
  1610   expand_fun_below ~> fun_below_iff
  1611   below_fun_ext ~> fun_belowI
  1612   expand_cfun_eq ~> cfun_eq_iff
  1613   ext_cfun ~> cfun_eqI
  1614   expand_cfun_below ~> cfun_below_iff
  1615   below_cfun_ext ~> cfun_belowI
  1616   monofun_fun_fun ~> fun_belowD
  1617   monofun_fun_arg ~> monofunE
  1618   monofun_lub_fun ~> adm_monofun [THEN admD]
  1619   cont_lub_fun ~> adm_cont [THEN admD]
  1620   cont2cont_Rep_CFun ~> cont2cont_APP
  1621   cont_Rep_CFun_app ~> cont_APP_app
  1622   cont_Rep_CFun_app_app ~> cont_APP_app_app
  1623   cont_cfun_fun ~> cont_Rep_cfun1 [THEN contE]
  1624   cont_cfun_arg ~> cont_Rep_cfun2 [THEN contE]
  1625   contlub_cfun ~> lub_APP [symmetric]
  1626   contlub_LAM ~> lub_LAM [symmetric]
  1627   thelubI ~> lub_eqI
  1628   UU_I ~> bottomI
  1629   lift_distinct1 ~> lift.distinct(1)
  1630   lift_distinct2 ~> lift.distinct(2)
  1631   Def_not_UU ~> lift.distinct(2)
  1632   Def_inject ~> lift.inject
  1633   below_UU_iff ~> below_bottom_iff
  1634   eq_UU_iff ~> eq_bottom_iff
  1635 
  1636 
  1637 *** Document preparation ***
  1638 
  1639 * Antiquotation @{rail} layouts railroad syntax diagrams, see also
  1640 isar-ref manual, both for description and actual application of the
  1641 same.
  1642 
  1643 * Antiquotation @{value} evaluates the given term and presents its
  1644 result.
  1645 
  1646 * Antiquotations: term style "isub" provides ad-hoc conversion of
  1647 variables x1, y23 into subscripted form x\<^isub>1,
  1648 y\<^isub>2\<^isub>3.
  1649 
  1650 * Predefined LaTeX macros for Isabelle symbols \<bind> and \<then>
  1651 (e.g. see ~~/src/HOL/Library/Monad_Syntax.thy).
  1652 
  1653 * Localized \isabellestyle switch can be used within blocks or groups
  1654 like this:
  1655 
  1656   \isabellestyle{it}  %preferred default
  1657   {\isabellestylett @{text "typewriter stuff"}}
  1658 
  1659 * Discontinued special treatment of hard tabulators.  Implicit
  1660 tab-width is now defined as 1.  Potential INCOMPATIBILITY for visual
  1661 layouts.
  1662 
  1663 
  1664 *** ML ***
  1665 
  1666 * The inner syntax of sort/type/term/prop supports inlined YXML
  1667 representations within quoted string tokens.  By encoding logical
  1668 entities via Term_XML (in ML or Scala) concrete syntax can be
  1669 bypassed, which is particularly useful for producing bits of text
  1670 under external program control.
  1671 
  1672 * Antiquotations for ML and document preparation are managed as theory
  1673 data, which requires explicit setup.
  1674 
  1675 * Isabelle_Process.is_active allows tools to check if the official
  1676 process wrapper is running (Isabelle/Scala/jEdit) or the old TTY loop
  1677 (better known as Proof General).
  1678 
  1679 * Structure Proof_Context follows standard naming scheme.  Old
  1680 ProofContext is still available for some time as legacy alias.
  1681 
  1682 * Structure Timing provides various operations for timing; supersedes
  1683 former start_timing/end_timing etc.
  1684 
  1685 * Path.print is the official way to show file-system paths to users
  1686 (including quotes etc.).
  1687 
  1688 * Inner syntax: identifiers in parse trees of generic categories
  1689 "logic", "aprop", "idt" etc. carry position information (disguised as
  1690 type constraints).  Occasional INCOMPATIBILITY with non-compliant
  1691 translations that choke on unexpected type constraints.  Positions can
  1692 be stripped in ML translations via Syntax.strip_positions /
  1693 Syntax.strip_positions_ast, or via the syntax constant
  1694 "_strip_positions" within parse trees.  As last resort, positions can
  1695 be disabled via the configuration option Syntax.positions, which is
  1696 called "syntax_positions" in Isar attribute syntax.
  1697 
  1698 * Discontinued special status of various ML structures that contribute
  1699 to structure Syntax (Ast, Lexicon, Mixfix, Parser, Printer etc.): less
  1700 pervasive content, no inclusion in structure Syntax.  INCOMPATIBILITY,
  1701 refer directly to Ast.Constant, Lexicon.is_identifier,
  1702 Syntax_Trans.mk_binder_tr etc.
  1703 
  1704 * Typed print translation: discontinued show_sorts argument, which is
  1705 already available via context of "advanced" translation.
  1706 
  1707 * Refined PARALLEL_GOALS tactical: degrades gracefully for schematic
  1708 goal states; body tactic needs to address all subgoals uniformly.
  1709 
  1710 * Slightly more special eq_list/eq_set, with shortcut involving
  1711 pointer equality (assumes that eq relation is reflexive).
  1712 
  1713 * Classical tactics use proper Proof.context instead of historic types
  1714 claset/clasimpset.  Old-style declarations like addIs, addEs, addDs
  1715 operate directly on Proof.context.  Raw type claset retains its use as
  1716 snapshot of the classical context, which can be recovered via
  1717 (put_claset HOL_cs) etc.  Type clasimpset has been discontinued.
  1718 INCOMPATIBILITY, classical tactics and derived proof methods require
  1719 proper Proof.context.
  1720 
  1721 
  1722 *** System ***
  1723 
  1724 * Discontinued support for Poly/ML 5.2, which was the last version
  1725 without proper multithreading and TimeLimit implementation.
  1726 
  1727 * Discontinued old lib/scripts/polyml-platform, which has been
  1728 obsolete since Isabelle2009-2.
  1729 
  1730 * Various optional external tools are referenced more robustly and
  1731 uniformly by explicit Isabelle settings as follows:
  1732 
  1733   ISABELLE_CSDP   (formerly CSDP_EXE)
  1734   ISABELLE_GHC    (formerly EXEC_GHC or GHC_PATH)
  1735   ISABELLE_OCAML  (formerly EXEC_OCAML)
  1736   ISABELLE_SWIPL  (formerly EXEC_SWIPL)
  1737   ISABELLE_YAP    (formerly EXEC_YAP)
  1738 
  1739 Note that automated detection from the file-system or search path has
  1740 been discontinued.  INCOMPATIBILITY.
  1741 
  1742 * Scala layer provides JVM method invocation service for static
  1743 methods of type (String)String, see Invoke_Scala.method in ML.  For
  1744 example:
  1745 
  1746   Invoke_Scala.method "java.lang.System.getProperty" "java.home"
  1747 
  1748 Together with YXML.string_of_body/parse_body and XML.Encode/Decode
  1749 this allows to pass structured values between ML and Scala.
  1750 
  1751 * The IsabelleText fonts includes some further glyphs to support the
  1752 Prover IDE.  Potential INCOMPATIBILITY: users who happen to have
  1753 installed a local copy (which is normally *not* required) need to
  1754 delete or update it from ~~/lib/fonts/.
  1755 
  1756 
  1757 
  1758 New in Isabelle2011 (January 2011)
  1759 ----------------------------------
  1760 
  1761 *** General ***
  1762 
  1763 * Experimental Prover IDE based on Isabelle/Scala and jEdit (see
  1764 src/Tools/jEdit).  This also serves as IDE for Isabelle/ML, with
  1765 useful tooltips and hyperlinks produced from its static analysis.  The
  1766 bundled component provides an executable Isabelle tool that can be run
  1767 like this:
  1768 
  1769   Isabelle2011/bin/isabelle jedit
  1770 
  1771 * Significantly improved Isabelle/Isar implementation manual.
  1772 
  1773 * System settings: ISABELLE_HOME_USER now includes ISABELLE_IDENTIFIER
  1774 (and thus refers to something like $HOME/.isabelle/Isabelle2011),
  1775 while the default heap location within that directory lacks that extra
  1776 suffix.  This isolates multiple Isabelle installations from each
  1777 other, avoiding problems with old settings in new versions.
  1778 INCOMPATIBILITY, need to copy/upgrade old user settings manually.
  1779 
  1780 * Source files are always encoded as UTF-8, instead of old-fashioned
  1781 ISO-Latin-1.  INCOMPATIBILITY.  Isabelle LaTeX documents might require
  1782 the following package declarations:
  1783 
  1784   \usepackage[utf8]{inputenc}
  1785   \usepackage{textcomp}
  1786 
  1787 * Explicit treatment of UTF-8 sequences as Isabelle symbols, such that
  1788 a Unicode character is treated as a single symbol, not a sequence of
  1789 non-ASCII bytes as before.  Since Isabelle/ML string literals may
  1790 contain symbols without further backslash escapes, Unicode can now be
  1791 used here as well.  Recall that Symbol.explode in ML provides a
  1792 consistent view on symbols, while raw explode (or String.explode)
  1793 merely give a byte-oriented representation.
  1794 
  1795 * Theory loader: source files are primarily located via the master
  1796 directory of each theory node (where the .thy file itself resides).
  1797 The global load path is still partially available as legacy feature.
  1798 Minor INCOMPATIBILITY due to subtle change in file lookup: use
  1799 explicit paths, relatively to the theory.
  1800 
  1801 * Special treatment of ML file names has been discontinued.
  1802 Historically, optional extensions .ML or .sml were added on demand --
  1803 at the cost of clarity of file dependencies.  Recall that Isabelle/ML
  1804 files exclusively use the .ML extension.  Minor INCOMPATIBILTY.
  1805 
  1806 * Various options that affect pretty printing etc. are now properly
  1807 handled within the context via configuration options, instead of
  1808 unsynchronized references or print modes.  There are both ML Config.T
  1809 entities and Isar declaration attributes to access these.
  1810 
  1811   ML (Config.T)                 Isar (attribute)
  1812 
  1813   eta_contract                  eta_contract
  1814   show_brackets                 show_brackets
  1815   show_sorts                    show_sorts
  1816   show_types                    show_types
  1817   show_question_marks           show_question_marks
  1818   show_consts                   show_consts
  1819   show_abbrevs                  show_abbrevs
  1820 
  1821   Syntax.ast_trace              syntax_ast_trace
  1822   Syntax.ast_stat               syntax_ast_stat
  1823   Syntax.ambiguity_level        syntax_ambiguity_level
  1824 
  1825   Goal_Display.goals_limit      goals_limit
  1826   Goal_Display.show_main_goal   show_main_goal
  1827 
  1828   Method.rule_trace             rule_trace
  1829 
  1830   Thy_Output.display            thy_output_display
  1831   Thy_Output.quotes             thy_output_quotes
  1832   Thy_Output.indent             thy_output_indent
  1833   Thy_Output.source             thy_output_source
  1834   Thy_Output.break              thy_output_break
  1835 
  1836 Note that corresponding "..._default" references in ML may only be
  1837 changed globally at the ROOT session setup, but *not* within a theory.
  1838 The option "show_abbrevs" supersedes the former print mode
  1839 "no_abbrevs" with inverted meaning.
  1840 
  1841 * More systematic naming of some configuration options.
  1842 INCOMPATIBILITY.
  1843 
  1844   trace_simp  ~>  simp_trace
  1845   debug_simp  ~>  simp_debug
  1846 
  1847 * Support for real valued configuration options, using simplistic
  1848 floating-point notation that coincides with the inner syntax for
  1849 float_token.
  1850 
  1851 * Support for real valued preferences (with approximative PGIP type):
  1852 front-ends need to accept "pgint" values in float notation.
  1853 INCOMPATIBILITY.
  1854 
  1855 * The IsabelleText font now includes Cyrillic, Hebrew, Arabic from
  1856 DejaVu Sans.
  1857 
  1858 * Discontinued support for Poly/ML 5.0 and 5.1 versions.
  1859 
  1860 
  1861 *** Pure ***
  1862 
  1863 * Command 'type_synonym' (with single argument) replaces somewhat
  1864 outdated 'types', which is still available as legacy feature for some
  1865 time.
  1866 
  1867 * Command 'nonterminal' (with 'and' separated list of arguments)
  1868 replaces somewhat outdated 'nonterminals'.  INCOMPATIBILITY.
  1869 
  1870 * Command 'notepad' replaces former 'example_proof' for
  1871 experimentation in Isar without any result.  INCOMPATIBILITY.
  1872 
  1873 * Locale interpretation commands 'interpret' and 'sublocale' accept
  1874 lists of equations to map definitions in a locale to appropriate
  1875 entities in the context of the interpretation.  The 'interpretation'
  1876 command already provided this functionality.
  1877 
  1878 * Diagnostic command 'print_dependencies' prints the locale instances
  1879 that would be activated if the specified expression was interpreted in
  1880 the current context.  Variant "print_dependencies!" assumes a context
  1881 without interpretations.
  1882 
  1883 * Diagnostic command 'print_interps' prints interpretations in proofs
  1884 in addition to interpretations in theories.
  1885 
  1886 * Discontinued obsolete 'global' and 'local' commands to manipulate
  1887 the theory name space.  Rare INCOMPATIBILITY.  The ML functions
  1888 Sign.root_path and Sign.local_path may be applied directly where this
  1889 feature is still required for historical reasons.
  1890 
  1891 * Discontinued obsolete 'constdefs' command.  INCOMPATIBILITY, use
  1892 'definition' instead.
  1893 
  1894 * The "prems" fact, which refers to the accidental collection of
  1895 foundational premises in the context, is now explicitly marked as
  1896 legacy feature and will be discontinued soon.  Consider using "assms"
  1897 of the head statement or reference facts by explicit names.
  1898 
  1899 * Document antiquotations @{class} and @{type} print classes and type
  1900 constructors.
  1901 
  1902 * Document antiquotation @{file} checks file/directory entries within
  1903 the local file system.
  1904 
  1905 
  1906 *** HOL ***
  1907 
  1908 * Coercive subtyping: functions can be declared as coercions and type
  1909 inference will add them as necessary upon input of a term.  Theory
  1910 Complex_Main declares real :: nat => real and real :: int => real as
  1911 coercions. A coercion function f is declared like this:
  1912 
  1913   declare [[coercion f]]
  1914 
  1915 To lift coercions through type constructors (e.g. from nat => real to
  1916 nat list => real list), map functions can be declared, e.g.
  1917 
  1918   declare [[coercion_map map]]
  1919 
  1920 Currently coercion inference is activated only in theories including
  1921 real numbers, i.e. descendants of Complex_Main.  This is controlled by
  1922 the configuration option "coercion_enabled", e.g. it can be enabled in
  1923 other theories like this:
  1924 
  1925   declare [[coercion_enabled]]
  1926 
  1927 * Command 'partial_function' provides basic support for recursive
  1928 function definitions over complete partial orders.  Concrete instances
  1929 are provided for i) the option type, ii) tail recursion on arbitrary
  1930 types, and iii) the heap monad of Imperative_HOL.  See
  1931 src/HOL/ex/Fundefs.thy and src/HOL/Imperative_HOL/ex/Linked_Lists.thy
  1932 for examples.
  1933 
  1934 * Function package: f.psimps rules are no longer implicitly declared
  1935 as [simp].  INCOMPATIBILITY.
  1936 
  1937 * Datatype package: theorems generated for executable equality (class
  1938 "eq") carry proper names and are treated as default code equations.
  1939 
  1940 * Inductive package: now offers command 'inductive_simps' to
  1941 automatically derive instantiated and simplified equations for
  1942 inductive predicates, similar to 'inductive_cases'.
  1943 
  1944 * Command 'enriched_type' allows to register properties of the
  1945 functorial structure of types.
  1946 
  1947 * Improved infrastructure for term evaluation using code generator
  1948 techniques, in particular static evaluation conversions.
  1949 
  1950 * Code generator: Scala (2.8 or higher) has been added to the target
  1951 languages.
  1952 
  1953 * Code generator: globbing constant expressions "*" and "Theory.*"
  1954 have been replaced by the more idiomatic "_" and "Theory._".
  1955 INCOMPATIBILITY.
  1956 
  1957 * Code generator: export_code without explicit file declaration prints
  1958 to standard output.  INCOMPATIBILITY.
  1959 
  1960 * Code generator: do not print function definitions for case
  1961 combinators any longer.
  1962 
  1963 * Code generator: simplification with rules determined with
  1964 src/Tools/Code/code_simp.ML and method "code_simp".
  1965 
  1966 * Code generator for records: more idiomatic representation of record
  1967 types.  Warning: records are not covered by ancient SML code
  1968 generation any longer.  INCOMPATIBILITY.  In cases of need, a suitable
  1969 rep_datatype declaration helps to succeed then:
  1970 
  1971   record 'a foo = ...
  1972   ...
  1973   rep_datatype foo_ext ...
  1974 
  1975 * Records: logical foundation type for records does not carry a
  1976 '_type' suffix any longer (obsolete due to authentic syntax).
  1977 INCOMPATIBILITY.
  1978 
  1979 * Quickcheck now by default uses exhaustive testing instead of random
  1980 testing.  Random testing can be invoked by "quickcheck [random]",
  1981 exhaustive testing by "quickcheck [exhaustive]".
  1982 
  1983 * Quickcheck instantiates polymorphic types with small finite
  1984 datatypes by default. This enables a simple execution mechanism to
  1985 handle quantifiers and function equality over the finite datatypes.
  1986 
  1987 * Quickcheck random generator has been renamed from "code" to
  1988 "random".  INCOMPATIBILITY.
  1989 
  1990 * Quickcheck now has a configurable time limit which is set to 30
  1991 seconds by default. This can be changed by adding [timeout = n] to the
  1992 quickcheck command. The time limit for Auto Quickcheck is still set
  1993 independently.
  1994 
  1995 * Quickcheck in locales considers interpretations of that locale for
  1996 counter example search.
  1997 
  1998 * Sledgehammer:
  1999   - Added "smt" and "remote_smt" provers based on the "smt" proof
  2000     method. See the Sledgehammer manual for details ("isabelle doc
  2001     sledgehammer").
  2002   - Renamed commands:
  2003     sledgehammer atp_info ~> sledgehammer running_provers
  2004     sledgehammer atp_kill ~> sledgehammer kill_provers
  2005     sledgehammer available_atps ~> sledgehammer available_provers
  2006     INCOMPATIBILITY.
  2007   - Renamed options:
  2008     sledgehammer [atps = ...] ~> sledgehammer [provers = ...]
  2009     sledgehammer [atp = ...] ~> sledgehammer [prover = ...]
  2010     sledgehammer [timeout = 77 s] ~> sledgehammer [timeout = 77]
  2011     (and "ms" and "min" are no longer supported)
  2012     INCOMPATIBILITY.
  2013 
  2014 * Nitpick:
  2015   - Renamed options:
  2016     nitpick [timeout = 77 s] ~> nitpick [timeout = 77]
  2017     nitpick [tac_timeout = 777 ms] ~> nitpick [tac_timeout = 0.777]
  2018     INCOMPATIBILITY.
  2019   - Added support for partial quotient types.
  2020   - Added local versions of the "Nitpick.register_xxx" functions.
  2021   - Added "whack" option.
  2022   - Allow registration of quotient types as codatatypes.
  2023   - Improved "merge_type_vars" option to merge more types.
  2024   - Removed unsound "fast_descrs" option.
  2025   - Added custom symmetry breaking for datatypes, making it possible to reach
  2026     higher cardinalities.
  2027   - Prevent the expansion of too large definitions.
  2028 
  2029 * Proof methods "metis" and "meson" now have configuration options
  2030 "meson_trace", "metis_trace", and "metis_verbose" that can be enabled
  2031 to diagnose these tools. E.g.
  2032 
  2033     using [[metis_trace = true]]
  2034 
  2035 * Auto Solve: Renamed "Auto Solve Direct".  The tool is now available
  2036 manually as command 'solve_direct'.
  2037 
  2038 * The default SMT solver Z3 must be enabled explicitly (due to
  2039 licensing issues) by setting the environment variable
  2040 Z3_NON_COMMERCIAL in etc/settings of the component, for example.  For
  2041 commercial applications, the SMT solver CVC3 is provided as fall-back;
  2042 changing the SMT solver is done via the configuration option
  2043 "smt_solver".
  2044 
  2045 * Remote SMT solvers need to be referred to by the "remote_" prefix,
  2046 i.e. "remote_cvc3" and "remote_z3".
  2047 
  2048 * Added basic SMT support for datatypes, records, and typedefs using
  2049 the oracle mode (no proofs).  Direct support of pairs has been dropped
  2050 in exchange (pass theorems fst_conv snd_conv pair_collapse to the SMT
  2051 support for a similar behavior).  Minor INCOMPATIBILITY.
  2052 
  2053 * Changed SMT configuration options:
  2054   - Renamed:
  2055     z3_proofs ~> smt_oracle (with inverted meaning)
  2056     z3_trace_assms ~> smt_trace_used_facts
  2057     INCOMPATIBILITY.
  2058   - Added:
  2059     smt_verbose
  2060     smt_random_seed
  2061     smt_datatypes
  2062     smt_infer_triggers
  2063     smt_monomorph_limit
  2064     cvc3_options
  2065     remote_cvc3_options
  2066     remote_z3_options
  2067     yices_options
  2068 
  2069 * Boogie output files (.b2i files) need to be declared in the theory
  2070 header.
  2071 
  2072 * Simplification procedure "list_to_set_comprehension" rewrites list
  2073 comprehensions applied to List.set to set comprehensions.  Occasional
  2074 INCOMPATIBILITY, may be deactivated like this:
  2075 
  2076   declare [[simproc del: list_to_set_comprehension]]
  2077 
  2078 * Removed old version of primrec package.  INCOMPATIBILITY.
  2079 
  2080 * Removed simplifier congruence rule of "prod_case", as has for long
  2081 been the case with "split".  INCOMPATIBILITY.
  2082 
  2083 * String.literal is a type, but not a datatype.  INCOMPATIBILITY.
  2084 
  2085 * Removed [split_format ... and ... and ...] version of
  2086 [split_format].  Potential INCOMPATIBILITY.
  2087 
  2088 * Predicate "sorted" now defined inductively, with nice induction
  2089 rules.  INCOMPATIBILITY: former sorted.simps now named sorted_simps.
  2090 
  2091 * Constant "contents" renamed to "the_elem", to free the generic name
  2092 contents for other uses.  INCOMPATIBILITY.
  2093 
  2094 * Renamed class eq and constant eq (for code generation) to class
  2095 equal and constant equal, plus renaming of related facts and various
  2096 tuning.  INCOMPATIBILITY.
  2097 
  2098 * Dropped type classes mult_mono and mult_mono1.  INCOMPATIBILITY.
  2099 
  2100 * Removed output syntax "'a ~=> 'b" for "'a => 'b option".
  2101 INCOMPATIBILITY.
  2102 
  2103 * Renamed theory Fset to Cset, type Fset.fset to Cset.set, in order to
  2104 avoid confusion with finite sets.  INCOMPATIBILITY.
  2105 
  2106 * Abandoned locales equiv, congruent and congruent2 for equivalence
  2107 relations.  INCOMPATIBILITY: use equivI rather than equiv_intro (same
  2108 for congruent(2)).
  2109 
  2110 * Some previously unqualified names have been qualified:
  2111 
  2112   types
  2113     bool ~> HOL.bool
  2114     nat ~> Nat.nat
  2115 
  2116   constants
  2117     Trueprop ~> HOL.Trueprop
  2118     True ~> HOL.True
  2119     False ~> HOL.False
  2120     op & ~> HOL.conj
  2121     op | ~> HOL.disj
  2122     op --> ~> HOL.implies
  2123     op = ~> HOL.eq
  2124     Not ~> HOL.Not
  2125     The ~> HOL.The
  2126     All ~> HOL.All
  2127     Ex ~> HOL.Ex
  2128     Ex1 ~> HOL.Ex1
  2129     Let ~> HOL.Let
  2130     If ~> HOL.If
  2131     Ball ~> Set.Ball
  2132     Bex ~> Set.Bex
  2133     Suc ~> Nat.Suc
  2134     Pair ~> Product_Type.Pair
  2135     fst ~> Product_Type.fst
  2136     snd ~> Product_Type.snd
  2137     curry ~> Product_Type.curry
  2138     op : ~> Set.member
  2139     Collect ~> Set.Collect
  2140 
  2141 INCOMPATIBILITY.
  2142 
  2143 * More canonical naming convention for some fundamental definitions:
  2144 
  2145     bot_bool_eq ~> bot_bool_def
  2146     top_bool_eq ~> top_bool_def
  2147     inf_bool_eq ~> inf_bool_def
  2148     sup_bool_eq ~> sup_bool_def
  2149     bot_fun_eq  ~> bot_fun_def
  2150     top_fun_eq  ~> top_fun_def
  2151     inf_fun_eq  ~> inf_fun_def
  2152     sup_fun_eq  ~> sup_fun_def
  2153 
  2154 INCOMPATIBILITY.
  2155 
  2156 * More stylized fact names:
  2157 
  2158   expand_fun_eq ~> fun_eq_iff
  2159   expand_set_eq ~> set_eq_iff
  2160   set_ext       ~> set_eqI
  2161   nat_number    ~> eval_nat_numeral
  2162 
  2163 INCOMPATIBILITY.
  2164 
  2165 * Refactoring of code-generation specific operations in theory List:
  2166 
  2167   constants
  2168     null ~> List.null
  2169 
  2170   facts
  2171     mem_iff ~> member_def
  2172     null_empty ~> null_def
  2173 
  2174 INCOMPATIBILITY.  Note that these were not supposed to be used
  2175 regularly unless for striking reasons; their main purpose was code
  2176 generation.
  2177 
  2178 Various operations from the Haskell prelude are used for generating
  2179 Haskell code.
  2180 
  2181 * Term "bij f" is now an abbreviation of "bij_betw f UNIV UNIV".  Term
  2182 "surj f" is now an abbreviation of "range f = UNIV".  The theorems
  2183 bij_def and surj_def are unchanged.  INCOMPATIBILITY.
  2184 
  2185 * Abolished some non-alphabetic type names: "prod" and "sum" replace
  2186 "*" and "+" respectively.  INCOMPATIBILITY.
  2187 
  2188 * Name "Plus" of disjoint sum operator "<+>" is now hidden.  Write
  2189 "Sum_Type.Plus" instead.
  2190 
  2191 * Constant "split" has been merged with constant "prod_case"; names of
  2192 ML functions, facts etc. involving split have been retained so far,
  2193 though.  INCOMPATIBILITY.
  2194 
  2195 * Dropped old infix syntax "_ mem _" for List.member; use "_ : set _"
  2196 instead.  INCOMPATIBILITY.
  2197 
  2198 * Removed lemma "Option.is_none_none" which duplicates "is_none_def".
  2199 INCOMPATIBILITY.
  2200 
  2201 * Former theory Library/Enum is now part of the HOL-Main image.
  2202 INCOMPATIBILITY: all constants of the Enum theory now have to be
  2203 referred to by its qualified name.
  2204 
  2205   enum    ~>  Enum.enum
  2206   nlists  ~>  Enum.nlists
  2207   product ~>  Enum.product
  2208 
  2209 * Theory Library/Monad_Syntax provides do-syntax for monad types.
  2210 Syntax in Library/State_Monad has been changed to avoid ambiguities.
  2211 INCOMPATIBILITY.
  2212 
  2213 * Theory Library/SetsAndFunctions has been split into
  2214 Library/Function_Algebras and Library/Set_Algebras; canonical names
  2215 for instance definitions for functions; various improvements.
  2216 INCOMPATIBILITY.
  2217 
  2218 * Theory Library/Multiset provides stable quicksort implementation of
  2219 sort_key.
  2220 
  2221 * Theory Library/Multiset: renamed empty_idemp ~> empty_neutral.
  2222 INCOMPATIBILITY.
  2223 
  2224 * Session Multivariate_Analysis: introduced a type class for euclidean
  2225 space.  Most theorems are now stated in terms of euclidean spaces
  2226 instead of finite cartesian products.
  2227 
  2228   types
  2229     real ^ 'n ~>  'a::real_vector
  2230               ~>  'a::euclidean_space
  2231               ~>  'a::ordered_euclidean_space
  2232         (depends on your needs)
  2233 
  2234   constants
  2235      _ $ _        ~> _ $$ _
  2236      \<chi> x. _  ~> \<chi>\<chi> x. _
  2237      CARD('n)     ~> DIM('a)
  2238 
  2239 Also note that the indices are now natural numbers and not from some
  2240 finite type. Finite cartesian products of euclidean spaces, products
  2241 of euclidean spaces the real and complex numbers are instantiated to
  2242 be euclidean_spaces.  INCOMPATIBILITY.
  2243 
  2244 * Session Probability: introduced pextreal as positive extended real
  2245 numbers.  Use pextreal as value for measures.  Introduce the
  2246 Radon-Nikodym derivative, product spaces and Fubini's theorem for
  2247 arbitrary sigma finite measures.  Introduces Lebesgue measure based on
  2248 the integral in Multivariate Analysis.  INCOMPATIBILITY.
  2249 
  2250 * Session Imperative_HOL: revamped, corrected dozens of inadequacies.
  2251 INCOMPATIBILITY.
  2252 
  2253 * Session SPARK (with image HOL-SPARK) provides commands to load and
  2254 prove verification conditions generated by the SPARK Ada program
  2255 verifier.  See also src/HOL/SPARK and src/HOL/SPARK/Examples.
  2256 
  2257 
  2258 *** HOL-Algebra ***
  2259 
  2260 * Theorems for additive ring operations (locale abelian_monoid and
  2261 descendants) are generated by interpretation from their multiplicative
  2262 counterparts.  Names (in particular theorem names) have the mandatory
  2263 qualifier 'add'.  Previous theorem names are redeclared for
  2264 compatibility.
  2265 
  2266 * Structure "int_ring" is now an abbreviation (previously a
  2267 definition).  This fits more natural with advanced interpretations.
  2268 
  2269 
  2270 *** HOLCF ***
  2271 
  2272 * The domain package now runs in definitional mode by default: The
  2273 former command 'new_domain' is now called 'domain'.  To use the domain
  2274 package in its original axiomatic mode, use 'domain (unsafe)'.
  2275 INCOMPATIBILITY.
  2276 
  2277 * The new class "domain" is now the default sort.  Class "predomain"
  2278 is an unpointed version of "domain". Theories can be updated by
  2279 replacing sort annotations as shown below.  INCOMPATIBILITY.
  2280 
  2281   'a::type ~> 'a::countable
  2282   'a::cpo  ~> 'a::predomain
  2283   'a::pcpo ~> 'a::domain
  2284 
  2285 * The old type class "rep" has been superseded by class "domain".
  2286 Accordingly, users of the definitional package must remove any
  2287 "default_sort rep" declarations.  INCOMPATIBILITY.
  2288 
  2289 * The domain package (definitional mode) now supports unpointed
  2290 predomain argument types, as long as they are marked 'lazy'. (Strict
  2291 arguments must be in class "domain".) For example, the following
  2292 domain definition now works:
  2293 
  2294   domain natlist = nil | cons (lazy "nat discr") (lazy "natlist")
  2295 
  2296 * Theory HOLCF/Library/HOL_Cpo provides cpo and predomain class
  2297 instances for types from main HOL: bool, nat, int, char, 'a + 'b,
  2298 'a option, and 'a list.  Additionally, it configures fixrec and the
  2299 domain package to work with these types.  For example:
  2300 
  2301   fixrec isInl :: "('a + 'b) u -> tr"
  2302     where "isInl$(up$(Inl x)) = TT" | "isInl$(up$(Inr y)) = FF"
  2303 
  2304   domain V = VFun (lazy "V -> V") | VCon (lazy "nat") (lazy "V list")
  2305 
  2306 * The "(permissive)" option of fixrec has been replaced with a
  2307 per-equation "(unchecked)" option. See
  2308 src/HOL/HOLCF/Tutorial/Fixrec_ex.thy for examples. INCOMPATIBILITY.
  2309 
  2310 * The "bifinite" class no longer fixes a constant "approx"; the class
  2311 now just asserts that such a function exists.  INCOMPATIBILITY.
  2312 
  2313 * Former type "alg_defl" has been renamed to "defl".  HOLCF no longer
  2314 defines an embedding of type 'a defl into udom by default; instances
  2315 of "bifinite" and "domain" classes are available in
  2316 src/HOL/HOLCF/Library/Defl_Bifinite.thy.
  2317 
  2318 * The syntax "REP('a)" has been replaced with "DEFL('a)".
  2319 
  2320 * The predicate "directed" has been removed.  INCOMPATIBILITY.
  2321 
  2322 * The type class "finite_po" has been removed.  INCOMPATIBILITY.
  2323 
  2324 * The function "cprod_map" has been renamed to "prod_map".
  2325 INCOMPATIBILITY.
  2326 
  2327 * The monadic bind operator on each powerdomain has new binder syntax
  2328 similar to sets, e.g. "\<Union>\<sharp>x\<in>xs. t" represents
  2329 "upper_bind\<cdot>xs\<cdot>(\<Lambda> x. t)".
  2330 
  2331 * The infix syntax for binary union on each powerdomain has changed
  2332 from e.g. "+\<sharp>" to "\<union>\<sharp>", for consistency with set
  2333 syntax.  INCOMPATIBILITY.
  2334 
  2335 * The constant "UU" has been renamed to "bottom".  The syntax "UU" is
  2336 still supported as an input translation.
  2337 
  2338 * Renamed some theorems (the original names are also still available).
  2339 
  2340   expand_fun_below   ~> fun_below_iff
  2341   below_fun_ext      ~> fun_belowI
  2342   expand_cfun_eq     ~> cfun_eq_iff
  2343   ext_cfun           ~> cfun_eqI
  2344   expand_cfun_below  ~> cfun_below_iff
  2345   below_cfun_ext     ~> cfun_belowI
  2346   cont2cont_Rep_CFun ~> cont2cont_APP
  2347 
  2348 * The Abs and Rep functions for various types have changed names.
  2349 Related theorem names have also changed to match. INCOMPATIBILITY.
  2350 
  2351   Rep_CFun  ~> Rep_cfun
  2352   Abs_CFun  ~> Abs_cfun
  2353   Rep_Sprod ~> Rep_sprod
  2354   Abs_Sprod ~> Abs_sprod
  2355   Rep_Ssum  ~> Rep_ssum
  2356   Abs_Ssum  ~> Abs_ssum
  2357 
  2358 * Lemmas with names of the form *_defined_iff or *_strict_iff have
  2359 been renamed to *_bottom_iff.  INCOMPATIBILITY.
  2360 
  2361 * Various changes to bisimulation/coinduction with domain package:
  2362 
  2363   - Definitions of "bisim" constants no longer mention definedness.
  2364   - With mutual recursion, "bisim" predicate is now curried.
  2365   - With mutual recursion, each type gets a separate coind theorem.
  2366   - Variable names in bisim_def and coinduct rules have changed.
  2367 
  2368 INCOMPATIBILITY.
  2369 
  2370 * Case combinators generated by the domain package for type "foo" are
  2371 now named "foo_case" instead of "foo_when".  INCOMPATIBILITY.
  2372 
  2373 * Several theorems have been renamed to more accurately reflect the
  2374 names of constants and types involved.  INCOMPATIBILITY.
  2375 
  2376   thelub_const    ~> lub_const
  2377   lub_const       ~> is_lub_const
  2378   thelubI         ~> lub_eqI
  2379   is_lub_lub      ~> is_lubD2
  2380   lubI            ~> is_lub_lub
  2381   unique_lub      ~> is_lub_unique
  2382   is_ub_lub       ~> is_lub_rangeD1
  2383   lub_bin_chain   ~> is_lub_bin_chain
  2384   lub_fun         ~> is_lub_fun
  2385   thelub_fun      ~> lub_fun
  2386   thelub_cfun     ~> lub_cfun
  2387   thelub_Pair     ~> lub_Pair
  2388   lub_cprod       ~> is_lub_prod
  2389   thelub_cprod    ~> lub_prod
  2390   minimal_cprod   ~> minimal_prod
  2391   inst_cprod_pcpo ~> inst_prod_pcpo
  2392   UU_I            ~> bottomI
  2393   compact_UU      ~> compact_bottom
  2394   deflation_UU    ~> deflation_bottom
  2395   finite_deflation_UU ~> finite_deflation_bottom
  2396 
  2397 * Many legacy theorem names have been discontinued.  INCOMPATIBILITY.
  2398 
  2399   sq_ord_less_eq_trans ~> below_eq_trans
  2400   sq_ord_eq_less_trans ~> eq_below_trans
  2401   refl_less            ~> below_refl
  2402   trans_less           ~> below_trans
  2403   antisym_less         ~> below_antisym
  2404   antisym_less_inverse ~> po_eq_conv [THEN iffD1]
  2405   box_less             ~> box_below
  2406   rev_trans_less       ~> rev_below_trans
  2407   not_less2not_eq      ~> not_below2not_eq
  2408   less_UU_iff          ~> below_UU_iff
  2409   flat_less_iff        ~> flat_below_iff
  2410   adm_less             ~> adm_below
  2411   adm_not_less         ~> adm_not_below
  2412   adm_compact_not_less ~> adm_compact_not_below
  2413   less_fun_def         ~> below_fun_def
  2414   expand_fun_less      ~> fun_below_iff
  2415   less_fun_ext         ~> fun_belowI
  2416   less_discr_def       ~> below_discr_def
  2417   discr_less_eq        ~> discr_below_eq
  2418   less_unit_def        ~> below_unit_def
  2419   less_cprod_def       ~> below_prod_def
  2420   prod_lessI           ~> prod_belowI
  2421   Pair_less_iff        ~> Pair_below_iff
  2422   fst_less_iff         ~> fst_below_iff
  2423   snd_less_iff         ~> snd_below_iff
  2424   expand_cfun_less     ~> cfun_below_iff
  2425   less_cfun_ext        ~> cfun_belowI
  2426   injection_less       ~> injection_below
  2427   less_up_def          ~> below_up_def
  2428   not_Iup_less         ~> not_Iup_below
  2429   Iup_less             ~> Iup_below
  2430   up_less              ~> up_below
  2431   Def_inject_less_eq   ~> Def_below_Def
  2432   Def_less_is_eq       ~> Def_below_iff
  2433   spair_less_iff       ~> spair_below_iff
  2434   less_sprod           ~> below_sprod
  2435   spair_less           ~> spair_below
  2436   sfst_less_iff        ~> sfst_below_iff
  2437   ssnd_less_iff        ~> ssnd_below_iff
  2438   fix_least_less       ~> fix_least_below
  2439   dist_less_one        ~> dist_below_one
  2440   less_ONE             ~> below_ONE
  2441   ONE_less_iff         ~> ONE_below_iff
  2442   less_sinlD           ~> below_sinlD
  2443   less_sinrD           ~> below_sinrD
  2444 
  2445 
  2446 *** FOL and ZF ***
  2447 
  2448 * All constant names are now qualified internally and use proper
  2449 identifiers, e.g. "IFOL.eq" instead of "op =".  INCOMPATIBILITY.
  2450 
  2451 
  2452 *** ML ***
  2453 
  2454 * Antiquotation @{assert} inlines a function bool -> unit that raises
  2455 Fail if the argument is false.  Due to inlining the source position of
  2456 failed assertions is included in the error output.
  2457 
  2458 * Discontinued antiquotation @{theory_ref}, which is obsolete since ML
  2459 text is in practice always evaluated with a stable theory checkpoint.
  2460 Minor INCOMPATIBILITY, use (Theory.check_thy @{theory}) instead.
  2461 
  2462 * Antiquotation @{theory A} refers to theory A from the ancestry of
  2463 the current context, not any accidental theory loader state as before.
  2464 Potential INCOMPATIBILITY, subtle change in semantics.
  2465 
  2466 * Syntax.pretty_priority (default 0) configures the required priority
  2467 of pretty-printed output and thus affects insertion of parentheses.
  2468 
  2469 * Syntax.default_root (default "any") configures the inner syntax
  2470 category (nonterminal symbol) for parsing of terms.
  2471 
  2472 * Former exception Library.UnequalLengths now coincides with
  2473 ListPair.UnequalLengths.
  2474 
  2475 * Renamed structure MetaSimplifier to Raw_Simplifier.  Note that the
  2476 main functionality is provided by structure Simplifier.
  2477 
  2478 * Renamed raw "explode" function to "raw_explode" to emphasize its
  2479 meaning.  Note that internally to Isabelle, Symbol.explode is used in
  2480 almost all situations.
  2481 
  2482 * Discontinued obsolete function sys_error and exception SYS_ERROR.
  2483 See implementation manual for further details on exceptions in
  2484 Isabelle/ML.
  2485 
  2486 * Renamed setmp_noncritical to Unsynchronized.setmp to emphasize its
  2487 meaning.
  2488 
  2489 * Renamed structure PureThy to Pure_Thy and moved most of its
  2490 operations to structure Global_Theory, to emphasize that this is
  2491 rarely-used global-only stuff.
  2492 
  2493 * Discontinued Output.debug.  Minor INCOMPATIBILITY, use plain writeln
  2494 instead (or tracing for high-volume output).
  2495 
  2496 * Configuration option show_question_marks only affects regular pretty
  2497 printing of types and terms, not raw Term.string_of_vname.
  2498 
  2499 * ML_Context.thm and ML_Context.thms are no longer pervasive.  Rare
  2500 INCOMPATIBILITY, superseded by static antiquotations @{thm} and
  2501 @{thms} for most purposes.
  2502 
  2503 * ML structure Unsynchronized is never opened, not even in Isar
  2504 interaction mode as before.  Old Unsynchronized.set etc. have been
  2505 discontinued -- use plain := instead.  This should be *rare* anyway,
  2506 since modern tools always work via official context data, notably
  2507 configuration options.
  2508 
  2509 * Parallel and asynchronous execution requires special care concerning
  2510 interrupts.  Structure Exn provides some convenience functions that
  2511 avoid working directly with raw Interrupt.  User code must not absorb
  2512 interrupts -- intermediate handling (for cleanup etc.) needs to be
  2513 followed by re-raising of the original exception.  Another common
  2514 source of mistakes are "handle _" patterns, which make the meaning of
  2515 the program subject to physical effects of the environment.
  2516 
  2517 
  2518 
  2519 New in Isabelle2009-2 (June 2010)
  2520 ---------------------------------
  2521 
  2522 *** General ***
  2523 
  2524 * Authentic syntax for *all* logical entities (type classes, type
  2525 constructors, term constants): provides simple and robust
  2526 correspondence between formal entities and concrete syntax.  Within
  2527 the parse tree / AST representations, "constants" are decorated by
  2528 their category (class, type, const) and spelled out explicitly with
  2529 their full internal name.
  2530 
  2531 Substantial INCOMPATIBILITY concerning low-level syntax declarations
  2532 and translations (translation rules and translation functions in ML).
  2533 Some hints on upgrading:
  2534 
  2535   - Many existing uses of 'syntax' and 'translations' can be replaced
  2536     by more modern 'type_notation', 'notation' and 'abbreviation',
  2537     which are independent of this issue.
  2538 
  2539   - 'translations' require markup within the AST; the term syntax
  2540     provides the following special forms:
  2541 
  2542       CONST c   -- produces syntax version of constant c from context
  2543       XCONST c  -- literally c, checked as constant from context
  2544       c         -- literally c, if declared by 'syntax'
  2545 
  2546     Plain identifiers are treated as AST variables -- occasionally the
  2547     system indicates accidental variables via the error "rhs contains
  2548     extra variables".
  2549 
  2550     Type classes and type constructors are marked according to their
  2551     concrete syntax.  Some old translations rules need to be written
  2552     for the "type" category, using type constructor application
  2553     instead of pseudo-term application of the default category
  2554     "logic".
  2555 
  2556   - 'parse_translation' etc. in ML may use the following
  2557     antiquotations:
  2558 
  2559       @{class_syntax c}   -- type class c within parse tree / AST
  2560       @{term_syntax c}    -- type constructor c within parse tree / AST
  2561       @{const_syntax c}   -- ML version of "CONST c" above
  2562       @{syntax_const c}   -- literally c (checked wrt. 'syntax' declarations)
  2563 
  2564   - Literal types within 'typed_print_translations', i.e. those *not*
  2565     represented as pseudo-terms are represented verbatim.  Use @{class
  2566     c} or @{type_name c} here instead of the above syntax
  2567     antiquotations.
  2568 
  2569 Note that old non-authentic syntax was based on unqualified base
  2570 names, so all of the above "constant" names would coincide.  Recall
  2571 that 'print_syntax' and ML_command "set Syntax.trace_ast" help to
  2572 diagnose syntax problems.
  2573 
  2574 * Type constructors admit general mixfix syntax, not just infix.
  2575 
  2576 * Concrete syntax may be attached to local entities without a proof
  2577 body, too.  This works via regular mixfix annotations for 'fix',
  2578 'def', 'obtain' etc. or via the explicit 'write' command, which is
  2579 similar to the 'notation' command in theory specifications.
  2580 
  2581 * Discontinued unnamed infix syntax (legacy feature for many years) --
  2582 need to specify constant name and syntax separately.  Internal ML
  2583 datatype constructors have been renamed from InfixName to Infix etc.
  2584 Minor INCOMPATIBILITY.
  2585 
  2586 * Schematic theorem statements need to be explicitly markup as such,
  2587 via commands 'schematic_lemma', 'schematic_theorem',
  2588 'schematic_corollary'.  Thus the relevance of the proof is made
  2589 syntactically clear, which impacts performance in a parallel or
  2590 asynchronous interactive environment.  Minor INCOMPATIBILITY.
  2591 
  2592 * Use of cumulative prems via "!" in some proof methods has been
  2593 discontinued (old legacy feature).
  2594 
  2595 * References 'trace_simp' and 'debug_simp' have been replaced by
  2596 configuration options stored in the context. Enabling tracing (the
  2597 case of debugging is similar) in proofs works via
  2598 
  2599   using [[trace_simp = true]]
  2600 
  2601 Tracing is then active for all invocations of the simplifier in
  2602 subsequent goal refinement steps. Tracing may also still be enabled or
  2603 disabled via the ProofGeneral settings menu.
  2604 
  2605 * Separate commands 'hide_class', 'hide_type', 'hide_const',
  2606 'hide_fact' replace the former 'hide' KIND command.  Minor
  2607 INCOMPATIBILITY.
  2608 
  2609 * Improved parallelism of proof term normalization: usedir -p2 -q0 is
  2610 more efficient than combinations with -q1 or -q2.
  2611 
  2612 
  2613 *** Pure ***
  2614 
  2615 * Proofterms record type-class reasoning explicitly, using the
  2616 "unconstrain" operation internally.  This eliminates all sort
  2617 constraints from a theorem and proof, introducing explicit
  2618 OFCLASS-premises.  On the proof term level, this operation is
  2619 automatically applied at theorem boundaries, such that closed proofs
  2620 are always free of sort constraints.  INCOMPATIBILITY for tools that
  2621 inspect proof terms.
  2622 
  2623 * Local theory specifications may depend on extra type variables that
  2624 are not present in the result type -- arguments TYPE('a) :: 'a itself
  2625 are added internally.  For example:
  2626 
  2627   definition unitary :: bool where "unitary = (ALL (x::'a) y. x = y)"
  2628 
  2629 * Predicates of locales introduced by classes carry a mandatory
  2630 "class" prefix.  INCOMPATIBILITY.
  2631 
  2632 * Vacuous class specifications observe default sort.  INCOMPATIBILITY.
  2633 
  2634 * Old 'axclass' command has been discontinued.  INCOMPATIBILITY, use
  2635 'class' instead.
  2636 
  2637 * Command 'code_reflect' allows to incorporate generated ML code into
  2638 runtime environment; replaces immature code_datatype antiquotation.
  2639 INCOMPATIBILITY.
  2640 
  2641 * Code generator: simple concept for abstract datatypes obeying
  2642 invariants.
  2643 
  2644 * Code generator: details of internal data cache have no impact on the
  2645 user space functionality any longer.
  2646 
  2647 * Methods "unfold_locales" and "intro_locales" ignore non-locale
  2648 subgoals.  This is more appropriate for interpretations with 'where'.
  2649 INCOMPATIBILITY.
  2650 
  2651 * Command 'example_proof' opens an empty proof body.  This allows to
  2652 experiment with Isar, without producing any persistent result.
  2653 
  2654 * Commands 'type_notation' and 'no_type_notation' declare type syntax
  2655 within a local theory context, with explicit checking of the
  2656 constructors involved (in contrast to the raw 'syntax' versions).
  2657 
  2658 * Commands 'types' and 'typedecl' now work within a local theory
  2659 context -- without introducing dependencies on parameters or
  2660 assumptions, which is not possible in Isabelle/Pure.
  2661 
  2662 * Command 'defaultsort' has been renamed to 'default_sort', it works
  2663 within a local theory context.  Minor INCOMPATIBILITY.
  2664 
  2665 
  2666 *** HOL ***
  2667 
  2668 * Command 'typedef' now works within a local theory context -- without
  2669 introducing dependencies on parameters or assumptions, which is not
  2670 possible in Isabelle/Pure/HOL.  Note that the logical environment may
  2671 contain multiple interpretations of local typedefs (with different
  2672 non-emptiness proofs), even in a global theory context.
  2673 
  2674 * New package for quotient types.  Commands 'quotient_type' and
  2675 'quotient_definition' may be used for defining types and constants by
  2676 quotient constructions.  An example is the type of integers created by
  2677 quotienting pairs of natural numbers:
  2678 
  2679   fun
  2680     intrel :: "(nat * nat) => (nat * nat) => bool"
  2681   where
  2682     "intrel (x, y) (u, v) = (x + v = u + y)"
  2683 
  2684   quotient_type int = "nat * nat" / intrel
  2685     by (auto simp add: equivp_def expand_fun_eq)
  2686 
  2687   quotient_definition
  2688     "0::int" is "(0::nat, 0::nat)"
  2689 
  2690 The method "lifting" can be used to lift of theorems from the
  2691 underlying "raw" type to the quotient type.  The example
  2692 src/HOL/Quotient_Examples/FSet.thy includes such a quotient
  2693 construction and provides a reasoning infrastructure for finite sets.
  2694 
  2695 * Renamed Library/Quotient.thy to Library/Quotient_Type.thy to avoid
  2696 clash with new theory Quotient in Main HOL.
  2697 
  2698 * Moved the SMT binding into the main HOL session, eliminating
  2699 separate HOL-SMT session.
  2700 
  2701 * List membership infix mem operation is only an input abbreviation.
  2702 INCOMPATIBILITY.
  2703 
  2704 * Theory Library/Word.thy has been removed.  Use library Word/Word.thy
  2705 for future developements; former Library/Word.thy is still present in
  2706 the AFP entry RSAPPS.
  2707 
  2708 * Theorem Int.int_induct renamed to Int.int_of_nat_induct and is no
  2709 longer shadowed.  INCOMPATIBILITY.
  2710 
  2711 * Dropped theorem duplicate comp_arith; use semiring_norm instead.
  2712 INCOMPATIBILITY.
  2713 
  2714 * Dropped theorem RealPow.real_sq_order; use power2_le_imp_le instead.
  2715 INCOMPATIBILITY.
  2716 
  2717 * Dropped normalizing_semiring etc; use the facts in semiring classes
  2718 instead.  INCOMPATIBILITY.
  2719 
  2720 * Dropped several real-specific versions of lemmas about floor and
  2721 ceiling; use the generic lemmas from theory "Archimedean_Field"
  2722 instead.  INCOMPATIBILITY.
  2723 
  2724   floor_number_of_eq         ~> floor_number_of
  2725   le_floor_eq_number_of      ~> number_of_le_floor
  2726   le_floor_eq_zero           ~> zero_le_floor
  2727   le_floor_eq_one            ~> one_le_floor
  2728   floor_less_eq_number_of    ~> floor_less_number_of
  2729   floor_less_eq_zero         ~> floor_less_zero
  2730   floor_less_eq_one          ~> floor_less_one
  2731   less_floor_eq_number_of    ~> number_of_less_floor
  2732   less_floor_eq_zero         ~> zero_less_floor
  2733   less_floor_eq_one          ~> one_less_floor
  2734   floor_le_eq_number_of      ~> floor_le_number_of
  2735   floor_le_eq_zero           ~> floor_le_zero
  2736   floor_le_eq_one            ~> floor_le_one
  2737   floor_subtract_number_of   ~> floor_diff_number_of
  2738   floor_subtract_one         ~> floor_diff_one
  2739   ceiling_number_of_eq       ~> ceiling_number_of
  2740   ceiling_le_eq_number_of    ~> ceiling_le_number_of
  2741   ceiling_le_zero_eq         ~> ceiling_le_zero
  2742   ceiling_le_eq_one          ~> ceiling_le_one
  2743   less_ceiling_eq_number_of  ~> number_of_less_ceiling
  2744   less_ceiling_eq_zero       ~> zero_less_ceiling
  2745   less_ceiling_eq_one        ~> one_less_ceiling
  2746   ceiling_less_eq_number_of  ~> ceiling_less_number_of
  2747   ceiling_less_eq_zero       ~> ceiling_less_zero
  2748   ceiling_less_eq_one        ~> ceiling_less_one
  2749   le_ceiling_eq_number_of    ~> number_of_le_ceiling
  2750   le_ceiling_eq_zero         ~> zero_le_ceiling
  2751   le_ceiling_eq_one          ~> one_le_ceiling
  2752   ceiling_subtract_number_of ~> ceiling_diff_number_of
  2753   ceiling_subtract_one       ~> ceiling_diff_one
  2754 
  2755 * Theory "Finite_Set": various folding_XXX locales facilitate the
  2756 application of the various fold combinators on finite sets.
  2757 
  2758 * Library theory "RBT" renamed to "RBT_Impl"; new library theory "RBT"
  2759 provides abstract red-black tree type which is backed by "RBT_Impl" as
  2760 implementation.  INCOMPATIBILTY.
  2761 
  2762 * Theory Library/Coinductive_List has been removed -- superseded by
  2763 AFP/thys/Coinductive.
  2764 
  2765 * Theory PReal, including the type "preal" and related operations, has
  2766 been removed.  INCOMPATIBILITY.
  2767 
  2768 * Real: new development using Cauchy Sequences.
  2769 
  2770 * Split off theory "Big_Operators" containing setsum, setprod,
  2771 Inf_fin, Sup_fin, Min, Max from theory Finite_Set.  INCOMPATIBILITY.
  2772 
  2773 * Theory "Rational" renamed to "Rat", for consistency with "Nat",
  2774 "Int" etc.  INCOMPATIBILITY.
  2775 
  2776 * Constant Rat.normalize needs to be qualified.  INCOMPATIBILITY.
  2777 
  2778 * New set of rules "ac_simps" provides combined assoc / commute
  2779 rewrites for all interpretations of the appropriate generic locales.
  2780 
  2781 * Renamed theory "OrderedGroup" to "Groups" and split theory
  2782 "Ring_and_Field" into theories "Rings" and "Fields"; for more
  2783 appropriate and more consistent names suitable for name prefixes
  2784 within the HOL theories.  INCOMPATIBILITY.
  2785 
  2786 * Some generic constants have been put to appropriate theories:
  2787   - less_eq, less: Orderings
  2788   - zero, one, plus, minus, uminus, times, abs, sgn: Groups
  2789   - inverse, divide: Rings
  2790 INCOMPATIBILITY.
  2791 
  2792 * More consistent naming of type classes involving orderings (and
  2793 lattices):
  2794 
  2795     lower_semilattice                   ~> semilattice_inf
  2796     upper_semilattice                   ~> semilattice_sup
  2797 
  2798     dense_linear_order                  ~> dense_linorder
  2799 
  2800     pordered_ab_group_add               ~> ordered_ab_group_add
  2801     pordered_ab_group_add_abs           ~> ordered_ab_group_add_abs
  2802     pordered_ab_semigroup_add           ~> ordered_ab_semigroup_add
  2803     pordered_ab_semigroup_add_imp_le    ~> ordered_ab_semigroup_add_imp_le
  2804     pordered_cancel_ab_semigroup_add    ~> ordered_cancel_ab_semigroup_add
  2805     pordered_cancel_comm_semiring       ~> ordered_cancel_comm_semiring
  2806     pordered_cancel_semiring            ~> ordered_cancel_semiring
  2807     pordered_comm_monoid_add            ~> ordered_comm_monoid_add
  2808     pordered_comm_ring                  ~> ordered_comm_ring
  2809     pordered_comm_semiring              ~> ordered_comm_semiring
  2810     pordered_ring                       ~> ordered_ring
  2811     pordered_ring_abs                   ~> ordered_ring_abs
  2812     pordered_semiring                   ~> ordered_semiring
  2813 
  2814     ordered_ab_group_add                ~> linordered_ab_group_add
  2815     ordered_ab_semigroup_add            ~> linordered_ab_semigroup_add
  2816     ordered_cancel_ab_semigroup_add     ~> linordered_cancel_ab_semigroup_add
  2817     ordered_comm_semiring_strict        ~> linordered_comm_semiring_strict
  2818     ordered_field                       ~> linordered_field
  2819     ordered_field_no_lb                 ~> linordered_field_no_lb
  2820     ordered_field_no_ub                 ~> linordered_field_no_ub
  2821     ordered_field_dense_linear_order    ~> dense_linordered_field
  2822     ordered_idom                        ~> linordered_idom
  2823     ordered_ring                        ~> linordered_ring
  2824     ordered_ring_le_cancel_factor       ~> linordered_ring_le_cancel_factor
  2825     ordered_ring_less_cancel_factor     ~> linordered_ring_less_cancel_factor
  2826     ordered_ring_strict                 ~> linordered_ring_strict
  2827     ordered_semidom                     ~> linordered_semidom
  2828     ordered_semiring                    ~> linordered_semiring
  2829     ordered_semiring_1                  ~> linordered_semiring_1
  2830     ordered_semiring_1_strict           ~> linordered_semiring_1_strict
  2831     ordered_semiring_strict             ~> linordered_semiring_strict
  2832 
  2833   The following slightly odd type classes have been moved to a
  2834   separate theory Library/Lattice_Algebras:
  2835 
  2836     lordered_ab_group_add               ~> lattice_ab_group_add
  2837     lordered_ab_group_add_abs           ~> lattice_ab_group_add_abs
  2838     lordered_ab_group_add_meet          ~> semilattice_inf_ab_group_add
  2839     lordered_ab_group_add_join          ~> semilattice_sup_ab_group_add
  2840     lordered_ring                       ~> lattice_ring
  2841 
  2842 INCOMPATIBILITY.
  2843 
  2844 * Refined field classes:
  2845   - classes division_ring_inverse_zero, field_inverse_zero,
  2846     linordered_field_inverse_zero include rule inverse 0 = 0 --
  2847     subsumes former division_by_zero class;
  2848   - numerous lemmas have been ported from field to division_ring.
  2849 INCOMPATIBILITY.
  2850 
  2851 * Refined algebra theorem collections:
  2852   - dropped theorem group group_simps, use algebra_simps instead;
  2853   - dropped theorem group ring_simps, use field_simps instead;
  2854   - proper theorem collection field_simps subsumes former theorem
  2855     groups field_eq_simps and field_simps;
  2856   - dropped lemma eq_minus_self_iff which is a duplicate for
  2857     equal_neg_zero.
  2858 INCOMPATIBILITY.
  2859 
  2860 * Theory Finite_Set and List: some lemmas have been generalized from
  2861 sets to lattices:
  2862 
  2863   fun_left_comm_idem_inter      ~> fun_left_comm_idem_inf
  2864   fun_left_comm_idem_union      ~> fun_left_comm_idem_sup
  2865   inter_Inter_fold_inter        ~> inf_Inf_fold_inf
  2866   union_Union_fold_union        ~> sup_Sup_fold_sup
  2867   Inter_fold_inter              ~> Inf_fold_inf
  2868   Union_fold_union              ~> Sup_fold_sup
  2869   inter_INTER_fold_inter        ~> inf_INFI_fold_inf
  2870   union_UNION_fold_union        ~> sup_SUPR_fold_sup
  2871   INTER_fold_inter              ~> INFI_fold_inf
  2872   UNION_fold_union              ~> SUPR_fold_sup
  2873 
  2874 * Theory "Complete_Lattice": lemmas top_def and bot_def have been
  2875 replaced by the more convenient lemmas Inf_empty and Sup_empty.
  2876 Dropped lemmas Inf_insert_simp and Sup_insert_simp, which are subsumed
  2877 by Inf_insert and Sup_insert.  Lemmas Inf_UNIV and Sup_UNIV replace
  2878 former Inf_Univ and Sup_Univ.  Lemmas inf_top_right and sup_bot_right
  2879 subsume inf_top and sup_bot respectively.  INCOMPATIBILITY.
  2880 
  2881 * Reorganized theory Multiset: swapped notation of pointwise and
  2882 multiset order:
  2883 
  2884   - pointwise ordering is instance of class order with standard syntax
  2885     <= and <;
  2886   - multiset ordering has syntax <=# and <#; partial order properties
  2887     are provided by means of interpretation with prefix
  2888     multiset_order;
  2889   - less duplication, less historical organization of sections,
  2890     conversion from associations lists to multisets, rudimentary code
  2891     generation;
  2892   - use insert_DiffM2 [symmetric] instead of elem_imp_eq_diff_union,
  2893     if needed.
  2894 
  2895 Renamed:
  2896 
  2897   multiset_eq_conv_count_eq  ~>  multiset_ext_iff
  2898   multi_count_ext  ~>  multiset_ext
  2899   diff_union_inverse2  ~>  diff_union_cancelR
  2900 
  2901 INCOMPATIBILITY.
  2902 
  2903 * Theory Permutation: replaced local "remove" by List.remove1.
  2904 
  2905 * Code generation: ML and OCaml code is decorated with signatures.
  2906 
  2907 * Theory List: added transpose.
  2908 
  2909 * Library/Nat_Bijection.thy is a collection of bijective functions
  2910 between nat and other types, which supersedes the older libraries
  2911 Library/Nat_Int_Bij.thy and HOLCF/NatIso.thy.  INCOMPATIBILITY.
  2912 
  2913   Constants:
  2914   Nat_Int_Bij.nat2_to_nat         ~> prod_encode
  2915   Nat_Int_Bij.nat_to_nat2         ~> prod_decode
  2916   Nat_Int_Bij.int_to_nat_bij      ~> int_encode
  2917   Nat_Int_Bij.nat_to_int_bij      ~> int_decode
  2918   Countable.pair_encode           ~> prod_encode
  2919   NatIso.prod2nat                 ~> prod_encode
  2920   NatIso.nat2prod                 ~> prod_decode
  2921   NatIso.sum2nat                  ~> sum_encode
  2922   NatIso.nat2sum                  ~> sum_decode
  2923   NatIso.list2nat                 ~> list_encode
  2924   NatIso.nat2list                 ~> list_decode
  2925   NatIso.set2nat                  ~> set_encode
  2926   NatIso.nat2set                  ~> set_decode
  2927 
  2928   Lemmas:
  2929   Nat_Int_Bij.bij_nat_to_int_bij  ~> bij_int_decode
  2930   Nat_Int_Bij.nat2_to_nat_inj     ~> inj_prod_encode
  2931   Nat_Int_Bij.nat2_to_nat_surj    ~> surj_prod_encode
  2932   Nat_Int_Bij.nat_to_nat2_inj     ~> inj_prod_decode
  2933   Nat_Int_Bij.nat_to_nat2_surj    ~> surj_prod_decode
  2934   Nat_Int_Bij.i2n_n2i_id          ~> int_encode_inverse
  2935   Nat_Int_Bij.n2i_i2n_id          ~> int_decode_inverse
  2936   Nat_Int_Bij.surj_nat_to_int_bij ~> surj_int_encode
  2937   Nat_Int_Bij.surj_int_to_nat_bij ~> surj_int_decode
  2938   Nat_Int_Bij.inj_nat_to_int_bij  ~> inj_int_encode
  2939   Nat_Int_Bij.inj_int_to_nat_bij  ~> inj_int_decode
  2940   Nat_Int_Bij.bij_nat_to_int_bij  ~> bij_int_encode
  2941   Nat_Int_Bij.bij_int_to_nat_bij  ~> bij_int_decode
  2942 
  2943 * Sledgehammer:
  2944   - Renamed ATP commands:
  2945     atp_info     ~> sledgehammer running_atps
  2946     atp_kill     ~> sledgehammer kill_atps
  2947     atp_messages ~> sledgehammer messages
  2948     atp_minimize ~> sledgehammer minimize
  2949     print_atps   ~> sledgehammer available_atps
  2950     INCOMPATIBILITY.
  2951   - Added user's manual ("isabelle doc sledgehammer").
  2952   - Added option syntax and "sledgehammer_params" to customize
  2953     Sledgehammer's behavior.  See the manual for details.
  2954   - Modified the Isar proof reconstruction code so that it produces
  2955     direct proofs rather than proofs by contradiction.  (This feature
  2956     is still experimental.)
  2957   - Made Isar proof reconstruction work for SPASS, remote ATPs, and in
  2958     full-typed mode.
  2959   - Added support for TPTP syntax for SPASS via the "spass_tptp" ATP.
  2960 
  2961 * Nitpick:
  2962   - Added and implemented "binary_ints" and "bits" options.
  2963   - Added "std" option and implemented support for nonstandard models.
  2964   - Added and implemented "finitize" option to improve the precision
  2965     of infinite datatypes based on a monotonicity analysis.
  2966   - Added support for quotient types.
  2967   - Added support for "specification" and "ax_specification"
  2968     constructs.
  2969   - Added support for local definitions (for "function" and
  2970     "termination" proofs).
  2971   - Added support for term postprocessors.
  2972   - Optimized "Multiset.multiset" and "FinFun.finfun".
  2973   - Improved efficiency of "destroy_constrs" optimization.
  2974   - Fixed soundness bugs related to "destroy_constrs" optimization and
  2975     record getters.
  2976   - Fixed soundness bug related to higher-order constructors.
  2977   - Fixed soundness bug when "full_descrs" is enabled.
  2978   - Improved precision of set constructs.
  2979   - Added "atoms" option.
  2980   - Added cache to speed up repeated Kodkod invocations on the same
  2981     problems.
  2982   - Renamed "MiniSatJNI", "zChaffJNI", "BerkMinAlloy", and
  2983     "SAT4JLight" to "MiniSat_JNI", "zChaff_JNI", "BerkMin_Alloy", and
  2984     "SAT4J_Light".  INCOMPATIBILITY.
  2985   - Removed "skolemize", "uncurry", "sym_break", "flatten_prop",
  2986     "sharing_depth", and "show_skolems" options.  INCOMPATIBILITY.
  2987   - Removed "nitpick_intro" attribute.  INCOMPATIBILITY.
  2988 
  2989 * Method "induct" now takes instantiations of the form t, where t is not
  2990   a variable, as a shorthand for "x == t", where x is a fresh variable.
  2991   If this is not intended, t has to be enclosed in parentheses.
  2992   By default, the equalities generated by definitional instantiations
  2993   are pre-simplified, which may cause parameters of inductive cases
  2994   to disappear, or may even delete some of the inductive cases.
  2995   Use "induct (no_simp)" instead of "induct" to restore the old
  2996   behaviour. The (no_simp) option is also understood by the "cases"
  2997   and "nominal_induct" methods, which now perform pre-simplification, too.
  2998   INCOMPATIBILITY.
  2999 
  3000 
  3001 *** HOLCF ***
  3002 
  3003 * Variable names in lemmas generated by the domain package have
  3004 changed; the naming scheme is now consistent with the HOL datatype
  3005 package.  Some proof scripts may be affected, INCOMPATIBILITY.
  3006 
  3007 * The domain package no longer defines the function "foo_copy" for
  3008 recursive domain "foo".  The reach lemma is now stated directly in
  3009 terms of "foo_take".  Lemmas and proofs that mention "foo_copy" must
  3010 be reformulated in terms of "foo_take", INCOMPATIBILITY.
  3011 
  3012 * Most definedness lemmas generated by the domain package (previously
  3013 of the form "x ~= UU ==> foo$x ~= UU") now have an if-and-only-if form
  3014 like "foo$x = UU <-> x = UU", which works better as a simp rule.
  3015 Proofs that used definedness lemmas as intro rules may break,
  3016 potential INCOMPATIBILITY.
  3017 
  3018 * Induction and casedist rules generated by the domain package now
  3019 declare proper case_names (one called "bottom", and one named for each
  3020 constructor).  INCOMPATIBILITY.
  3021 
  3022 * For mutually-recursive domains, separate "reach" and "take_lemma"
  3023 rules are generated for each domain, INCOMPATIBILITY.
  3024 
  3025   foo_bar.reach       ~> foo.reach  bar.reach
  3026   foo_bar.take_lemmas ~> foo.take_lemma  bar.take_lemma
  3027 
  3028 * Some lemmas generated by the domain package have been renamed for
  3029 consistency with the datatype package, INCOMPATIBILITY.
  3030 
  3031   foo.ind        ~> foo.induct
  3032   foo.finite_ind ~> foo.finite_induct
  3033   foo.coind      ~> foo.coinduct
  3034   foo.casedist   ~> foo.exhaust
  3035   foo.exhaust    ~> foo.nchotomy
  3036 
  3037 * For consistency with other definition packages, the fixrec package
  3038 now generates qualified theorem names, INCOMPATIBILITY.
  3039 
  3040   foo_simps  ~> foo.simps
  3041   foo_unfold ~> foo.unfold
  3042   foo_induct ~> foo.induct
  3043 
  3044 * The "fixrec_simp" attribute has been removed.  The "fixrec_simp"
  3045 method and internal fixrec proofs now use the default simpset instead.
  3046 INCOMPATIBILITY.
  3047 
  3048 * The "contlub" predicate has been removed.  Proof scripts should use
  3049 lemma contI2 in place of monocontlub2cont, INCOMPATIBILITY.
  3050 
  3051 * The "admw" predicate has been removed, INCOMPATIBILITY.
  3052 
  3053 * The constants cpair, cfst, and csnd have been removed in favor of
  3054 Pair, fst, and snd from Isabelle/HOL, INCOMPATIBILITY.
  3055 
  3056 
  3057 *** ML ***
  3058 
  3059 * Antiquotations for basic formal entities:
  3060 
  3061     @{class NAME}         -- type class
  3062     @{class_syntax NAME}  -- syntax representation of the above
  3063 
  3064     @{type_name NAME}     -- logical type
  3065     @{type_abbrev NAME}   -- type abbreviation
  3066     @{nonterminal NAME}   -- type of concrete syntactic category
  3067     @{type_syntax NAME}   -- syntax representation of any of the above
  3068 
  3069     @{const_name NAME}    -- logical constant (INCOMPATIBILITY)
  3070     @{const_abbrev NAME}  -- abbreviated constant
  3071     @{const_syntax NAME}  -- syntax representation of any of the above
  3072 
  3073 * Antiquotation @{syntax_const NAME} ensures that NAME refers to a raw
  3074 syntax constant (cf. 'syntax' command).
  3075 
  3076 * Antiquotation @{make_string} inlines a function to print arbitrary
  3077 values similar to the ML toplevel.  The result is compiler dependent
  3078 and may fall back on "?" in certain situations.
  3079 
  3080 * Diagnostic commands 'ML_val' and 'ML_command' may refer to
  3081 antiquotations @{Isar.state} and @{Isar.goal}.  This replaces impure
  3082 Isar.state() and Isar.goal(), which belong to the old TTY loop and do
  3083 not work with the asynchronous Isar document model.
  3084 
  3085 * Configuration options now admit dynamic default values, depending on
  3086 the context or even global references.
  3087 
  3088 * SHA1.digest digests strings according to SHA-1 (see RFC 3174).  It
  3089 uses an efficient external library if available (for Poly/ML).
  3090 
  3091 * Renamed some important ML structures, while keeping the old names
  3092 for some time as aliases within the structure Legacy:
  3093 
  3094   OuterKeyword  ~>  Keyword
  3095   OuterLex      ~>  Token
  3096   OuterParse    ~>  Parse
  3097   OuterSyntax   ~>  Outer_Syntax
  3098   PrintMode     ~>  Print_Mode
  3099   SpecParse     ~>  Parse_Spec
  3100   ThyInfo       ~>  Thy_Info
  3101   ThyLoad       ~>  Thy_Load
  3102   ThyOutput     ~>  Thy_Output
  3103   TypeInfer     ~>  Type_Infer
  3104 
  3105 Note that "open Legacy" simplifies porting of sources, but forgetting
  3106 to remove it again will complicate porting again in the future.
  3107 
  3108 * Most operations that refer to a global context are named
  3109 accordingly, e.g. Simplifier.global_context or
  3110 ProofContext.init_global.  There are some situations where a global
  3111 context actually works, but under normal circumstances one needs to
  3112 pass the proper local context through the code!
  3113 
  3114 * Discontinued old TheoryDataFun with its copy/init operation -- data
  3115 needs to be pure.  Functor Theory_Data_PP retains the traditional
  3116 Pretty.pp argument to merge, which is absent in the standard
  3117 Theory_Data version.
  3118 
  3119 * Sorts.certify_sort and derived "cert" operations for types and terms
  3120 no longer minimize sorts.  Thus certification at the boundary of the
  3121 inference kernel becomes invariant under addition of class relations,
  3122 which is an important monotonicity principle.  Sorts are now minimized
  3123 in the syntax layer only, at the boundary between the end-user and the
  3124 system.  Subtle INCOMPATIBILITY, may have to use Sign.minimize_sort
  3125 explicitly in rare situations.
  3126 
  3127 * Renamed old-style Drule.standard to Drule.export_without_context, to
  3128 emphasize that this is in no way a standard operation.
  3129 INCOMPATIBILITY.
  3130 
  3131 * Subgoal.FOCUS (and variants): resulting goal state is normalized as
  3132 usual for resolution.  Rare INCOMPATIBILITY.
  3133 
  3134 * Renamed varify/unvarify operations to varify_global/unvarify_global
  3135 to emphasize that these only work in a global situation (which is
  3136 quite rare).
  3137 
  3138 * Curried take and drop in library.ML; negative length is interpreted
  3139 as infinity (as in chop).  Subtle INCOMPATIBILITY.
  3140 
  3141 * Proof terms: type substitutions on proof constants now use canonical
  3142 order of type variables.  INCOMPATIBILITY for tools working with proof
  3143 terms.
  3144 
  3145 * Raw axioms/defs may no longer carry sort constraints, and raw defs
  3146 may no longer carry premises.  User-level specifications are
  3147 transformed accordingly by Thm.add_axiom/add_def.
  3148 
  3149 
  3150 *** System ***
  3151 
  3152 * Discontinued special HOL_USEDIR_OPTIONS for the main HOL image;
  3153 ISABELLE_USEDIR_OPTIONS applies uniformly to all sessions.  Note that
  3154 proof terms are enabled unconditionally in the new HOL-Proofs image.
  3155 
  3156 * Discontinued old ISABELLE and ISATOOL environment settings (legacy
  3157 feature since Isabelle2009).  Use ISABELLE_PROCESS and ISABELLE_TOOL,
  3158 respectively.
  3159 
  3160 * Old lib/scripts/polyml-platform is superseded by the
  3161 ISABELLE_PLATFORM setting variable, which defaults to the 32 bit
  3162 variant, even on a 64 bit machine.  The following example setting
  3163 prefers 64 bit if available:
  3164 
  3165   ML_PLATFORM="${ISABELLE_PLATFORM64:-$ISABELLE_PLATFORM}"
  3166 
  3167 * The preliminary Isabelle/jEdit application demonstrates the emerging
  3168 Isabelle/Scala layer for advanced prover interaction and integration.
  3169 See src/Tools/jEdit or "isabelle jedit" provided by the properly built
  3170 component.
  3171 
  3172 * "IsabelleText" is a Unicode font derived from Bitstream Vera Mono
  3173 and Bluesky TeX fonts.  It provides the usual Isabelle symbols,
  3174 similar to the default assignment of the document preparation system
  3175 (cf. isabellesym.sty).  The Isabelle/Scala class Isabelle_System
  3176 provides some operations for direct access to the font without asking
  3177 the user for manual installation.
  3178 
  3179 
  3180 
  3181 New in Isabelle2009-1 (December 2009)
  3182 -------------------------------------
  3183 
  3184 *** General ***
  3185 
  3186 * Discontinued old form of "escaped symbols" such as \\<forall>.  Only
  3187 one backslash should be used, even in ML sources.
  3188 
  3189 
  3190 *** Pure ***
  3191 
  3192 * Locale interpretation propagates mixins along the locale hierarchy.
  3193 The currently only available mixins are the equations used to map
  3194 local definitions to terms of the target domain of an interpretation.
  3195 
  3196 * Reactivated diagnostic command 'print_interps'.  Use "print_interps
  3197 loc" to print all interpretations of locale "loc" in the theory.
  3198 Interpretations in proofs are not shown.
  3199 
  3200 * Thoroughly revised locales tutorial.  New section on conditional
  3201 interpretation.
  3202 
  3203 * On instantiation of classes, remaining undefined class parameters
  3204 are formally declared.  INCOMPATIBILITY.
  3205 
  3206 
  3207 *** Document preparation ***
  3208 
  3209 * New generalized style concept for printing terms: @{foo (style) ...}
  3210 instead of @{foo_style style ...}  (old form is still retained for
  3211 backward compatibility).  Styles can be also applied for
  3212 antiquotations prop, term_type and typeof.
  3213 
  3214 
  3215 *** HOL ***
  3216 
  3217 * New proof method "smt" for a combination of first-order logic with
  3218 equality, linear and nonlinear (natural/integer/real) arithmetic, and
  3219 fixed-size bitvectors; there is also basic support for higher-order
  3220 features (esp. lambda abstractions).  It is an incomplete decision
  3221 procedure based on external SMT solvers using the oracle mechanism;
  3222 for the SMT solver Z3, this method is proof-producing.  Certificates
  3223 are provided to avoid calling the external solvers solely for
  3224 re-checking proofs.  Due to a remote SMT service there is no need for
  3225 installing SMT solvers locally.  See src/HOL/SMT.
  3226 
  3227 * New commands to load and prove verification conditions generated by
  3228 the Boogie program verifier or derived systems (e.g. the Verifying C
  3229 Compiler (VCC) or Spec#).  See src/HOL/Boogie.
  3230 
  3231 * New counterexample generator tool 'nitpick' based on the Kodkod
  3232 relational model finder.  See src/HOL/Tools/Nitpick and
  3233 src/HOL/Nitpick_Examples.
  3234 
  3235 * New commands 'code_pred' and 'values' to invoke the predicate
  3236 compiler and to enumerate values of inductive predicates.
  3237 
  3238 * A tabled implementation of the reflexive transitive closure.
  3239 
  3240 * New implementation of quickcheck uses generic code generator;
  3241 default generators are provided for all suitable HOL types, records
  3242 and datatypes.  Old quickcheck can be re-activated importing theory
  3243 Library/SML_Quickcheck.
  3244 
  3245 * New testing tool Mirabelle for automated proof tools.  Applies
  3246 several tools and tactics like sledgehammer, metis, or quickcheck, to
  3247 every proof step in a theory.  To be used in batch mode via the
  3248 "mirabelle" utility.
  3249 
  3250 * New proof method "sos" (sum of squares) for nonlinear real
  3251 arithmetic (originally due to John Harison). It requires theory
  3252 Library/Sum_Of_Squares.  It is not a complete decision procedure but
  3253 works well in practice on quantifier-free real arithmetic with +, -,
  3254 *, ^, =, <= and <, i.e. boolean combinations of equalities and
  3255 inequalities between polynomials.  It makes use of external
  3256 semidefinite programming solvers.  Method "sos" generates a
  3257 certificate that can be pasted into the proof thus avoiding the need
  3258 to call an external tool every time the proof is checked.  See
  3259 src/HOL/Library/Sum_Of_Squares.
  3260 
  3261 * New method "linarith" invokes existing linear arithmetic decision
  3262 procedure only.
  3263 
  3264 * New command 'atp_minimal' reduces result produced by Sledgehammer.
  3265 
  3266 * New Sledgehammer option "Full Types" in Proof General settings menu.
  3267 Causes full type information to be output to the ATPs.  This slows
  3268 ATPs down considerably but eliminates a source of unsound "proofs"
  3269 that fail later.
  3270 
  3271 * New method "metisFT": A version of metis that uses full type
  3272 information in order to avoid failures of proof reconstruction.
  3273 
  3274 * New evaluator "approximate" approximates an real valued term using
  3275 the same method as the approximation method.
  3276 
  3277 * Method "approximate" now supports arithmetic expressions as
  3278 boundaries of intervals and implements interval splitting and Taylor
  3279 series expansion.
  3280 
  3281 * ML antiquotation @{code_datatype} inserts definition of a datatype
  3282 generated by the code generator; e.g. see src/HOL/Predicate.thy.
  3283 
  3284 * New theory SupInf of the supremum and infimum operators for sets of
  3285 reals.
  3286 
  3287 * New theory Probability, which contains a development of measure
  3288 theory, eventually leading to Lebesgue integration and probability.
  3289 
  3290 * Extended Multivariate Analysis to include derivation and Brouwer's
  3291 fixpoint theorem.
  3292 
  3293 * Reorganization of number theory, INCOMPATIBILITY:
  3294   - new number theory development for nat and int, in theories Divides
  3295     and GCD as well as in new session Number_Theory
  3296   - some constants and facts now suffixed with _nat and _int
  3297     accordingly
  3298   - former session NumberTheory now named Old_Number_Theory, including
  3299     theories Legacy_GCD and Primes (prefer Number_Theory if possible)
  3300   - moved theory Pocklington from src/HOL/Library to
  3301     src/HOL/Old_Number_Theory
  3302 
  3303 * Theory GCD includes functions Gcd/GCD and Lcm/LCM for the gcd and
  3304 lcm of finite and infinite sets. It is shown that they form a complete
  3305 lattice.
  3306 
  3307 * Class semiring_div requires superclass no_zero_divisors and proof of
  3308 div_mult_mult1; theorems div_mult_mult1, div_mult_mult2,
  3309 div_mult_mult1_if, div_mult_mult1 and div_mult_mult2 have been
  3310 generalized to class semiring_div, subsuming former theorems
  3311 zdiv_zmult_zmult1, zdiv_zmult_zmult1_if, zdiv_zmult_zmult1 and
  3312 zdiv_zmult_zmult2.  div_mult_mult1 is now [simp] by default.
  3313 INCOMPATIBILITY.
  3314 
  3315 * Refinements to lattice classes and sets:
  3316   - less default intro/elim rules in locale variant, more default
  3317     intro/elim rules in class variant: more uniformity
  3318   - lemma ge_sup_conv renamed to le_sup_iff, in accordance with
  3319     le_inf_iff
  3320   - dropped lemma alias inf_ACI for inf_aci (same for sup_ACI and
  3321     sup_aci)
  3322   - renamed ACI to inf_sup_aci
  3323   - new class "boolean_algebra"
  3324   - class "complete_lattice" moved to separate theory
  3325     "Complete_Lattice"; corresponding constants (and abbreviations)
  3326     renamed and with authentic syntax:
  3327     Set.Inf ~>    Complete_Lattice.Inf
  3328     Set.Sup ~>    Complete_Lattice.Sup
  3329     Set.INFI ~>   Complete_Lattice.INFI
  3330     Set.SUPR ~>   Complete_Lattice.SUPR
  3331     Set.Inter ~>  Complete_Lattice.Inter
  3332     Set.Union ~>  Complete_Lattice.Union
  3333     Set.INTER ~>  Complete_Lattice.INTER
  3334     Set.UNION ~>  Complete_Lattice.UNION
  3335   - authentic syntax for
  3336     Set.Pow
  3337     Set.image
  3338   - mere abbreviations:
  3339     Set.empty               (for bot)
  3340     Set.UNIV                (for top)
  3341     Set.inter               (for inf, formerly Set.Int)
  3342     Set.union               (for sup, formerly Set.Un)
  3343     Complete_Lattice.Inter  (for Inf)
  3344     Complete_Lattice.Union  (for Sup)
  3345     Complete_Lattice.INTER  (for INFI)
  3346     Complete_Lattice.UNION  (for SUPR)
  3347   - object-logic definitions as far as appropriate
  3348 
  3349 INCOMPATIBILITY.  Care is required when theorems Int_subset_iff or
  3350 Un_subset_iff are explicitly deleted as default simp rules; then also
  3351 their lattice counterparts le_inf_iff and le_sup_iff have to be
  3352 deleted to achieve the desired effect.
  3353 
  3354 * Rules inf_absorb1, inf_absorb2, sup_absorb1, sup_absorb2 are no simp
  3355 rules by default any longer; the same applies to min_max.inf_absorb1
  3356 etc.  INCOMPATIBILITY.
  3357 
  3358 * Rules sup_Int_eq and sup_Un_eq are no longer declared as
  3359 pred_set_conv by default.  INCOMPATIBILITY.
  3360 
  3361 * Power operations on relations and functions are now one dedicated
  3362 constant "compow" with infix syntax "^^".  Power operation on
  3363 multiplicative monoids retains syntax "^" and is now defined generic
  3364 in class power.  INCOMPATIBILITY.
  3365 
  3366 * Relation composition "R O S" now has a more standard argument order:
  3367 "R O S = {(x, z). EX y. (x, y) : R & (y, z) : S}".  INCOMPATIBILITY,
  3368 rewrite propositions with "S O R" --> "R O S". Proofs may occasionally
  3369 break, since the O_assoc rule was not rewritten like this.  Fix using
  3370 O_assoc[symmetric].  The same applies to the curried version "R OO S".
  3371 
  3372 * Function "Inv" is renamed to "inv_into" and function "inv" is now an
  3373 abbreviation for "inv_into UNIV".  Lemmas are renamed accordingly.
  3374 INCOMPATIBILITY.
  3375 
  3376 * Most rules produced by inductive and datatype package have mandatory
  3377 prefixes.  INCOMPATIBILITY.
  3378 
  3379 * Changed "DERIV_intros" to a dynamic fact, which can be augmented by
  3380 the attribute of the same name.  Each of the theorems in the list
  3381 DERIV_intros assumes composition with an additional function and
  3382 matches a variable to the derivative, which has to be solved by the
  3383 Simplifier.  Hence (auto intro!: DERIV_intros) computes the derivative
  3384 of most elementary terms.  Former Maclauren.DERIV_tac and
  3385 Maclauren.deriv_tac should be replaced by (auto intro!: DERIV_intros).
  3386 INCOMPATIBILITY.
  3387 
  3388 * Code generator attributes follow the usual underscore convention:
  3389     code_unfold     replaces    code unfold
  3390     code_post       replaces    code post
  3391     etc.
  3392   INCOMPATIBILITY.
  3393 
  3394 * Renamed methods:
  3395     sizechange -> size_change
  3396     induct_scheme -> induction_schema
  3397   INCOMPATIBILITY.
  3398 
  3399 * Discontinued abbreviation "arbitrary" of constant "undefined".
  3400 INCOMPATIBILITY, use "undefined" directly.
  3401 
  3402 * Renamed theorems:
  3403     Suc_eq_add_numeral_1 -> Suc_eq_plus1
  3404     Suc_eq_add_numeral_1_left -> Suc_eq_plus1_left
  3405     Suc_plus1 -> Suc_eq_plus1
  3406     *anti_sym -> *antisym*
  3407     vector_less_eq_def -> vector_le_def
  3408   INCOMPATIBILITY.
  3409 
  3410 * Added theorem List.map_map as [simp].  Removed List.map_compose.
  3411 INCOMPATIBILITY.
  3412 
  3413 * Removed predicate "M hassize n" (<--> card M = n & finite M).
  3414 INCOMPATIBILITY.
  3415 
  3416 
  3417 *** HOLCF ***
  3418 
  3419 * Theory Representable defines a class "rep" of domains that are
  3420 representable (via an ep-pair) in the universal domain type "udom".
  3421 Instances are provided for all type constructors defined in HOLCF.
  3422 
  3423 * The 'new_domain' command is a purely definitional version of the
  3424 domain package, for representable domains.  Syntax is identical to the
  3425 old domain package.  The 'new_domain' package also supports indirect
  3426 recursion using previously-defined type constructors.  See
  3427 src/HOLCF/ex/New_Domain.thy for examples.
  3428 
  3429 * Method "fixrec_simp" unfolds one step of a fixrec-defined constant
  3430 on the left-hand side of an equation, and then performs
  3431 simplification.  Rewriting is done using rules declared with the
  3432 "fixrec_simp" attribute.  The "fixrec_simp" method is intended as a
  3433 replacement for "fixpat"; see src/HOLCF/ex/Fixrec_ex.thy for examples.
  3434 
  3435 * The pattern-match compiler in 'fixrec' can now handle constructors
  3436 with HOL function types.  Pattern-match combinators for the Pair
  3437 constructor are pre-configured.
  3438 
  3439 * The 'fixrec' package now produces better fixed-point induction rules
  3440 for mutually-recursive definitions:  Induction rules have conclusions
  3441 of the form "P foo bar" instead of "P <foo, bar>".
  3442 
  3443 * The constant "sq_le" (with infix syntax "<<" or "\<sqsubseteq>") has
  3444 been renamed to "below".  The name "below" now replaces "less" in many
  3445 theorem names.  (Legacy theorem names using "less" are still supported
  3446 as well.)
  3447 
  3448 * The 'fixrec' package now supports "bottom patterns".  Bottom
  3449 patterns can be used to generate strictness rules, or to make
  3450 functions more strict (much like the bang-patterns supported by the
  3451 Glasgow Haskell Compiler).  See src/HOLCF/ex/Fixrec_ex.thy for
  3452 examples.
  3453 
  3454 
  3455 *** ML ***
  3456 
  3457 * Support for Poly/ML 5.3.0, with improved reporting of compiler
  3458 errors and run-time exceptions, including detailed source positions.
  3459 
  3460 * Structure Name_Space (formerly NameSpace) now manages uniquely
  3461 identified entries, with some additional information such as source
  3462 position, logical grouping etc.
  3463 
  3464 * Theory and context data is now introduced by the simplified and
  3465 modernized functors Theory_Data, Proof_Data, Generic_Data.  Data needs
  3466 to be pure, but the old TheoryDataFun for mutable data (with explicit
  3467 copy operation) is still available for some time.
  3468 
  3469 * Structure Synchronized (cf. src/Pure/Concurrent/synchronized.ML)
  3470 provides a high-level programming interface to synchronized state
  3471 variables with atomic update.  This works via pure function
  3472 application within a critical section -- its runtime should be as
  3473 short as possible; beware of deadlocks if critical code is nested,
  3474 either directly or indirectly via other synchronized variables!
  3475 
  3476 * Structure Unsynchronized (cf. src/Pure/ML-Systems/unsynchronized.ML)
  3477 wraps raw ML references, explicitly indicating their non-thread-safe
  3478 behaviour.  The Isar toplevel keeps this structure open, to
  3479 accommodate Proof General as well as quick and dirty interactive
  3480 experiments with references.
  3481 
  3482 * PARALLEL_CHOICE and PARALLEL_GOALS provide basic support for
  3483 parallel tactical reasoning.
  3484 
  3485 * Tacticals Subgoal.FOCUS, Subgoal.FOCUS_PREMS, Subgoal.FOCUS_PARAMS
  3486 are similar to SUBPROOF, but are slightly more flexible: only the
  3487 specified parts of the subgoal are imported into the context, and the
  3488 body tactic may introduce new subgoals and schematic variables.
  3489 
  3490 * Old tactical METAHYPS, which does not observe the proof context, has
  3491 been renamed to Old_Goals.METAHYPS and awaits deletion.  Use SUBPROOF
  3492 or Subgoal.FOCUS etc.
  3493 
  3494 * Renamed functor TableFun to Table, and GraphFun to Graph.  (Since
  3495 functors have their own ML name space there is no point to mark them
  3496 separately.)  Minor INCOMPATIBILITY.
  3497 
  3498 * Renamed NamedThmsFun to Named_Thms.  INCOMPATIBILITY.
  3499 
  3500 * Renamed several structures FooBar to Foo_Bar.  Occasional,
  3501 INCOMPATIBILITY.
  3502 
  3503 * Operations of structure Skip_Proof no longer require quick_and_dirty
  3504 mode, which avoids critical setmp.
  3505 
  3506 * Eliminated old Attrib.add_attributes, Method.add_methods and related
  3507 combinators for "args".  INCOMPATIBILITY, need to use simplified
  3508 Attrib/Method.setup introduced in Isabelle2009.
  3509 
  3510 * Proper context for simpset_of, claset_of, clasimpset_of.  May fall
  3511 back on global_simpset_of, global_claset_of, global_clasimpset_of as
  3512 last resort.  INCOMPATIBILITY.
  3513 
  3514 * Display.pretty_thm now requires a proper context (cf. former
  3515 ProofContext.pretty_thm).  May fall back on Display.pretty_thm_global
  3516 or even Display.pretty_thm_without_context as last resort.
  3517 INCOMPATIBILITY.
  3518 
  3519 * Discontinued Display.pretty_ctyp/cterm etc.  INCOMPATIBILITY, use
  3520 Syntax.pretty_typ/term directly, preferably with proper context
  3521 instead of global theory.
  3522 
  3523 
  3524 *** System ***
  3525 
  3526 * Further fine tuning of parallel proof checking, scales up to 8 cores
  3527 (max. speedup factor 5.0).  See also Goal.parallel_proofs in ML and
  3528 usedir option -q.
  3529 
  3530 * Support for additional "Isabelle components" via etc/components, see
  3531 also the system manual.
  3532 
  3533 * The isabelle makeall tool now operates on all components with
  3534 IsaMakefile, not just hardwired "logics".
  3535 
  3536 * Removed "compress" option from isabelle-process and isabelle usedir;
  3537 this is always enabled.
  3538 
  3539 * Discontinued support for Poly/ML 4.x versions.
  3540 
  3541 * Isabelle tool "wwwfind" provides web interface for 'find_theorems'
  3542 on a given logic image.  This requires the lighttpd webserver and is
  3543 currently supported on Linux only.
  3544 
  3545 
  3546 
  3547 New in Isabelle2009 (April 2009)
  3548 --------------------------------
  3549 
  3550 *** General ***
  3551 
  3552 * Simplified main Isabelle executables, with less surprises on
  3553 case-insensitive file-systems (such as Mac OS).
  3554 
  3555   - The main Isabelle tool wrapper is now called "isabelle" instead of
  3556     "isatool."
  3557 
  3558   - The former "isabelle" alias for "isabelle-process" has been
  3559     removed (should rarely occur to regular users).
  3560 
  3561   - The former "isabelle-interface" and its alias "Isabelle" have been
  3562     removed (interfaces are now regular Isabelle tools).
  3563 
  3564 Within scripts and make files, the Isabelle environment variables
  3565 ISABELLE_TOOL and ISABELLE_PROCESS replace old ISATOOL and ISABELLE,
  3566 respectively.  (The latter are still available as legacy feature.)
  3567 
  3568 The old isabelle-interface wrapper could react in confusing ways if
  3569 the interface was uninstalled or changed otherwise.  Individual
  3570 interface tool configuration is now more explicit, see also the
  3571 Isabelle system manual.  In particular, Proof General is now available
  3572 via "isabelle emacs".
  3573 
  3574 INCOMPATIBILITY, need to adapt derivative scripts.  Users may need to
  3575 purge installed copies of Isabelle executables and re-run "isabelle
  3576 install -p ...", or use symlinks.
  3577 
  3578 * The default for ISABELLE_HOME_USER is now ~/.isabelle instead of the
  3579 old ~/isabelle, which was slightly non-standard and apt to cause
  3580 surprises on case-insensitive file-systems (such as Mac OS).
  3581 
  3582 INCOMPATIBILITY, need to move existing ~/isabelle/etc,
  3583 ~/isabelle/heaps, ~/isabelle/browser_info to the new place.  Special
  3584 care is required when using older releases of Isabelle.  Note that
  3585 ISABELLE_HOME_USER can be changed in Isabelle/etc/settings of any
  3586 Isabelle distribution, in order to use the new ~/.isabelle uniformly.
  3587 
  3588 * Proofs of fully specified statements are run in parallel on
  3589 multi-core systems.  A speedup factor of 2.5 to 3.2 can be expected on
  3590 a regular 4-core machine, if the initial heap space is made reasonably
  3591 large (cf. Poly/ML option -H).  (Requires Poly/ML 5.2.1 or later.)
  3592 
  3593 * The main reference manuals ("isar-ref", "implementation", and
  3594 "system") have been updated and extended.  Formally checked references
  3595 as hyperlinks are now available uniformly.
  3596 
  3597 
  3598 *** Pure ***
  3599 
  3600 * Complete re-implementation of locales.  INCOMPATIBILITY in several
  3601 respects.  The most important changes are listed below.  See the
  3602 Tutorial on Locales ("locales" manual) for details.
  3603 
  3604 - In locale expressions, instantiation replaces renaming.  Parameters
  3605 must be declared in a for clause.  To aid compatibility with previous
  3606 parameter inheritance, in locale declarations, parameters that are not
  3607 'touched' (instantiation position "_" or omitted) are implicitly added
  3608 with their syntax at the beginning of the for clause.
  3609 
  3610 - Syntax from abbreviations and definitions in locales is available in
  3611 locale expressions and context elements.  The latter is particularly
  3612 useful in locale declarations.
  3613 
  3614 - More flexible mechanisms to qualify names generated by locale
  3615 expressions.  Qualifiers (prefixes) may be specified in locale
  3616 expressions, and can be marked as mandatory (syntax: "name!:") or
  3617 optional (syntax "name?:").  The default depends for plain "name:"
  3618 depends on the situation where a locale expression is used: in
  3619 commands 'locale' and 'sublocale' prefixes are optional, in
  3620 'interpretation' and 'interpret' prefixes are mandatory.  The old
  3621 implicit qualifiers derived from the parameter names of a locale are
  3622 no longer generated.
  3623 
  3624 - Command "sublocale l < e" replaces "interpretation l < e".  The
  3625 instantiation clause in "interpretation" and "interpret" (square
  3626 brackets) is no longer available.  Use locale expressions.
  3627 
  3628 - When converting proof scripts, mandatory qualifiers in
  3629 'interpretation' and 'interpret' should be retained by default, even
  3630 if this is an INCOMPATIBILITY compared to former behavior.  In the
  3631 worst case, use the "name?:" form for non-mandatory ones.  Qualifiers
  3632 in locale expressions range over a single locale instance only.
  3633 
  3634 - Dropped locale element "includes".  This is a major INCOMPATIBILITY.
  3635 In existing theorem specifications replace the includes element by the
  3636 respective context elements of the included locale, omitting those
  3637 that are already present in the theorem specification.  Multiple
  3638 assume elements of a locale should be replaced by a single one
  3639 involving the locale predicate.  In the proof body, declarations (most
  3640 notably theorems) may be regained by interpreting the respective
  3641 locales in the proof context as required (command "interpret").
  3642 
  3643 If using "includes" in replacement of a target solely because the
  3644 parameter types in the theorem are not as general as in the target,
  3645 consider declaring a new locale with additional type constraints on
  3646 the parameters (context element "constrains").
  3647 
  3648 - Discontinued "locale (open)".  INCOMPATIBILITY.
  3649 
  3650 - Locale interpretation commands no longer attempt to simplify goal.
  3651 INCOMPATIBILITY: in rare situations the generated goal differs.  Use
  3652 methods intro_locales and unfold_locales to clarify.
  3653 
  3654 - Locale interpretation commands no longer accept interpretation
  3655 attributes.  INCOMPATIBILITY.
  3656 
  3657 * Class declaration: so-called "base sort" must not be given in import
  3658 list any longer, but is inferred from the specification.  Particularly
  3659 in HOL, write
  3660 
  3661     class foo = ...
  3662 
  3663 instead of
  3664 
  3665     class foo = type + ...
  3666 
  3667 * Class target: global versions of theorems stemming do not carry a
  3668 parameter prefix any longer.  INCOMPATIBILITY.
  3669 
  3670 * Class 'instance' command no longer accepts attached definitions.
  3671 INCOMPATIBILITY, use proper 'instantiation' target instead.
  3672 
  3673 * Recovered hiding of consts, which was accidentally broken in
  3674 Isabelle2007.  Potential INCOMPATIBILITY, ``hide const c'' really
  3675 makes c inaccessible; consider using ``hide (open) const c'' instead.
  3676 
  3677 * Slightly more coherent Pure syntax, with updated documentation in
  3678 isar-ref manual.  Removed locales meta_term_syntax and
  3679 meta_conjunction_syntax: TERM and &&& (formerly &&) are now permanent,
  3680 INCOMPATIBILITY in rare situations.  Note that &&& should not be used
  3681 directly in regular applications.
  3682 
  3683 * There is a new syntactic category "float_const" for signed decimal
  3684 fractions (e.g. 123.45 or -123.45).
  3685 
  3686 * Removed exotic 'token_translation' command.  INCOMPATIBILITY, use ML
  3687 interface with 'setup' command instead.
  3688 
  3689 * Command 'local_setup' is similar to 'setup', but operates on a local
  3690 theory context.
  3691 
  3692 * The 'axiomatization' command now only works within a global theory
  3693 context.  INCOMPATIBILITY.
  3694 
  3695 * Goal-directed proof now enforces strict proof irrelevance wrt. sort
  3696 hypotheses.  Sorts required in the course of reasoning need to be
  3697 covered by the constraints in the initial statement, completed by the
  3698 type instance information of the background theory.  Non-trivial sort
  3699 hypotheses, which rarely occur in practice, may be specified via
  3700 vacuous propositions of the form SORT_CONSTRAINT('a::c).  For example:
  3701 
  3702   lemma assumes "SORT_CONSTRAINT('a::empty)" shows False ...
  3703 
  3704 The result contains an implicit sort hypotheses as before --
  3705 SORT_CONSTRAINT premises are eliminated as part of the canonical rule
  3706 normalization.
  3707 
  3708 * Generalized Isar history, with support for linear undo, direct state
  3709 addressing etc.
  3710 
  3711 * Changed defaults for unify configuration options:
  3712 
  3713   unify_trace_bound = 50 (formerly 25)
  3714   unify_search_bound = 60 (formerly 30)
  3715 
  3716 * Different bookkeeping for code equations (INCOMPATIBILITY):
  3717 
  3718   a) On theory merge, the last set of code equations for a particular
  3719      constant is taken (in accordance with the policy applied by other
  3720      parts of the code generator framework).
  3721 
  3722   b) Code equations stemming from explicit declarations (e.g. code
  3723      attribute) gain priority over default code equations stemming
  3724      from definition, primrec, fun etc.
  3725 
  3726 * Keyword 'code_exception' now named 'code_abort'.  INCOMPATIBILITY.
  3727 
  3728 * Unified theorem tables for both code generators.  Thus [code
  3729 func] has disappeared and only [code] remains.  INCOMPATIBILITY.
  3730 
  3731 * Command 'find_consts' searches for constants based on type and name
  3732 patterns, e.g.
  3733 
  3734     find_consts "_ => bool"
  3735 
  3736 By default, matching is against subtypes, but it may be restricted to
  3737 the whole type.  Searching by name is possible.  Multiple queries are
  3738 conjunctive and queries may be negated by prefixing them with a
  3739 hyphen:
  3740 
  3741     find_consts strict: "_ => bool" name: "Int" -"int => int"
  3742 
  3743 * New 'find_theorems' criterion "solves" matches theorems that
  3744 directly solve the current goal (modulo higher-order unification).
  3745 
  3746 * Auto solve feature for main theorem statements: whenever a new goal
  3747 is stated, "find_theorems solves" is called; any theorems that could
  3748 solve the lemma directly are listed as part of the goal state.
  3749 Cf. associated options in Proof General Isabelle settings menu,
  3750 enabled by default, with reasonable timeout for pathological cases of
  3751 higher-order unification.
  3752 
  3753 
  3754 *** Document preparation ***
  3755 
  3756 * Antiquotation @{lemma} now imitates a regular terminal proof,
  3757 demanding keyword 'by' and supporting the full method expression
  3758 syntax just like the Isar command 'by'.
  3759 
  3760 
  3761 *** HOL ***
  3762 
  3763 * Integrated main parts of former image HOL-Complex with HOL.  Entry
  3764 points Main and Complex_Main remain as before.
  3765 
  3766 * Logic image HOL-Plain provides a minimal HOL with the most important
  3767 tools available (inductive, datatype, primrec, ...).  This facilitates
  3768 experimentation and tool development.  Note that user applications
  3769 (and library theories) should never refer to anything below theory
  3770 Main, as before.
  3771 
  3772 * Logic image HOL-Main stops at theory Main, and thus facilitates
  3773 experimentation due to shorter build times.
  3774 
  3775 * Logic image HOL-NSA contains theories of nonstandard analysis which
  3776 were previously part of former HOL-Complex.  Entry point Hyperreal
  3777 remains valid, but theories formerly using Complex_Main should now use
  3778 new entry point Hypercomplex.
  3779 
  3780 * Generic ATP manager for Sledgehammer, based on ML threads instead of
  3781 Posix processes.  Avoids potentially expensive forking of the ML
  3782 process.  New thread-based implementation also works on non-Unix
  3783 platforms (Cygwin).  Provers are no longer hardwired, but defined
  3784 within the theory via plain ML wrapper functions.  Basic Sledgehammer
  3785 commands are covered in the isar-ref manual.
  3786 
  3787 * Wrapper scripts for remote SystemOnTPTP service allows to use
  3788 sledgehammer without local ATP installation (Vampire etc.). Other
  3789 provers may be included via suitable ML wrappers, see also
  3790 src/HOL/ATP_Linkup.thy.
  3791 
  3792 * ATP selection (E/Vampire/Spass) is now via Proof General's settings
  3793 menu.
  3794 
  3795 * The metis method no longer fails because the theorem is too trivial
  3796 (contains the empty clause).
  3797 
  3798 * The metis method now fails in the usual manner, rather than raising
  3799 an exception, if it determines that it cannot prove the theorem.
  3800 
  3801 * Method "coherent" implements a prover for coherent logic (see also
  3802 src/Tools/coherent.ML).
  3803 
  3804 * Constants "undefined" and "default" replace "arbitrary".  Usually
  3805 "undefined" is the right choice to replace "arbitrary", though
  3806 logically there is no difference.  INCOMPATIBILITY.
  3807 
  3808 * Command "value" now integrates different evaluation mechanisms.  The
  3809 result of the first successful evaluation mechanism is printed.  In
  3810 square brackets a particular named evaluation mechanisms may be
  3811 specified (currently, [SML], [code] or [nbe]).  See further
  3812 src/HOL/ex/Eval_Examples.thy.
  3813 
  3814 * Normalization by evaluation now allows non-leftlinear equations.
  3815 Declare with attribute [code nbe].
  3816 
  3817 * Methods "case_tac" and "induct_tac" now refer to the very same rules
  3818 as the structured Isar versions "cases" and "induct", cf. the
  3819 corresponding "cases" and "induct" attributes.  Mutual induction rules
  3820 are now presented as a list of individual projections
  3821 (e.g. foo_bar.inducts for types foo and bar); the old format with
  3822 explicit HOL conjunction is no longer supported.  INCOMPATIBILITY, in
  3823 rare situations a different rule is selected --- notably nested tuple
  3824 elimination instead of former prod.exhaust: use explicit (case_tac t
  3825 rule: prod.exhaust) here.
  3826 
  3827 * Attributes "cases", "induct", "coinduct" support "del" option.
  3828 
  3829 * Removed fact "case_split_thm", which duplicates "case_split".
  3830 
  3831 * The option datatype has been moved to a new theory Option.  Renamed
  3832 option_map to Option.map, and o2s to Option.set, INCOMPATIBILITY.
  3833 
  3834 * New predicate "strict_mono" classifies strict functions on partial
  3835 orders.  With strict functions on linear orders, reasoning about
  3836 (in)equalities is facilitated by theorems "strict_mono_eq",
  3837 "strict_mono_less_eq" and "strict_mono_less".
  3838 
  3839 * Some set operations are now proper qualified constants with
  3840 authentic syntax.  INCOMPATIBILITY:
  3841 
  3842     op Int ~>   Set.Int
  3843     op Un ~>    Set.Un
  3844     INTER ~>    Set.INTER
  3845     UNION ~>    Set.UNION
  3846     Inter ~>    Set.Inter
  3847     Union ~>    Set.Union
  3848     {} ~>       Set.empty
  3849     UNIV ~>     Set.UNIV
  3850 
  3851 * Class complete_lattice with operations Inf, Sup, INFI, SUPR now in
  3852 theory Set.
  3853 
  3854 * Auxiliary class "itself" has disappeared -- classes without any
  3855 parameter are treated as expected by the 'class' command.
  3856 
  3857 * Leibnitz's Series for Pi and the arcus tangens and logarithm series.
  3858 
  3859 * Common decision procedures (Cooper, MIR, Ferrack, Approximation,
  3860 Dense_Linear_Order) are now in directory HOL/Decision_Procs.
  3861 
  3862 * Theory src/HOL/Decision_Procs/Approximation provides the new proof
  3863 method "approximation".  It proves formulas on real values by using
  3864 interval arithmetic.  In the formulas are also the transcendental
  3865 functions sin, cos, tan, atan, ln, exp and the constant pi are
  3866 allowed. For examples see
  3867 src/HOL/Descision_Procs/ex/Approximation_Ex.thy.
  3868 
  3869 * Theory "Reflection" now resides in HOL/Library.
  3870 
  3871 * Entry point to Word library now simply named "Word".
  3872 INCOMPATIBILITY.
  3873 
  3874 * Made source layout more coherent with logical distribution
  3875 structure:
  3876 
  3877     src/HOL/Library/RType.thy ~> src/HOL/Typerep.thy
  3878     src/HOL/Library/Code_Message.thy ~> src/HOL/
  3879     src/HOL/Library/GCD.thy ~> src/HOL/
  3880     src/HOL/Library/Order_Relation.thy ~> src/HOL/
  3881     src/HOL/Library/Parity.thy ~> src/HOL/
  3882     src/HOL/Library/Univ_Poly.thy ~> src/HOL/
  3883     src/HOL/Real/ContNotDenum.thy ~> src/HOL/Library/
  3884     src/HOL/Real/Lubs.thy ~> src/HOL/
  3885     src/HOL/Real/PReal.thy ~> src/HOL/
  3886     src/HOL/Real/Rational.thy ~> src/HOL/
  3887     src/HOL/Real/RComplete.thy ~> src/HOL/
  3888     src/HOL/Real/RealDef.thy ~> src/HOL/
  3889     src/HOL/Real/RealPow.thy ~> src/HOL/
  3890     src/HOL/Real/Real.thy ~> src/HOL/
  3891     src/HOL/Complex/Complex_Main.thy ~> src/HOL/
  3892     src/HOL/Complex/Complex.thy ~> src/HOL/
  3893     src/HOL/Complex/FrechetDeriv.thy ~> src/HOL/Library/
  3894     src/HOL/Complex/Fundamental_Theorem_Algebra.thy ~> src/HOL/Library/
  3895     src/HOL/Hyperreal/Deriv.thy ~> src/HOL/
  3896     src/HOL/Hyperreal/Fact.thy ~> src/HOL/
  3897     src/HOL/Hyperreal/Integration.thy ~> src/HOL/
  3898     src/HOL/Hyperreal/Lim.thy ~> src/HOL/
  3899     src/HOL/Hyperreal/Ln.thy ~> src/HOL/
  3900     src/HOL/Hyperreal/Log.thy ~> src/HOL/
  3901     src/HOL/Hyperreal/MacLaurin.thy ~> src/HOL/
  3902     src/HOL/Hyperreal/NthRoot.thy ~> src/HOL/
  3903     src/HOL/Hyperreal/Series.thy ~> src/HOL/
  3904     src/HOL/Hyperreal/SEQ.thy ~> src/HOL/
  3905     src/HOL/Hyperreal/Taylor.thy ~> src/HOL/
  3906     src/HOL/Hyperreal/Transcendental.thy ~> src/HOL/
  3907     src/HOL/Real/Float ~> src/HOL/Library/
  3908     src/HOL/Real/HahnBanach ~> src/HOL/HahnBanach
  3909     src/HOL/Real/RealVector.thy ~> src/HOL/
  3910 
  3911     src/HOL/arith_data.ML ~> src/HOL/Tools
  3912     src/HOL/hologic.ML ~> src/HOL/Tools
  3913     src/HOL/simpdata.ML ~> src/HOL/Tools
  3914     src/HOL/int_arith1.ML ~> src/HOL/Tools/int_arith.ML
  3915     src/HOL/int_factor_simprocs.ML ~> src/HOL/Tools
  3916     src/HOL/nat_simprocs.ML ~> src/HOL/Tools
  3917     src/HOL/Real/float_arith.ML ~> src/HOL/Tools
  3918     src/HOL/Real/float_syntax.ML ~> src/HOL/Tools
  3919     src/HOL/Real/rat_arith.ML ~> src/HOL/Tools
  3920     src/HOL/Real/real_arith.ML ~> src/HOL/Tools
  3921 
  3922     src/HOL/Library/Array.thy ~> src/HOL/Imperative_HOL
  3923     src/HOL/Library/Heap_Monad.thy ~> src/HOL/Imperative_HOL
  3924     src/HOL/Library/Heap.thy ~> src/HOL/Imperative_HOL
  3925     src/HOL/Library/Imperative_HOL.thy ~> src/HOL/Imperative_HOL
  3926     src/HOL/Library/Ref.thy ~> src/HOL/Imperative_HOL
  3927     src/HOL/Library/Relational.thy ~> src/HOL/Imperative_HOL
  3928 
  3929 * If methods "eval" and "evaluation" encounter a structured proof
  3930 state with !!/==>, only the conclusion is evaluated to True (if
  3931 possible), avoiding strange error messages.
  3932 
  3933 * Method "sizechange" automates termination proofs using (a
  3934 modification of) the size-change principle.  Requires SAT solver.  See
  3935 src/HOL/ex/Termination.thy for examples.
  3936 
  3937 * Simplifier: simproc for let expressions now unfolds if bound
  3938 variable occurs at most once in let expression body.  INCOMPATIBILITY.
  3939 
  3940 * Method "arith": Linear arithmetic now ignores all inequalities when
  3941 fast_arith_neq_limit is exceeded, instead of giving up entirely.
  3942 
  3943 * New attribute "arith" for facts that should always be used
  3944 automatically by arithmetic. It is intended to be used locally in
  3945 proofs, e.g.
  3946 
  3947   assumes [arith]: "x > 0"
  3948 
  3949 Global usage is discouraged because of possible performance impact.
  3950 
  3951 * New classes "top" and "bot" with corresponding operations "top" and
  3952 "bot" in theory Orderings; instantiation of class "complete_lattice"
  3953 requires instantiation of classes "top" and "bot".  INCOMPATIBILITY.
  3954 
  3955 * Changed definition lemma "less_fun_def" in order to provide an
  3956 instance for preorders on functions; use lemma "less_le" instead.
  3957 INCOMPATIBILITY.
  3958 
  3959 * Theory Orderings: class "wellorder" moved here, with explicit
  3960 induction rule "less_induct" as assumption.  For instantiation of
  3961 "wellorder" by means of predicate "wf", use rule wf_wellorderI.
  3962 INCOMPATIBILITY.
  3963 
  3964 * Theory Orderings: added class "preorder" as superclass of "order".
  3965 INCOMPATIBILITY: Instantiation proofs for order, linorder
  3966 etc. slightly changed.  Some theorems named order_class.* now named
  3967 preorder_class.*.
  3968 
  3969 * Theory Relation: renamed "refl" to "refl_on", "reflexive" to "refl,
  3970 "diag" to "Id_on".
  3971 
  3972 * Theory Finite_Set: added a new fold combinator of type
  3973 
  3974   ('a => 'b => 'b) => 'b => 'a set => 'b
  3975 
  3976 Occasionally this is more convenient than the old fold combinator
  3977 which is now defined in terms of the new one and renamed to
  3978 fold_image.
  3979 
  3980 * Theories Ring_and_Field and OrderedGroup: The lemmas "group_simps"
  3981 and "ring_simps" have been replaced by "algebra_simps" (which can be
  3982 extended with further lemmas!).  At the moment both still exist but
  3983 the former will disappear at some point.
  3984 
  3985 * Theory Power: Lemma power_Suc is now declared as a simp rule in
  3986 class recpower.  Type-specific simp rules for various recpower types
  3987 have been removed.  INCOMPATIBILITY, rename old lemmas as follows:
  3988 
  3989 rat_power_0    -> power_0
  3990 rat_power_Suc  -> power_Suc
  3991 realpow_0      -> power_0
  3992 realpow_Suc    -> power_Suc
  3993 complexpow_0   -> power_0
  3994 complexpow_Suc -> power_Suc
  3995 power_poly_0   -> power_0
  3996 power_poly_Suc -> power_Suc
  3997 
  3998 * Theories Ring_and_Field and Divides: Definition of "op dvd" has been
  3999 moved to separate class dvd in Ring_and_Field; a couple of lemmas on
  4000 dvd has been generalized to class comm_semiring_1.  Likewise a bunch
  4001 of lemmas from Divides has been generalized from nat to class
  4002 semiring_div.  INCOMPATIBILITY.  This involves the following theorem
  4003 renames resulting from duplicate elimination:
  4004 
  4005     dvd_def_mod ~>          dvd_eq_mod_eq_0
  4006     zero_dvd_iff ~>         dvd_0_left_iff
  4007     dvd_0 ~>                dvd_0_right
  4008     DIVISION_BY_ZERO_DIV ~> div_by_0
  4009     DIVISION_BY_ZERO_MOD ~> mod_by_0
  4010     mult_div ~>             div_mult_self2_is_id
  4011     mult_mod ~>             mod_mult_self2_is_0
  4012 
  4013 * Theory IntDiv: removed many lemmas that are instances of class-based
  4014 generalizations (from Divides and Ring_and_Field).  INCOMPATIBILITY,
  4015 rename old lemmas as follows:
  4016 
  4017 dvd_diff               -> nat_dvd_diff
  4018 dvd_zminus_iff         -> dvd_minus_iff
  4019 mod_add1_eq            -> mod_add_eq
  4020 mod_mult1_eq           -> mod_mult_right_eq
  4021 mod_mult1_eq'          -> mod_mult_left_eq
  4022 mod_mult_distrib_mod   -> mod_mult_eq
  4023 nat_mod_add_left_eq    -> mod_add_left_eq
  4024 nat_mod_add_right_eq   -> mod_add_right_eq
  4025 nat_mod_div_trivial    -> mod_div_trivial
  4026 nat_mod_mod_trivial    -> mod_mod_trivial
  4027 zdiv_zadd_self1        -> div_add_self1
  4028 zdiv_zadd_self2        -> div_add_self2
  4029 zdiv_zmult_self1       -> div_mult_self2_is_id
  4030 zdiv_zmult_self2       -> div_mult_self1_is_id
  4031 zdvd_triv_left         -> dvd_triv_left
  4032 zdvd_triv_right        -> dvd_triv_right
  4033 zdvd_zmult_cancel_disj -> dvd_mult_cancel_left
  4034 zmod_eq0_zdvd_iff      -> dvd_eq_mod_eq_0[symmetric]
  4035 zmod_zadd_left_eq      -> mod_add_left_eq
  4036 zmod_zadd_right_eq     -> mod_add_right_eq
  4037 zmod_zadd_self1        -> mod_add_self1
  4038 zmod_zadd_self2        -> mod_add_self2
  4039 zmod_zadd1_eq          -> mod_add_eq
  4040 zmod_zdiff1_eq         -> mod_diff_eq
  4041 zmod_zdvd_zmod         -> mod_mod_cancel
  4042 zmod_zmod_cancel       -> mod_mod_cancel
  4043 zmod_zmult_self1       -> mod_mult_self2_is_0
  4044 zmod_zmult_self2       -> mod_mult_self1_is_0
  4045 zmod_1                 -> mod_by_1
  4046 zdiv_1                 -> div_by_1
  4047 zdvd_abs1              -> abs_dvd_iff
  4048 zdvd_abs2              -> dvd_abs_iff
  4049 zdvd_refl              -> dvd_refl
  4050 zdvd_trans             -> dvd_trans
  4051 zdvd_zadd              -> dvd_add
  4052 zdvd_zdiff             -> dvd_diff
  4053 zdvd_zminus_iff        -> dvd_minus_iff
  4054 zdvd_zminus2_iff       -> minus_dvd_iff
  4055 zdvd_zmultD            -> dvd_mult_right
  4056 zdvd_zmultD2           -> dvd_mult_left
  4057 zdvd_zmult_mono        -> mult_dvd_mono
  4058 zdvd_0_right           -> dvd_0_right
  4059 zdvd_0_left            -> dvd_0_left_iff
  4060 zdvd_1_left            -> one_dvd
  4061 zminus_dvd_iff         -> minus_dvd_iff
  4062 
  4063 * Theory Rational: 'Fract k 0' now equals '0'.  INCOMPATIBILITY.
  4064 
  4065 * The real numbers offer decimal input syntax: 12.34 is translated
  4066 into 1234/10^2. This translation is not reversed upon output.
  4067 
  4068 * Theory Library/Polynomial defines an abstract type 'a poly of
  4069 univariate polynomials with coefficients of type 'a.  In addition to
  4070 the standard ring operations, it also supports div and mod.  Code
  4071 generation is also supported, using list-style constructors.
  4072 
  4073 * Theory Library/Inner_Product defines a class of real_inner for real
  4074 inner product spaces, with an overloaded operation inner :: 'a => 'a
  4075 => real.  Class real_inner is a subclass of real_normed_vector from
  4076 theory RealVector.
  4077 
  4078 * Theory Library/Product_Vector provides instances for the product
  4079 type 'a * 'b of several classes from RealVector and Inner_Product.
  4080 Definitions of addition, subtraction, scalar multiplication, norms,
  4081 and inner products are included.
  4082 
  4083 * Theory Library/Bit defines the field "bit" of integers modulo 2.  In
  4084 addition to the field operations, numerals and case syntax are also
  4085 supported.
  4086 
  4087 * Theory Library/Diagonalize provides constructive version of Cantor's
  4088 first diagonalization argument.
  4089 
  4090 * Theory Library/GCD: Curried operations gcd, lcm (for nat) and zgcd,
  4091 zlcm (for int); carried together from various gcd/lcm developements in
  4092 the HOL Distribution.  Constants zgcd and zlcm replace former igcd and
  4093 ilcm; corresponding theorems renamed accordingly.  INCOMPATIBILITY,
  4094 may recover tupled syntax as follows:
  4095 
  4096     hide (open) const gcd
  4097     abbreviation gcd where
  4098       "gcd == (%(a, b). GCD.gcd a b)"
  4099     notation (output)
  4100       GCD.gcd ("gcd '(_, _')")
  4101 
  4102 The same works for lcm, zgcd, zlcm.
  4103 
  4104 * Theory Library/Nat_Infinity: added addition, numeral syntax and more
  4105 instantiations for algebraic structures.  Removed some duplicate
  4106 theorems.  Changes in simp rules.  INCOMPATIBILITY.
  4107 
  4108 * ML antiquotation @{code} takes a constant as argument and generates
  4109 corresponding code in background and inserts name of the corresponding
  4110 resulting ML value/function/datatype constructor binding in place.
  4111 All occurrences of @{code} with a single ML block are generated
  4112 simultaneously.  Provides a generic and safe interface for
  4113 instrumentalizing code generation.  See
  4114 src/HOL/Decision_Procs/Ferrack.thy for a more ambitious application.
  4115 In future you ought to refrain from ad-hoc compiling generated SML
  4116 code on the ML toplevel.  Note that (for technical reasons) @{code}
  4117 cannot refer to constants for which user-defined serializations are
  4118 set.  Refer to the corresponding ML counterpart directly in that
  4119 cases.
  4120 
  4121 * Command 'rep_datatype': instead of theorem names the command now
  4122 takes a list of terms denoting the constructors of the type to be
  4123 represented as datatype.  The characteristic theorems have to be
  4124 proven.  INCOMPATIBILITY.  Also observe that the following theorems
  4125 have disappeared in favour of existing ones:
  4126 
  4127     unit_induct                 ~> unit.induct
  4128     prod_induct                 ~> prod.induct
  4129     sum_induct                  ~> sum.induct
  4130     Suc_Suc_eq                  ~> nat.inject
  4131     Suc_not_Zero Zero_not_Suc   ~> nat.distinct
  4132 
  4133 
  4134 *** HOL-Algebra ***
  4135 
  4136 * New locales for orders and lattices where the equivalence relation
  4137 is not restricted to equality.  INCOMPATIBILITY: all order and lattice
  4138 locales use a record structure with field eq for the equivalence.
  4139 
  4140 * New theory of factorial domains.
  4141 
  4142 * Units_l_inv and Units_r_inv are now simp rules by default.
  4143 INCOMPATIBILITY.  Simplifier proof that require deletion of l_inv
  4144 and/or r_inv will now also require deletion of these lemmas.
  4145 
  4146 * Renamed the following theorems, INCOMPATIBILITY:
  4147 
  4148 UpperD ~> Upper_memD
  4149 LowerD ~> Lower_memD
  4150 least_carrier ~> least_closed
  4151 greatest_carrier ~> greatest_closed
  4152 greatest_Lower_above ~> greatest_Lower_below
  4153 one_zero ~> carrier_one_zero
  4154 one_not_zero ~> carrier_one_not_zero  (collision with assumption)
  4155 
  4156 
  4157 *** HOL-Nominal ***
  4158 
  4159 * Nominal datatypes can now contain type-variables.
  4160 
  4161 * Commands 'nominal_inductive' and 'equivariance' work with local
  4162 theory targets.
  4163 
  4164 * Nominal primrec can now works with local theory targets and its
  4165 specification syntax now conforms to the general format as seen in
  4166 'inductive' etc.
  4167 
  4168 * Method "perm_simp" honours the standard simplifier attributes
  4169 (no_asm), (no_asm_use) etc.
  4170 
  4171 * The new predicate #* is defined like freshness, except that on the
  4172 left hand side can be a set or list of atoms.
  4173 
  4174 * Experimental command 'nominal_inductive2' derives strong induction
  4175 principles for inductive definitions.  In contrast to
  4176 'nominal_inductive', which can only deal with a fixed number of
  4177 binders, it can deal with arbitrary expressions standing for sets of
  4178 atoms to be avoided.  The only inductive definition we have at the
  4179 moment that needs this generalisation is the typing rule for Lets in
  4180 the algorithm W:
  4181 
  4182  Gamma |- t1 : T1   (x,close Gamma T1)::Gamma |- t2 : T2   x#Gamma
  4183  -----------------------------------------------------------------
  4184          Gamma |- Let x be t1 in t2 : T2
  4185 
  4186 In this rule one wants to avoid all the binders that are introduced by
  4187 "close Gamma T1".  We are looking for other examples where this
  4188 feature might be useful.  Please let us know.
  4189 
  4190 
  4191 *** HOLCF ***
  4192 
  4193 * Reimplemented the simplification procedure for proving continuity
  4194 subgoals.  The new simproc is extensible; users can declare additional
  4195 continuity introduction rules with the attribute [cont2cont].
  4196 
  4197 * The continuity simproc now uses a different introduction rule for
  4198 solving continuity subgoals on terms with lambda abstractions.  In
  4199 some rare cases the new simproc may fail to solve subgoals that the
  4200 old one could solve, and "simp add: cont2cont_LAM" may be necessary.
  4201 Potential INCOMPATIBILITY.
  4202 
  4203 * Command 'fixrec': specification syntax now conforms to the general
  4204 format as seen in 'inductive' etc.  See src/HOLCF/ex/Fixrec_ex.thy for
  4205 examples.  INCOMPATIBILITY.
  4206 
  4207 
  4208 *** ZF ***
  4209 
  4210 * Proof of Zorn's Lemma for partial orders.
  4211 
  4212 
  4213 *** ML ***
  4214 
  4215 * Multithreading for Poly/ML 5.1/5.2 is no longer supported, only for
  4216 Poly/ML 5.2.1 or later.  Important note: the TimeLimit facility
  4217 depends on multithreading, so timouts will not work before Poly/ML
  4218 5.2.1!
  4219 
  4220 * High-level support for concurrent ML programming, see
  4221 src/Pure/Cuncurrent.  The data-oriented model of "future values" is
  4222 particularly convenient to organize independent functional
  4223 computations.  The concept of "synchronized variables" provides a
  4224 higher-order interface for components with shared state, avoiding the
  4225 delicate details of mutexes and condition variables.  (Requires
  4226 Poly/ML 5.2.1 or later.)
  4227 
  4228 * ML bindings produced via Isar commands are stored within the Isar
  4229 context (theory or proof).  Consequently, commands like 'use' and 'ML'
  4230 become thread-safe and work with undo as expected (concerning
  4231 top-level bindings, not side-effects on global references).
  4232 INCOMPATIBILITY, need to provide proper Isar context when invoking the
  4233 compiler at runtime; really global bindings need to be given outside a
  4234 theory.  (Requires Poly/ML 5.2 or later.)
  4235 
  4236 * Command 'ML_prf' is analogous to 'ML' but works within a proof
  4237 context.  Top-level ML bindings are stored within the proof context in
  4238 a purely sequential fashion, disregarding the nested proof structure.
  4239 ML bindings introduced by 'ML_prf' are discarded at the end of the
  4240 proof.  (Requires Poly/ML 5.2 or later.)
  4241 
  4242 * Simplified ML attribute and method setup, cf. functions Attrib.setup
  4243 and Method.setup, as well as Isar commands 'attribute_setup' and
  4244 'method_setup'.  INCOMPATIBILITY for 'method_setup', need to simplify
  4245 existing code accordingly, or use plain 'setup' together with old
  4246 Method.add_method.
  4247 
  4248 * Simplified ML oracle interface Thm.add_oracle promotes 'a -> cterm
  4249 to 'a -> thm, while results are always tagged with an authentic oracle
  4250 name.  The Isar command 'oracle' is now polymorphic, no argument type
  4251 is specified.  INCOMPATIBILITY, need to simplify existing oracle code
  4252 accordingly.  Note that extra performance may be gained by producing
  4253 the cterm carefully, avoiding slow Thm.cterm_of.
  4254 
  4255 * Simplified interface for defining document antiquotations via
  4256 ThyOutput.antiquotation, ThyOutput.output, and optionally
  4257 ThyOutput.maybe_pretty_source.  INCOMPATIBILITY, need to simplify user
  4258 antiquotations accordingly, see src/Pure/Thy/thy_output.ML for common
  4259 examples.
  4260 
  4261 * More systematic treatment of long names, abstract name bindings, and
  4262 name space operations.  Basic operations on qualified names have been
  4263 move from structure NameSpace to Long_Name, e.g. Long_Name.base_name,
  4264 Long_Name.append.  Old type bstring has been mostly replaced by
  4265 abstract type binding (see structure Binding), which supports precise
  4266 qualification by packages and local theory targets, as well as proper
  4267 tracking of source positions.  INCOMPATIBILITY, need to wrap old
  4268 bstring values into Binding.name, or better pass through abstract
  4269 bindings everywhere.  See further src/Pure/General/long_name.ML,
  4270 src/Pure/General/binding.ML and src/Pure/General/name_space.ML
  4271 
  4272 * Result facts (from PureThy.note_thms, ProofContext.note_thms,
  4273 LocalTheory.note etc.) now refer to the *full* internal name, not the
  4274 bstring as before.  INCOMPATIBILITY, not detected by ML type-checking!
  4275 
  4276 * Disposed old type and term read functions (Sign.read_def_typ,
  4277 Sign.read_typ, Sign.read_def_terms, Sign.read_term,
  4278 Thm.read_def_cterms, Thm.read_cterm etc.).  INCOMPATIBILITY, should
  4279 use regular Syntax.read_typ, Syntax.read_term, Syntax.read_typ_global,
  4280 Syntax.read_term_global etc.; see also OldGoals.read_term as last
  4281 resort for legacy applications.
  4282 
  4283 * Disposed old declarations, tactics, tactic combinators that refer to
  4284 the simpset or claset of an implicit theory (such as Addsimps,
  4285 Simp_tac, SIMPSET).  INCOMPATIBILITY, should use @{simpset} etc. in
  4286 embedded ML text, or local_simpset_of with a proper context passed as
  4287 explicit runtime argument.
  4288 
  4289 * Rules and tactics that read instantiations (read_instantiate,
  4290 res_inst_tac, thin_tac, subgoal_tac etc.) now demand a proper proof
  4291 context, which is required for parsing and type-checking.  Moreover,
  4292 the variables are specified as plain indexnames, not string encodings
  4293 thereof.  INCOMPATIBILITY.
  4294 
  4295 * Generic Toplevel.add_hook interface allows to analyze the result of
  4296 transactions.  E.g. see src/Pure/ProofGeneral/proof_general_pgip.ML
  4297 for theorem dependency output of transactions resulting in a new
  4298 theory state.
  4299 
  4300 * ML antiquotations: block-structured compilation context indicated by
  4301 \<lbrace> ... \<rbrace>; additional antiquotation forms:
  4302 
  4303   @{binding name}                         - basic name binding
  4304   @{let ?pat = term}                      - term abbreviation (HO matching)
  4305   @{note name = fact}                     - fact abbreviation
  4306   @{thm fact}                             - singleton fact (with attributes)
  4307   @{thms fact}                            - general fact (with attributes)
  4308   @{lemma prop by method}                 - singleton goal
  4309   @{lemma prop by meth1 meth2}            - singleton goal
  4310   @{lemma prop1 ... propN by method}      - general goal
  4311   @{lemma prop1 ... propN by meth1 meth2} - general goal
  4312   @{lemma (open) ...}                     - open derivation
  4313 
  4314 
  4315 *** System ***
  4316 
  4317 * The Isabelle "emacs" tool provides a specific interface to invoke
  4318 Proof General / Emacs, with more explicit failure if that is not
  4319 installed (the old isabelle-interface script silently falls back on
  4320 isabelle-process).  The PROOFGENERAL_HOME setting determines the
  4321 installation location of the Proof General distribution.
  4322 
  4323 * Isabelle/lib/classes/Pure.jar provides basic support to integrate
  4324 the Isabelle process into a JVM/Scala application.  See
  4325 Isabelle/lib/jedit/plugin for a minimal example.  (The obsolete Java
  4326 process wrapper has been discontinued.)
  4327 
  4328 * Added homegrown Isabelle font with unicode layout, see lib/fonts.
  4329 
  4330 * Various status messages (with exact source position information) are
  4331 emitted, if proper markup print mode is enabled.  This allows
  4332 user-interface components to provide detailed feedback on internal
  4333 prover operations.
  4334 
  4335 
  4336 
  4337 New in Isabelle2008 (June 2008)
  4338 -------------------------------
  4339 
  4340 *** General ***
  4341 
  4342 * The Isabelle/Isar Reference Manual (isar-ref) has been reorganized
  4343 and updated, with formally checked references as hyperlinks.
  4344 
  4345 * Theory loader: use_thy (and similar operations) no longer set the
  4346 implicit ML context, which was occasionally hard to predict and in
  4347 conflict with concurrency.  INCOMPATIBILITY, use ML within Isar which
  4348 provides a proper context already.
  4349 
  4350 * Theory loader: old-style ML proof scripts being *attached* to a thy
  4351 file are no longer supported.  INCOMPATIBILITY, regular 'uses' and
  4352 'use' within a theory file will do the job.
  4353 
  4354 * Name space merge now observes canonical order, i.e. the second space
  4355 is inserted into the first one, while existing entries in the first
  4356 space take precedence.  INCOMPATIBILITY in rare situations, may try to
  4357 swap theory imports.
  4358 
  4359 * Syntax: symbol \<chi> is now considered a letter.  Potential
  4360 INCOMPATIBILITY in identifier syntax etc.
  4361 
  4362 * Outer syntax: string tokens no longer admit escaped white space,
  4363 which was an accidental (undocumented) feature.  INCOMPATIBILITY, use
  4364 white space without escapes.
  4365 
  4366 * Outer syntax: string tokens may contain arbitrary character codes
  4367 specified via 3 decimal digits (as in SML).  E.g. "foo\095bar" for
  4368 "foo_bar".
  4369 
  4370 
  4371 *** Pure ***
  4372 
  4373 * Context-dependent token translations.  Default setup reverts locally
  4374 fixed variables, and adds hilite markup for undeclared frees.
  4375 
  4376 * Unused theorems can be found using the new command 'unused_thms'.
  4377 There are three ways of invoking it:
  4378 
  4379 (1) unused_thms
  4380      Only finds unused theorems in the current theory.
  4381 
  4382 (2) unused_thms thy_1 ... thy_n -
  4383      Finds unused theorems in the current theory and all of its ancestors,
  4384      excluding the theories thy_1 ... thy_n and all of their ancestors.
  4385 
  4386 (3) unused_thms thy_1 ... thy_n - thy'_1 ... thy'_m
  4387      Finds unused theorems in the theories thy'_1 ... thy'_m and all of
  4388      their ancestors, excluding the theories thy_1 ... thy_n and all of
  4389      their ancestors.
  4390 
  4391 In order to increase the readability of the list produced by
  4392 unused_thms, theorems that have been created by a particular instance
  4393 of a theory command such as 'inductive' or 'function' are considered
  4394 to belong to the same "group", meaning that if at least one theorem in
  4395 this group is used, the other theorems in the same group are no longer
  4396 reported as unused.  Moreover, if all theorems in the group are
  4397 unused, only one theorem in the group is displayed.
  4398 
  4399 Note that proof objects have to be switched on in order for
  4400 unused_thms to work properly (i.e. !proofs must be >= 1, which is
  4401 usually the case when using Proof General with the default settings).
  4402 
  4403 * Authentic naming of facts disallows ad-hoc overwriting of previous
  4404 theorems within the same name space.  INCOMPATIBILITY, need to remove
  4405 duplicate fact bindings, or even accidental fact duplications.  Note
  4406 that tools may maintain dynamically scoped facts systematically, using
  4407 PureThy.add_thms_dynamic.
  4408 
  4409 * Command 'hide' now allows to hide from "fact" name space as well.
  4410 
  4411 * Eliminated destructive theorem database, simpset, claset, and
  4412 clasimpset.  Potential INCOMPATIBILITY, really need to observe linear
  4413 update of theories within ML code.
  4414 
  4415 * Eliminated theory ProtoPure and CPure, leaving just one Pure theory.
  4416 INCOMPATIBILITY, object-logics depending on former Pure require
  4417 additional setup PureThy.old_appl_syntax_setup; object-logics
  4418 depending on former CPure need to refer to Pure.
  4419 
  4420 * Commands 'use' and 'ML' are now purely functional, operating on
  4421 theory/local_theory.  Removed former 'ML_setup' (on theory), use 'ML'
  4422 instead.  Added 'ML_val' as mere diagnostic replacement for 'ML'.
  4423 INCOMPATIBILITY.
  4424 
  4425 * Command 'setup': discontinued implicit version with ML reference.
  4426 
  4427 * Instantiation target allows for simultaneous specification of class
  4428 instance operations together with an instantiation proof.
  4429 Type-checking phase allows to refer to class operations uniformly.
  4430 See src/HOL/Complex/Complex.thy for an Isar example and
  4431 src/HOL/Library/Eval.thy for an ML example.
  4432 
  4433 * Indexing of literal facts: be more serious about including only
  4434 facts from the visible specification/proof context, but not the
  4435 background context (locale etc.).  Affects `prop` notation and method
  4436 "fact".  INCOMPATIBILITY: need to name facts explicitly in rare
  4437 situations.
  4438 
  4439 * Method "cases", "induct", "coinduct": removed obsolete/undocumented
  4440 "(open)" option, which used to expose internal bound variables to the
  4441 proof text.
  4442 
  4443 * Isar statements: removed obsolete case "rule_context".
  4444 INCOMPATIBILITY, better use explicit fixes/assumes.
  4445 
  4446 * Locale proofs: default proof step now includes 'unfold_locales';
  4447 hence 'proof' without argument may be used to unfold locale
  4448 predicates.
  4449 
  4450 
  4451 *** Document preparation ***
  4452 
  4453 * Simplified pdfsetup.sty: color/hyperref is used unconditionally for
  4454 both pdf and dvi (hyperlinks usually work in xdvi as well); removed
  4455 obsolete thumbpdf setup (contemporary PDF viewers do this on the
  4456 spot); renamed link color from "darkblue" to "linkcolor" (default
  4457 value unchanged, can be redefined via \definecolor); no longer sets
  4458 "a4paper" option (unnecessary or even intrusive).
  4459 
  4460 * Antiquotation @{lemma A method} proves proposition A by the given
  4461 method (either a method name or a method name plus (optional) method
  4462 arguments in parentheses) and prints A just like @{prop A}.
  4463 
  4464 
  4465 *** HOL ***
  4466 
  4467 * New primrec package.  Specification syntax conforms in style to
  4468 definition/function/....  No separate induction rule is provided.  The
  4469 "primrec" command distinguishes old-style and new-style specifications
  4470 by syntax.  The former primrec package is now named OldPrimrecPackage.
  4471 When adjusting theories, beware: constants stemming from new-style
  4472 primrec specifications have authentic syntax.
  4473 
  4474 * Metis prover is now an order of magnitude faster, and also works
  4475 with multithreading.
  4476 
  4477 * Metis: the maximum number of clauses that can be produced from a
  4478 theorem is now given by the attribute max_clauses.  Theorems that
  4479 exceed this number are ignored, with a warning printed.
  4480 
  4481 * Sledgehammer no longer produces structured proofs by default. To
  4482 enable, declare [[sledgehammer_full = true]].  Attributes
  4483 reconstruction_modulus, reconstruction_sorts renamed
  4484 sledgehammer_modulus, sledgehammer_sorts.  INCOMPATIBILITY.
  4485 
  4486 * Method "induct_scheme" derives user-specified induction rules
  4487 from well-founded induction and completeness of patterns. This factors
  4488 out some operations that are done internally by the function package
  4489 and makes them available separately.  See
  4490 src/HOL/ex/Induction_Scheme.thy for examples.
  4491 
  4492 * More flexible generation of measure functions for termination
  4493 proofs: Measure functions can be declared by proving a rule of the
  4494 form "is_measure f" and giving it the [measure_function] attribute.
  4495 The "is_measure" predicate is logically meaningless (always true), and
  4496 just guides the heuristic.  To find suitable measure functions, the
  4497 termination prover sets up the goal "is_measure ?f" of the appropriate
  4498 type and generates all solutions by prolog-style backwards proof using
  4499 the declared rules.
  4500 
  4501 This setup also deals with rules like 
  4502 
  4503   "is_measure f ==> is_measure (list_size f)"
  4504 
  4505 which accommodates nested datatypes that recurse through lists.
  4506 Similar rules are predeclared for products and option types.
  4507 
  4508 * Turned the type of sets "'a set" into an abbreviation for "'a => bool"
  4509 
  4510   INCOMPATIBILITIES:
  4511 
  4512   - Definitions of overloaded constants on sets have to be replaced by
  4513     definitions on => and bool.
  4514 
  4515   - Some definitions of overloaded operators on sets can now be proved
  4516     using the definitions of the operators on => and bool.  Therefore,
  4517     the following theorems have been renamed:
  4518 
  4519       subset_def   -> subset_eq
  4520       psubset_def  -> psubset_eq
  4521       set_diff_def -> set_diff_eq
  4522       Compl_def    -> Compl_eq
  4523       Sup_set_def  -> Sup_set_eq
  4524       Inf_set_def  -> Inf_set_eq
  4525       sup_set_def  -> sup_set_eq
  4526       inf_set_def  -> inf_set_eq
  4527 
  4528   - Due to the incompleteness of the HO unification algorithm, some
  4529     rules such as subst may require manual instantiation, if some of
  4530     the unknowns in the rule is a set.
  4531 
  4532   - Higher order unification and forward proofs:
  4533     The proof pattern
  4534 
  4535       have "P (S::'a set)" <...>
  4536       then have "EX S. P S" ..
  4537 
  4538     no longer works (due to the incompleteness of the HO unification
  4539     algorithm) and must be replaced by the pattern
  4540 
  4541       have "EX S. P S"
  4542       proof
  4543         show "P S" <...>
  4544       qed
  4545 
  4546   - Calculational reasoning with subst (or similar rules):
  4547     The proof pattern
  4548 
  4549       have "P (S::'a set)" <...>
  4550       also have "S = T" <...>
  4551       finally have "P T" .
  4552 
  4553     no longer works (for similar reasons as the previous example) and
  4554     must be replaced by something like
  4555 
  4556       have "P (S::'a set)" <...>
  4557       moreover have "S = T" <...>
  4558       ultimately have "P T" by simp
  4559 
  4560   - Tactics or packages written in ML code:
  4561     Code performing pattern matching on types via
  4562 
  4563       Type ("set", [T]) => ...
  4564 
  4565     must be rewritten. Moreover, functions like strip_type or
  4566     binder_types no longer return the right value when applied to a
  4567     type of the form
  4568 
  4569       T1 => ... => Tn => U => bool
  4570 
  4571     rather than
  4572 
  4573       T1 => ... => Tn => U set
  4574 
  4575 * Merged theories Wellfounded_Recursion, Accessible_Part and
  4576 Wellfounded_Relations to theory Wellfounded.
  4577 
  4578 * Explicit class "eq" for executable equality.  INCOMPATIBILITY.
  4579 
  4580 * Class finite no longer treats UNIV as class parameter.  Use class
  4581 enum from theory Library/Enum instead to achieve a similar effect.
  4582 INCOMPATIBILITY.
  4583 
  4584 * Theory List: rule list_induct2 now has explicitly named cases "Nil"
  4585 and "Cons".  INCOMPATIBILITY.
  4586 
  4587 * HOL (and FOL): renamed variables in rules imp_elim and swap.
  4588 Potential INCOMPATIBILITY.
  4589 
  4590 * Theory Product_Type: duplicated lemmas split_Pair_apply and
  4591 injective_fst_snd removed, use split_eta and prod_eqI instead.
  4592 Renamed upd_fst to apfst and upd_snd to apsnd.  INCOMPATIBILITY.
  4593 
  4594 * Theory Nat: removed redundant lemmas that merely duplicate lemmas of
  4595 the same name in theory Orderings:
  4596 
  4597   less_trans
  4598   less_linear
  4599   le_imp_less_or_eq
  4600   le_less_trans
  4601   less_le_trans
  4602   less_not_sym
  4603   less_asym
  4604 
  4605 Renamed less_imp_le to less_imp_le_nat, and less_irrefl to
  4606 less_irrefl_nat.  Potential INCOMPATIBILITY due to more general types
  4607 and different variable names.
  4608 
  4609 * Library/Option_ord.thy: Canonical order on option type.
  4610 
  4611 * Library/RBT.thy: Red-black trees, an efficient implementation of
  4612 finite maps.
  4613 
  4614 * Library/Countable.thy: Type class for countable types.
  4615 
  4616 * Theory Int: The representation of numerals has changed.  The infix
  4617 operator BIT and the bit datatype with constructors B0 and B1 have
  4618 disappeared.  INCOMPATIBILITY, use "Int.Bit0 x" and "Int.Bit1 y" in
  4619 place of "x BIT bit.B0" and "y BIT bit.B1", respectively.  Theorems
  4620 involving BIT, B0, or B1 have been renamed with "Bit0" or "Bit1"
  4621 accordingly.
  4622 
  4623 * Theory Nat: definition of <= and < on natural numbers no longer
  4624 depend on well-founded relations.  INCOMPATIBILITY.  Definitions
  4625 le_def and less_def have disappeared.  Consider lemmas not_less
  4626 [symmetric, where ?'a = nat] and less_eq [symmetric] instead.
  4627 
  4628 * Theory Finite_Set: locales ACf, ACe, ACIf, ACIfSL and ACIfSLlin
  4629 (whose purpose mainly is for various fold_set functionals) have been
  4630 abandoned in favor of the existing algebraic classes
  4631 ab_semigroup_mult, comm_monoid_mult, ab_semigroup_idem_mult,
  4632 lower_semilattice (resp. upper_semilattice) and linorder.
  4633 INCOMPATIBILITY.
  4634 
  4635 * Theory Transitive_Closure: induct and cases rules now declare proper
  4636 case_names ("base" and "step").  INCOMPATIBILITY.
  4637 
  4638 * Theorem Inductive.lfp_ordinal_induct generalized to complete
  4639 lattices.  The form set-specific version is available as
  4640 Inductive.lfp_ordinal_induct_set.
  4641 
  4642 * Renamed theorems "power.simps" to "power_int.simps".
  4643 INCOMPATIBILITY.
  4644 
  4645 * Class semiring_div provides basic abstract properties of semirings
  4646 with division and modulo operations.  Subsumes former class dvd_mod.
  4647 
  4648 * Merged theories IntDef, Numeral and IntArith into unified theory
  4649 Int.  INCOMPATIBILITY.
  4650 
  4651 * Theory Library/Code_Index: type "index" now represents natural
  4652 numbers rather than integers.  INCOMPATIBILITY.
  4653 
  4654 * New class "uminus" with operation "uminus" (split of from class
  4655 "minus" which now only has operation "minus", binary).
  4656 INCOMPATIBILITY.
  4657 
  4658 * Constants "card", "internal_split", "option_map" now with authentic
  4659 syntax.  INCOMPATIBILITY.
  4660 
  4661 * Definitions subset_def, psubset_def, set_diff_def, Compl_def,
  4662 le_bool_def, less_bool_def, le_fun_def, less_fun_def, inf_bool_def,
  4663 sup_bool_def, Inf_bool_def, Sup_bool_def, inf_fun_def, sup_fun_def,
  4664 Inf_fun_def, Sup_fun_def, inf_set_def, sup_set_def, Inf_set_def,
  4665 Sup_set_def, le_def, less_def, option_map_def now with object
  4666 equality.  INCOMPATIBILITY.
  4667 
  4668 * Records. Removed K_record, and replaced it by pure lambda term
  4669 %x. c. The simplifier setup is now more robust against eta expansion.
  4670 INCOMPATIBILITY: in cases explicitly referring to K_record.
  4671 
  4672 * Library/Multiset: {#a, b, c#} abbreviates {#a#} + {#b#} + {#c#}.
  4673 
  4674 * Library/ListVector: new theory of arithmetic vector operations.
  4675 
  4676 * Library/Order_Relation: new theory of various orderings as sets of
  4677 pairs.  Defines preorders, partial orders, linear orders and
  4678 well-orders on sets and on types.
  4679 
  4680 
  4681 *** ZF ***
  4682 
  4683 * Renamed some theories to allow to loading both ZF and HOL in the
  4684 same session:
  4685 
  4686   Datatype  -> Datatype_ZF
  4687   Inductive -> Inductive_ZF
  4688   Int       -> Int_ZF
  4689   IntDiv    -> IntDiv_ZF
  4690   Nat       -> Nat_ZF
  4691   List      -> List_ZF
  4692   Main      -> Main_ZF
  4693 
  4694 INCOMPATIBILITY: ZF theories that import individual theories below
  4695 Main might need to be adapted.  Regular theory Main is still
  4696 available, as trivial extension of Main_ZF.
  4697 
  4698 
  4699 *** ML ***
  4700 
  4701 * ML within Isar: antiquotation @{const name} or @{const
  4702 name(typargs)} produces statically-checked Const term.
  4703 
  4704 * Functor NamedThmsFun: data is available to the user as dynamic fact
  4705 (of the same name).  Removed obsolete print command.
  4706 
  4707 * Removed obsolete "use_legacy_bindings" function.
  4708 
  4709 * The ``print mode'' is now a thread-local value derived from a global
  4710 template (the former print_mode reference), thus access becomes
  4711 non-critical.  The global print_mode reference is for session
  4712 management only; user-code should use print_mode_value,
  4713 print_mode_active, PrintMode.setmp etc.  INCOMPATIBILITY.
  4714 
  4715 * Functions system/system_out provide a robust way to invoke external
  4716 shell commands, with propagation of interrupts (requires Poly/ML
  4717 5.2.1).  Do not use OS.Process.system etc. from the basis library!
  4718 
  4719 
  4720 *** System ***
  4721 
  4722 * Default settings: PROOFGENERAL_OPTIONS no longer impose xemacs ---
  4723 in accordance with Proof General 3.7, which prefers GNU emacs.
  4724 
  4725 * isatool tty runs Isabelle process with plain tty interaction;
  4726 optional line editor may be specified via ISABELLE_LINE_EDITOR
  4727 setting, the default settings attempt to locate "ledit" and "rlwrap".
  4728 
  4729 * isatool browser now works with Cygwin as well, using general
  4730 "javapath" function defined in Isabelle process environment.
  4731 
  4732 * YXML notation provides a simple and efficient alternative to
  4733 standard XML transfer syntax.  See src/Pure/General/yxml.ML and
  4734 isatool yxml as described in the Isabelle system manual.
  4735 
  4736 * JVM class isabelle.IsabelleProcess (located in Isabelle/lib/classes)
  4737 provides general wrapper for managing an Isabelle process in a robust
  4738 fashion, with ``cooked'' output from stdin/stderr.
  4739 
  4740 * Rudimentary Isabelle plugin for jEdit (see Isabelle/lib/jedit),
  4741 based on Isabelle/JVM process wrapper (see Isabelle/lib/classes).
  4742 
  4743 * Removed obsolete THIS_IS_ISABELLE_BUILD feature.  NB: the documented
  4744 way of changing the user's settings is via
  4745 ISABELLE_HOME_USER/etc/settings, which is a fully featured bash
  4746 script.
  4747 
  4748 * Multithreading.max_threads := 0 refers to the number of actual CPU
  4749 cores of the underlying machine, which is a good starting point for
  4750 optimal performance tuning.  The corresponding usedir option -M allows
  4751 "max" as an alias for "0".  WARNING: does not work on certain versions
  4752 of Mac OS (with Poly/ML 5.1).
  4753 
  4754 * isabelle-process: non-ML sessions are run with "nice", to reduce the
  4755 adverse effect of Isabelle flooding interactive front-ends (notably
  4756 ProofGeneral / XEmacs).
  4757 
  4758 
  4759 
  4760 New in Isabelle2007 (November 2007)
  4761 -----------------------------------
  4762 
  4763 *** General ***
  4764 
  4765 * More uniform information about legacy features, notably a
  4766 warning/error of "Legacy feature: ...", depending on the state of the
  4767 tolerate_legacy_features flag (default true). FUTURE INCOMPATIBILITY:
  4768 legacy features will disappear eventually.
  4769 
  4770 * Theory syntax: the header format ``theory A = B + C:'' has been
  4771 discontinued in favour of ``theory A imports B C begin''.  Use isatool
  4772 fixheaders to convert existing theory files.  INCOMPATIBILITY.
  4773 
  4774 * Theory syntax: the old non-Isar theory file format has been
  4775 discontinued altogether.  Note that ML proof scripts may still be used
  4776 with Isar theories; migration is usually quite simple with the ML
  4777 function use_legacy_bindings.  INCOMPATIBILITY.
  4778 
  4779 * Theory syntax: some popular names (e.g. 'class', 'declaration',
  4780 'fun', 'help', 'if') are now keywords.  INCOMPATIBILITY, use double
  4781 quotes.
  4782 
  4783 * Theory loader: be more serious about observing the static theory
  4784 header specifications (including optional directories), but not the
  4785 accidental file locations of previously successful loads.  The strict
  4786 update policy of former update_thy is now already performed by
  4787 use_thy, so the former has been removed; use_thys updates several
  4788 theories simultaneously, just as 'imports' within a theory header
  4789 specification, but without merging the results.  Potential
  4790 INCOMPATIBILITY: may need to refine theory headers and commands
  4791 ROOT.ML which depend on load order.
  4792 
  4793 * Theory loader: optional support for content-based file
  4794 identification, instead of the traditional scheme of full physical
  4795 path plus date stamp; configured by the ISABELLE_FILE_IDENT setting
  4796 (cf. the system manual).  The new scheme allows to work with
  4797 non-finished theories in persistent session images, such that source
  4798 files may be moved later on without requiring reloads.
  4799 
  4800 * Theory loader: old-style ML proof scripts being *attached* to a thy
  4801 file (with the same base name as the theory) are considered a legacy
  4802 feature, which will disappear eventually. Even now, the theory loader
  4803 no longer maintains dependencies on such files.
  4804 
  4805 * Syntax: the scope for resolving ambiguities via type-inference is
  4806 now limited to individual terms, instead of whole simultaneous
  4807 specifications as before. This greatly reduces the complexity of the
  4808 syntax module and improves flexibility by separating parsing and
  4809 type-checking. INCOMPATIBILITY: additional type-constraints (explicit
  4810 'fixes' etc.) are required in rare situations.
  4811 
  4812 * Syntax: constants introduced by new-style packages ('definition',
  4813 'abbreviation' etc.) are passed through the syntax module in
  4814 ``authentic mode''. This means that associated mixfix annotations
  4815 really stick to such constants, independently of potential name space
  4816 ambiguities introduced later on. INCOMPATIBILITY: constants in parse
  4817 trees are represented slightly differently, may need to adapt syntax
  4818 translations accordingly. Use CONST marker in 'translations' and
  4819 @{const_syntax} antiquotation in 'parse_translation' etc.
  4820 
  4821 * Legacy goal package: reduced interface to the bare minimum required
  4822 to keep existing proof scripts running.  Most other user-level
  4823 functions are now part of the OldGoals structure, which is *not* open
  4824 by default (consider isatool expandshort before open OldGoals).
  4825 Removed top_sg, prin, printyp, pprint_term/typ altogether, because
  4826 these tend to cause confusion about the actual goal (!) context being
  4827 used here, which is not necessarily the same as the_context().
  4828 
  4829 * Command 'find_theorems': supports "*" wild-card in "name:"
  4830 criterion; "with_dups" option.  Certain ProofGeneral versions might
  4831 support a specific search form (see ProofGeneral/CHANGES).
  4832 
  4833 * The ``prems limit'' option (cf. ProofContext.prems_limit) is now -1
  4834 by default, which means that "prems" (and also "fixed variables") are
  4835 suppressed from proof state output.  Note that the ProofGeneral
  4836 settings mechanism allows to change and save options persistently, but
  4837 older versions of Isabelle will fail to start up if a negative prems
  4838 limit is imposed.
  4839 
  4840 * Local theory targets may be specified by non-nested blocks of
  4841 ``context/locale/class ... begin'' followed by ``end''.  The body may
  4842 contain definitions, theorems etc., including any derived mechanism
  4843 that has been implemented on top of these primitives.  This concept
  4844 generalizes the existing ``theorem (in ...)'' towards more versatility
  4845 and scalability.
  4846 
  4847 * Proof General interface: proper undo of final 'end' command;
  4848 discontinued Isabelle/classic mode (ML proof scripts).
  4849 
  4850 
  4851 *** Document preparation ***
  4852 
  4853 * Added antiquotation @{theory name} which prints the given name,
  4854 after checking that it refers to a valid ancestor theory in the
  4855 current context.
  4856 
  4857 * Added antiquotations @{ML_type text} and @{ML_struct text} which
  4858 check the given source text as ML type/structure, printing verbatim.
  4859 
  4860 * Added antiquotation @{abbrev "c args"} which prints the abbreviation
  4861 "c args == rhs" given in the current context.  (Any number of
  4862 arguments may be given on the LHS.)
  4863 
  4864 
  4865 *** Pure ***
  4866 
  4867 * The 'class' package offers a combination of axclass and locale to
  4868 achieve Haskell-like type classes in Isabelle.  Definitions and
  4869 theorems within a class context produce both relative results (with
  4870 implicit parameters according to the locale context), and polymorphic
  4871 constants with qualified polymorphism (according to the class
  4872 context).  Within the body context of a 'class' target, a separate
  4873 syntax layer ("user space type system") takes care of converting
  4874 between global polymorphic consts and internal locale representation.
  4875 See src/HOL/ex/Classpackage.thy for examples (as well as main HOL).
  4876 "isatool doc classes" provides a tutorial.
  4877 
  4878 * Generic code generator framework allows to generate executable
  4879 code for ML and Haskell (including Isabelle classes).  A short usage
  4880 sketch:
  4881 
  4882     internal compilation:
  4883         export_code <list of constants (term syntax)> in SML
  4884     writing SML code to a file:
  4885         export_code <list of constants (term syntax)> in SML <filename>
  4886     writing OCaml code to a file:
  4887         export_code <list of constants (term syntax)> in OCaml <filename>
  4888     writing Haskell code to a bunch of files:
  4889         export_code <list of constants (term syntax)> in Haskell <filename>
  4890 
  4891     evaluating closed propositions to True/False using code generation:
  4892         method ``eval''
  4893 
  4894 Reasonable default setup of framework in HOL.
  4895 
  4896 Theorem attributs for selecting and transforming function equations theorems:
  4897 
  4898     [code fun]:        select a theorem as function equation for a specific constant
  4899     [code fun del]:    deselect a theorem as function equation for a specific constant
  4900     [code inline]:     select an equation theorem for unfolding (inlining) in place
  4901     [code inline del]: deselect an equation theorem for unfolding (inlining) in place
  4902 
  4903 User-defined serializations (target in {SML, OCaml, Haskell}):
  4904 
  4905     code_const <and-list of constants (term syntax)>
  4906       {(target) <and-list of const target syntax>}+
  4907 
  4908     code_type <and-list of type constructors>
  4909       {(target) <and-list of type target syntax>}+
  4910 
  4911     code_instance <and-list of instances>
  4912       {(target)}+
  4913         where instance ::= <type constructor> :: <class>
  4914 
  4915     code_class <and_list of classes>
  4916       {(target) <and-list of class target syntax>}+
  4917         where class target syntax ::= <class name> {where {<classop> == <target syntax>}+}?
  4918 
  4919 code_instance and code_class only are effective to target Haskell.
  4920 
  4921 For example usage see src/HOL/ex/Codegenerator.thy and
  4922 src/HOL/ex/Codegenerator_Pretty.thy.  A separate tutorial on code
  4923 generation from Isabelle/HOL theories is available via "isatool doc
  4924 codegen".
  4925 
  4926 * Code generator: consts in 'consts_code' Isar commands are now
  4927 referred to by usual term syntax (including optional type
  4928 annotations).
  4929 
  4930 * Command 'no_translations' removes translation rules from theory
  4931 syntax.
  4932 
  4933 * Overloaded definitions are now actually checked for acyclic
  4934 dependencies.  The overloading scheme is slightly more general than
  4935 that of Haskell98, although Isabelle does not demand an exact
  4936 correspondence to type class and instance declarations.
  4937 INCOMPATIBILITY, use ``defs (unchecked overloaded)'' to admit more
  4938 exotic versions of overloading -- at the discretion of the user!
  4939 
  4940 Polymorphic constants are represented via type arguments, i.e. the
  4941 instantiation that matches an instance against the most general
  4942 declaration given in the signature.  For example, with the declaration
  4943 c :: 'a => 'a => 'a, an instance c :: nat => nat => nat is represented
  4944 as c(nat).  Overloading is essentially simultaneous structural
  4945 recursion over such type arguments.  Incomplete specification patterns
  4946 impose global constraints on all occurrences, e.g. c('a * 'a) on the
  4947 LHS means that more general c('a * 'b) will be disallowed on any RHS.
  4948 Command 'print_theory' outputs the normalized system of recursive
  4949 equations, see section "definitions".
  4950 
  4951 * Configuration options are maintained within the theory or proof
  4952 context (with name and type bool/int/string), providing a very simple
  4953 interface to a poor-man's version of general context data.  Tools may
  4954 declare options in ML (e.g. using Attrib.config_int) and then refer to
  4955 these values using Config.get etc.  Users may change options via an
  4956 associated attribute of the same name.  This form of context
  4957 declaration works particularly well with commands 'declare' or
  4958 'using', for example ``declare [[foo = 42]]''.  Thus it has become
  4959 very easy to avoid global references, which would not observe Isar
  4960 toplevel undo/redo and fail to work with multithreading.
  4961 
  4962 Various global ML references of Pure and HOL have been turned into
  4963 configuration options:
  4964 
  4965   Unify.search_bound		unify_search_bound
  4966   Unify.trace_bound		unify_trace_bound
  4967   Unify.trace_simp		unify_trace_simp
  4968   Unify.trace_types		unify_trace_types
  4969   Simplifier.simp_depth_limit	simp_depth_limit
  4970   Blast.depth_limit		blast_depth_limit
  4971   DatatypeProp.dtK		datatype_distinctness_limit
  4972   fast_arith_neq_limit  	fast_arith_neq_limit
  4973   fast_arith_split_limit	fast_arith_split_limit
  4974 
  4975 * Named collections of theorems may be easily installed as context
  4976 data using the functor NamedThmsFun (see also
  4977 src/Pure/Tools/named_thms.ML).  The user may add or delete facts via
  4978 attributes; there is also a toplevel print command.  This facility is
  4979 just a common case of general context data, which is the preferred way
  4980 for anything more complex than just a list of facts in canonical
  4981 order.
  4982 
  4983 * Isar: command 'declaration' augments a local theory by generic
  4984 declaration functions written in ML.  This enables arbitrary content
  4985 being added to the context, depending on a morphism that tells the
  4986 difference of the original declaration context wrt. the application
  4987 context encountered later on.
  4988 
  4989 * Isar: proper interfaces for simplification procedures.  Command
  4990 'simproc_setup' declares named simprocs (with match patterns, and body
  4991 text in ML).  Attribute "simproc" adds/deletes simprocs in the current
  4992 context.  ML antiquotation @{simproc name} retrieves named simprocs.
  4993 
  4994 * Isar: an extra pair of brackets around attribute declarations
  4995 abbreviates a theorem reference involving an internal dummy fact,
  4996 which will be ignored later --- only the effect of the attribute on
  4997 the background context will persist.  This form of in-place
  4998 declarations is particularly useful with commands like 'declare' and
  4999 'using', for example ``have A using [[simproc a]] by simp''.
  5000 
  5001 * Isar: method "assumption" (and implicit closing of subproofs) now
  5002 takes simple non-atomic goal assumptions into account: after applying
  5003 an assumption as a rule the resulting subgoals are solved by atomic
  5004 assumption steps.  This is particularly useful to finish 'obtain'
  5005 goals, such as "!!x. (!!x. P x ==> thesis) ==> P x ==> thesis",
  5006 without referring to the original premise "!!x. P x ==> thesis" in the
  5007 Isar proof context.  POTENTIAL INCOMPATIBILITY: method "assumption" is
  5008 more permissive.
  5009 
  5010 * Isar: implicit use of prems from the Isar proof context is
  5011 considered a legacy feature.  Common applications like ``have A .''
  5012 may be replaced by ``have A by fact'' or ``note `A`''.  In general,
  5013 referencing facts explicitly here improves readability and
  5014 maintainability of proof texts.
  5015 
  5016 * Isar: improper proof element 'guess' is like 'obtain', but derives
  5017 the obtained context from the course of reasoning!  For example:
  5018 
  5019   assume "EX x y. A x & B y"   -- "any previous fact"
  5020   then guess x and y by clarify
  5021 
  5022 This technique is potentially adventurous, depending on the facts and
  5023 proof tools being involved here.
  5024 
  5025 * Isar: known facts from the proof context may be specified as literal
  5026 propositions, using ASCII back-quote syntax.  This works wherever
  5027 named facts used to be allowed so far, in proof commands, proof
  5028 methods, attributes etc.  Literal facts are retrieved from the context
  5029 according to unification of type and term parameters.  For example,
  5030 provided that "A" and "A ==> B" and "!!x. P x ==> Q x" are known
  5031 theorems in the current context, then these are valid literal facts:
  5032 `A` and `A ==> B` and `!!x. P x ==> Q x" as well as `P a ==> Q a` etc.
  5033 
  5034 There is also a proof method "fact" which does the same composition
  5035 for explicit goal states, e.g. the following proof texts coincide with
  5036 certain special cases of literal facts:
  5037 
  5038   have "A" by fact                 ==  note `A`
  5039   have "A ==> B" by fact           ==  note `A ==> B`
  5040   have "!!x. P x ==> Q x" by fact  ==  note `!!x. P x ==> Q x`
  5041   have "P a ==> Q a" by fact       ==  note `P a ==> Q a`
  5042 
  5043 * Isar: ":" (colon) is no longer a symbolic identifier character in
  5044 outer syntax.  Thus symbolic identifiers may be used without
  5045 additional white space in declarations like this: ``assume *: A''.
  5046 
  5047 * Isar: 'print_facts' prints all local facts of the current context,
  5048 both named and unnamed ones.
  5049 
  5050 * Isar: 'def' now admits simultaneous definitions, e.g.:
  5051 
  5052   def x == "t" and y == "u"
  5053 
  5054 * Isar: added command 'unfolding', which is structurally similar to
  5055 'using', but affects both the goal state and facts by unfolding given
  5056 rewrite rules.  Thus many occurrences of the 'unfold' method or
  5057 'unfolded' attribute may be replaced by first-class proof text.
  5058 
  5059 * Isar: methods 'unfold' / 'fold', attributes 'unfolded' / 'folded',
  5060 and command 'unfolding' now all support object-level equalities
  5061 (potentially conditional).  The underlying notion of rewrite rule is
  5062 analogous to the 'rule_format' attribute, but *not* that of the
  5063 Simplifier (which is usually more generous).
  5064 
  5065 * Isar: the new attribute [rotated n] (default n = 1) rotates the
  5066 premises of a theorem by n. Useful in conjunction with drule.
  5067 
  5068 * Isar: the goal restriction operator [N] (default N = 1) evaluates a
  5069 method expression within a sandbox consisting of the first N
  5070 sub-goals, which need to exist.  For example, ``simp_all [3]''
  5071 simplifies the first three sub-goals, while (rule foo, simp_all)[]
  5072 simplifies all new goals that emerge from applying rule foo to the
  5073 originally first one.
  5074 
  5075 * Isar: schematic goals are no longer restricted to higher-order
  5076 patterns; e.g. ``lemma "?P(?x)" by (rule TrueI)'' now works as
  5077 expected.
  5078 
  5079 * Isar: the conclusion of a long theorem statement is now either
  5080 'shows' (a simultaneous conjunction, as before), or 'obtains'
  5081 (essentially a disjunction of cases with local parameters and
  5082 assumptions).  The latter allows to express general elimination rules
  5083 adequately; in this notation common elimination rules look like this:
  5084 
  5085   lemma exE:    -- "EX x. P x ==> (!!x. P x ==> thesis) ==> thesis"
  5086     assumes "EX x. P x"
  5087     obtains x where "P x"
  5088 
  5089   lemma conjE:  -- "A & B ==> (A ==> B ==> thesis) ==> thesis"
  5090     assumes "A & B"
  5091     obtains A and B
  5092 
  5093   lemma disjE:  -- "A | B ==> (A ==> thesis) ==> (B ==> thesis) ==> thesis"
  5094     assumes "A | B"
  5095     obtains
  5096       A
  5097     | B
  5098 
  5099 The subsequent classical rules even refer to the formal "thesis"
  5100 explicitly:
  5101 
  5102   lemma classical:     -- "(~ thesis ==> thesis) ==> thesis"
  5103     obtains "~ thesis"
  5104 
  5105   lemma Peirce's_Law:  -- "((thesis ==> something) ==> thesis) ==> thesis"
  5106     obtains "thesis ==> something"
  5107 
  5108 The actual proof of an 'obtains' statement is analogous to that of the
  5109 Isar proof element 'obtain', only that there may be several cases.
  5110 Optional case names may be specified in parentheses; these will be
  5111 available both in the present proof and as annotations in the
  5112 resulting rule, for later use with the 'cases' method (cf. attribute
  5113 case_names).
  5114 
  5115 * Isar: the assumptions of a long theorem statement are available as
  5116 "assms" fact in the proof context.  This is more appropriate than the
  5117 (historical) "prems", which refers to all assumptions of the current
  5118 context, including those from the target locale, proof body etc.
  5119 
  5120 * Isar: 'print_statement' prints theorems from the current theory or
  5121 proof context in long statement form, according to the syntax of a
  5122 top-level lemma.
  5123 
  5124 * Isar: 'obtain' takes an optional case name for the local context
  5125 introduction rule (default "that").
  5126 
  5127 * Isar: removed obsolete 'concl is' patterns.  INCOMPATIBILITY, use
  5128 explicit (is "_ ==> ?foo") in the rare cases where this still happens
  5129 to occur.
  5130 
  5131 * Pure: syntax "CONST name" produces a fully internalized constant
  5132 according to the current context.  This is particularly useful for
  5133 syntax translations that should refer to internal constant
  5134 representations independently of name spaces.
  5135 
  5136 * Pure: syntax constant for foo (binder "FOO ") is called "foo_binder"
  5137 instead of "FOO ". This allows multiple binder declarations to coexist
  5138 in the same context.  INCOMPATIBILITY.
  5139 
  5140 * Isar/locales: 'notation' provides a robust interface to the 'syntax'
  5141 primitive that also works in a locale context (both for constants and
  5142 fixed variables). Type declaration and internal syntactic representation
  5143 of given constants retrieved from the context. Likewise, the
  5144 'no_notation' command allows to remove given syntax annotations from the
  5145 current context.
  5146 
  5147 * Isar/locales: new derived specification elements 'axiomatization',
  5148 'definition', 'abbreviation', which support type-inference, admit
  5149 object-level specifications (equality, equivalence).  See also the
  5150 isar-ref manual.  Examples:
  5151 
  5152   axiomatization
  5153     eq  (infix "===" 50) where
  5154     eq_refl: "x === x" and eq_subst: "x === y ==> P x ==> P y"
  5155 
  5156   definition "f x y = x + y + 1"
  5157   definition g where "g x = f x x"
  5158 
  5159   abbreviation
  5160     neq  (infix "=!=" 50) where
  5161     "x =!= y == ~ (x === y)"
  5162 
  5163 These specifications may be also used in a locale context.  Then the
  5164 constants being introduced depend on certain fixed parameters, and the
  5165 constant name is qualified by the locale base name.  An internal
  5166 abbreviation takes care for convenient input and output, making the
  5167 parameters implicit and using the original short name.  See also
  5168 src/HOL/ex/Abstract_NAT.thy for an example of deriving polymorphic
  5169 entities from a monomorphic theory.
  5170 
  5171 Presently, abbreviations are only available 'in' a target locale, but
  5172 not inherited by general import expressions.  Also note that
  5173 'abbreviation' may be used as a type-safe replacement for 'syntax' +
  5174 'translations' in common applications.  The "no_abbrevs" print mode
  5175 prevents folding of abbreviations in term output.
  5176 
  5177 Concrete syntax is attached to specified constants in internal form,
  5178 independently of name spaces.  The parse tree representation is
  5179 slightly different -- use 'notation' instead of raw 'syntax', and
  5180 'translations' with explicit "CONST" markup to accommodate this.
  5181 
  5182 * Pure/Isar: unified syntax for new-style specification mechanisms
  5183 (e.g.  'definition', 'abbreviation', or 'inductive' in HOL) admits
  5184 full type inference and dummy patterns ("_").  For example:
  5185 
  5186   definition "K x _ = x"
  5187 
  5188   inductive conj for A B
  5189   where "A ==> B ==> conj A B"
  5190 
  5191 * Pure: command 'print_abbrevs' prints all constant abbreviations of
  5192 the current context.  Print mode "no_abbrevs" prevents inversion of
  5193 abbreviations on output.
  5194 
  5195 * Isar/locales: improved parameter handling: use of locales "var" and
  5196 "struct" no longer necessary; - parameter renamings are no longer
  5197 required to be injective.  For example, this allows to define
  5198 endomorphisms as locale endom = homom mult mult h.
  5199 
  5200 * Isar/locales: changed the way locales with predicates are defined.
  5201 Instead of accumulating the specification, the imported expression is
  5202 now an interpretation.  INCOMPATIBILITY: different normal form of
  5203 locale expressions.  In particular, in interpretations of locales with
  5204 predicates, goals repesenting already interpreted fragments are not
  5205 removed automatically.  Use methods `intro_locales' and
  5206 `unfold_locales'; see below.
  5207 
  5208 * Isar/locales: new methods `intro_locales' and `unfold_locales'
  5209 provide backward reasoning on locales predicates.  The methods are
  5210 aware of interpretations and discharge corresponding goals.
  5211 `intro_locales' is less aggressive then `unfold_locales' and does not
  5212 unfold predicates to assumptions.
  5213 
  5214 * Isar/locales: the order in which locale fragments are accumulated
  5215 has changed.  This enables to override declarations from fragments due
  5216 to interpretations -- for example, unwanted simp rules.
  5217 
  5218 * Isar/locales: interpretation in theories and proof contexts has been
  5219 extended.  One may now specify (and prove) equations, which are
  5220 unfolded in interpreted theorems.  This is useful for replacing
  5221 defined concepts (constants depending on locale parameters) by
  5222 concepts already existing in the target context.  Example:
  5223 
  5224   interpretation partial_order ["op <= :: [int, int] => bool"]
  5225     where "partial_order.less (op <=) (x::int) y = (x < y)"
  5226 
  5227 Typically, the constant `partial_order.less' is created by a
  5228 definition specification element in the context of locale
  5229 partial_order.
  5230 
  5231 * Method "induct": improved internal context management to support
  5232 local fixes and defines on-the-fly. Thus explicit meta-level
  5233 connectives !!  and ==> are rarely required anymore in inductive goals
  5234 (using object-logic connectives for this purpose has been long
  5235 obsolete anyway). Common proof patterns are explained in
  5236 src/HOL/Induct/Common_Patterns.thy, see also
  5237 src/HOL/Isar_examples/Puzzle.thy and src/HOL/Lambda for realistic
  5238 examples.
  5239 
  5240 * Method "induct": improved handling of simultaneous goals. Instead of
  5241 introducing object-level conjunction, the statement is now split into
  5242 several conclusions, while the corresponding symbolic cases are nested
  5243 accordingly. INCOMPATIBILITY, proofs need to be structured explicitly,
  5244 see src/HOL/Induct/Common_Patterns.thy, for example.
  5245 
  5246 * Method "induct": mutual induction rules are now specified as a list
  5247 of rule sharing the same induction cases. HOL packages usually provide
  5248 foo_bar.inducts for mutually defined items foo and bar (e.g. inductive
  5249 predicates/sets or datatypes). INCOMPATIBILITY, users need to specify
  5250 mutual induction rules differently, i.e. like this:
  5251 
  5252   (induct rule: foo_bar.inducts)
  5253   (induct set: foo bar)
  5254   (induct pred: foo bar)
  5255   (induct type: foo bar)
  5256 
  5257 The ML function ProjectRule.projections turns old-style rules into the
  5258 new format.
  5259 
  5260 * Method "coinduct": dual of induction, see
  5261 src/HOL/Library/Coinductive_List.thy for various examples.
  5262 
  5263 * Method "cases", "induct", "coinduct": the ``(open)'' option is
  5264 considered a legacy feature.
  5265 
  5266 * Attribute "symmetric" produces result with standardized schematic
  5267 variables (index 0).  Potential INCOMPATIBILITY.
  5268 
  5269 * Simplifier: by default the simplifier trace only shows top level
  5270 rewrites now. That is, trace_simp_depth_limit is set to 1 by
  5271 default. Thus there is less danger of being flooded by the trace. The
  5272 trace indicates where parts have been suppressed.
  5273   
  5274 * Provers/classical: removed obsolete classical version of elim_format
  5275 attribute; classical elim/dest rules are now treated uniformly when
  5276 manipulating the claset.
  5277 
  5278 * Provers/classical: stricter checks to ensure that supplied intro,
  5279 dest and elim rules are well-formed; dest and elim rules must have at
  5280 least one premise.
  5281 
  5282 * Provers/classical: attributes dest/elim/intro take an optional
  5283 weight argument for the rule (just as the Pure versions).  Weights are
  5284 ignored by automated tools, but determine the search order of single
  5285 rule steps.
  5286 
  5287 * Syntax: input syntax now supports dummy variable binding "%_. b",
  5288 where the body does not mention the bound variable.  Note that dummy
  5289 patterns implicitly depend on their context of bounds, which makes
  5290 "{_. _}" match any set comprehension as expected.  Potential
  5291 INCOMPATIBILITY -- parse translations need to cope with syntactic
  5292 constant "_idtdummy" in the binding position.
  5293 
  5294 * Syntax: removed obsolete syntactic constant "_K" and its associated
  5295 parse translation.  INCOMPATIBILITY -- use dummy abstraction instead,
  5296 for example "A -> B" => "Pi A (%_. B)".
  5297 
  5298 * Pure: 'class_deps' command visualizes the subclass relation, using
  5299 the graph browser tool.
  5300 
  5301 * Pure: 'print_theory' now suppresses certain internal declarations by
  5302 default; use '!' option for full details.
  5303 
  5304 
  5305 *** HOL ***
  5306 
  5307 * Method "metis" proves goals by applying the Metis general-purpose
  5308 resolution prover (see also http://gilith.com/software/metis/).
  5309 Examples are in the directory MetisExamples.  WARNING: the
  5310 Isabelle/HOL-Metis integration does not yet work properly with
  5311 multi-threading.
  5312   
  5313 * Command 'sledgehammer' invokes external automatic theorem provers as
  5314 background processes.  It generates calls to the "metis" method if
  5315 successful. These can be pasted into the proof.  Users do not have to
  5316 wait for the automatic provers to return.  WARNING: does not really
  5317 work with multi-threading.
  5318 
  5319 * New "auto_quickcheck" feature tests outermost goal statements for
  5320 potential counter-examples.  Controlled by ML references
  5321 auto_quickcheck (default true) and auto_quickcheck_time_limit (default
  5322 5000 milliseconds).  Fails silently if statements is outside of
  5323 executable fragment, or any other codgenerator problem occurs.
  5324 
  5325 * New constant "undefined" with axiom "undefined x = undefined".
  5326 
  5327 * Added class "HOL.eq", allowing for code generation with polymorphic
  5328 equality.
  5329 
  5330 * Some renaming of class constants due to canonical name prefixing in
  5331 the new 'class' package:
  5332 
  5333     HOL.abs ~> HOL.abs_class.abs
  5334     HOL.divide ~> HOL.divide_class.divide
  5335     0 ~> HOL.zero_class.zero
  5336     1 ~> HOL.one_class.one
  5337     op + ~> HOL.plus_class.plus
  5338     op - ~> HOL.minus_class.minus
  5339     uminus ~> HOL.minus_class.uminus
  5340     op * ~> HOL.times_class.times
  5341     op < ~> HOL.ord_class.less
  5342     op <= > HOL.ord_class.less_eq
  5343     Nat.power ~> Power.power_class.power
  5344     Nat.size ~> Nat.size_class.size
  5345     Numeral.number_of ~> Numeral.number_class.number_of
  5346     FixedPoint.Inf ~> Lattices.complete_lattice_class.Inf
  5347     FixedPoint.Sup ~> Lattices.complete_lattice_class.Sup
  5348     Orderings.min ~> Orderings.ord_class.min
  5349     Orderings.max ~> Orderings.ord_class.max
  5350     Divides.op div ~> Divides.div_class.div
  5351     Divides.op mod ~> Divides.div_class.mod
  5352     Divides.op dvd ~> Divides.div_class.dvd
  5353 
  5354 INCOMPATIBILITY.  Adaptions may be required in the following cases:
  5355 
  5356 a) User-defined constants using any of the names "plus", "minus",
  5357 "times", "less" or "less_eq". The standard syntax translations for
  5358 "+", "-" and "*" may go wrong.  INCOMPATIBILITY: use more specific
  5359 names.
  5360 
  5361 b) Variables named "plus", "minus", "times", "less", "less_eq"
  5362 INCOMPATIBILITY: use more specific names.
  5363 
  5364 c) Permutative equations (e.g. "a + b = b + a")
  5365 Since the change of names also changes the order of terms, permutative
  5366 rewrite rules may get applied in a different order. Experience shows
  5367 that this is rarely the case (only two adaptions in the whole Isabelle
  5368 distribution).  INCOMPATIBILITY: rewrite proofs
  5369 
  5370 d) ML code directly refering to constant names
  5371 This in general only affects hand-written proof tactics, simprocs and
  5372 so on.  INCOMPATIBILITY: grep your sourcecode and replace names.
  5373 Consider using @{const_name} antiquotation.
  5374 
  5375 * New class "default" with associated constant "default".
  5376 
  5377 * Function "sgn" is now overloaded and available on int, real, complex
  5378 (and other numeric types), using class "sgn".  Two possible defs of
  5379 sgn are given as equational assumptions in the classes sgn_if and
  5380 sgn_div_norm; ordered_idom now also inherits from sgn_if.
  5381 INCOMPATIBILITY.
  5382 
  5383 * Locale "partial_order" now unified with class "order" (cf. theory
  5384 Orderings), added parameter "less".  INCOMPATIBILITY.
  5385 
  5386 * Renamings in classes "order" and "linorder": facts "refl", "trans" and
  5387 "cases" to "order_refl", "order_trans" and "linorder_cases", to avoid
  5388 clashes with HOL "refl" and "trans".  INCOMPATIBILITY.
  5389 
  5390 * Classes "order" and "linorder": potential INCOMPATIBILITY due to
  5391 changed order of proof goals in instance proofs.
  5392 
  5393 * The transitivity reasoner for partial and linear orders is set up
  5394 for classes "order" and "linorder".  Instances of the reasoner are available
  5395 in all contexts importing or interpreting the corresponding locales.
  5396 Method "order" invokes the reasoner separately; the reasoner
  5397 is also integrated with the Simplifier as a solver.  Diagnostic
  5398 command 'print_orders' shows the available instances of the reasoner
  5399 in the current context.
  5400 
  5401 * Localized monotonicity predicate in theory "Orderings"; integrated
  5402 lemmas max_of_mono and min_of_mono with this predicate.
  5403 INCOMPATIBILITY.
  5404 
  5405 * Formulation of theorem "dense" changed slightly due to integration
  5406 with new class dense_linear_order.
  5407 
  5408 * Uniform lattice theory development in HOL.
  5409 
  5410     constants "meet" and "join" now named "inf" and "sup"
  5411     constant "Meet" now named "Inf"
  5412 
  5413     classes "meet_semilorder" and "join_semilorder" now named
  5414       "lower_semilattice" and "upper_semilattice"
  5415     class "lorder" now named "lattice"
  5416     class "comp_lat" now named "complete_lattice"
  5417 
  5418     Instantiation of lattice classes allows explicit definitions
  5419     for "inf" and "sup" operations (or "Inf" and "Sup" for complete lattices).
  5420 
  5421   INCOMPATIBILITY.  Theorem renames:
  5422 
  5423     meet_left_le            ~> inf_le1
  5424     meet_right_le           ~> inf_le2
  5425     join_left_le            ~> sup_ge1
  5426     join_right_le           ~> sup_ge2
  5427     meet_join_le            ~> inf_sup_ord
  5428     le_meetI                ~> le_infI
  5429     join_leI                ~> le_supI
  5430     le_meet                 ~> le_inf_iff
  5431     le_join                 ~> ge_sup_conv
  5432     meet_idempotent         ~> inf_idem
  5433     join_idempotent         ~> sup_idem
  5434     meet_comm               ~> inf_commute
  5435     join_comm               ~> sup_commute
  5436     meet_leI1               ~> le_infI1
  5437     meet_leI2               ~> le_infI2
  5438     le_joinI1               ~> le_supI1
  5439     le_joinI2               ~> le_supI2
  5440     meet_assoc              ~> inf_assoc
  5441     join_assoc              ~> sup_assoc
  5442     meet_left_comm          ~> inf_left_commute
  5443     meet_left_idempotent    ~> inf_left_idem
  5444     join_left_comm          ~> sup_left_commute
  5445     join_left_idempotent    ~> sup_left_idem
  5446     meet_aci                ~> inf_aci
  5447     join_aci                ~> sup_aci
  5448     le_def_meet             ~> le_iff_inf
  5449     le_def_join             ~> le_iff_sup
  5450     join_absorp2            ~> sup_absorb2
  5451     join_absorp1            ~> sup_absorb1
  5452     meet_absorp1            ~> inf_absorb1
  5453     meet_absorp2            ~> inf_absorb2
  5454     meet_join_absorp        ~> inf_sup_absorb
  5455     join_meet_absorp        ~> sup_inf_absorb
  5456     distrib_join_le         ~> distrib_sup_le
  5457     distrib_meet_le         ~> distrib_inf_le
  5458 
  5459     add_meet_distrib_left   ~> add_inf_distrib_left
  5460     add_join_distrib_left   ~> add_sup_distrib_left
  5461     is_join_neg_meet        ~> is_join_neg_inf
  5462     is_meet_neg_join        ~> is_meet_neg_sup
  5463     add_meet_distrib_right  ~> add_inf_distrib_right
  5464     add_join_distrib_right  ~> add_sup_distrib_right
  5465     add_meet_join_distribs  ~> add_sup_inf_distribs
  5466     join_eq_neg_meet        ~> sup_eq_neg_inf
  5467     meet_eq_neg_join        ~> inf_eq_neg_sup
  5468     add_eq_meet_join        ~> add_eq_inf_sup
  5469     meet_0_imp_0            ~> inf_0_imp_0
  5470     join_0_imp_0            ~> sup_0_imp_0
  5471     meet_0_eq_0             ~> inf_0_eq_0
  5472     join_0_eq_0             ~> sup_0_eq_0
  5473     neg_meet_eq_join        ~> neg_inf_eq_sup
  5474     neg_join_eq_meet        ~> neg_sup_eq_inf
  5475     join_eq_if              ~> sup_eq_if
  5476 
  5477     mono_meet               ~> mono_inf
  5478     mono_join               ~> mono_sup
  5479     meet_bool_eq            ~> inf_bool_eq
  5480     join_bool_eq            ~> sup_bool_eq
  5481     meet_fun_eq             ~> inf_fun_eq
  5482     join_fun_eq             ~> sup_fun_eq
  5483     meet_set_eq             ~> inf_set_eq
  5484     join_set_eq             ~> sup_set_eq
  5485     meet1_iff               ~> inf1_iff
  5486     meet2_iff               ~> inf2_iff
  5487     meet1I                  ~> inf1I
  5488     meet2I                  ~> inf2I
  5489     meet1D1                 ~> inf1D1
  5490     meet2D1                 ~> inf2D1
  5491     meet1D2                 ~> inf1D2
  5492     meet2D2                 ~> inf2D2
  5493     meet1E                  ~> inf1E
  5494     meet2E                  ~> inf2E
  5495     join1_iff               ~> sup1_iff
  5496     join2_iff               ~> sup2_iff
  5497     join1I1                 ~> sup1I1
  5498     join2I1                 ~> sup2I1
  5499     join1I1                 ~> sup1I1
  5500     join2I2                 ~> sup1I2
  5501     join1CI                 ~> sup1CI
  5502     join2CI                 ~> sup2CI
  5503     join1E                  ~> sup1E
  5504     join2E                  ~> sup2E
  5505 
  5506     is_meet_Meet            ~> is_meet_Inf
  5507     Meet_bool_def           ~> Inf_bool_def
  5508     Meet_fun_def            ~> Inf_fun_def
  5509     Meet_greatest           ~> Inf_greatest
  5510     Meet_lower              ~> Inf_lower
  5511     Meet_set_def            ~> Inf_set_def
  5512 
  5513     Sup_def                 ~> Sup_Inf
  5514     Sup_bool_eq             ~> Sup_bool_def
  5515     Sup_fun_eq              ~> Sup_fun_def
  5516     Sup_set_eq              ~> Sup_set_def
  5517 
  5518     listsp_meetI            ~> listsp_infI
  5519     listsp_meet_eq          ~> listsp_inf_eq
  5520 
  5521     meet_min                ~> inf_min
  5522     join_max                ~> sup_max
  5523 
  5524 * Added syntactic class "size"; overloaded constant "size" now has
  5525 type "'a::size ==> bool"
  5526 
  5527 * Internal reorganisation of `size' of datatypes: size theorems
  5528 "foo.size" are no longer subsumed by "foo.simps" (but are still
  5529 simplification rules by default!); theorems "prod.size" now named
  5530 "*.size".
  5531 
  5532 * Class "div" now inherits from class "times" rather than "type".
  5533 INCOMPATIBILITY.
  5534 
  5535 * HOL/Finite_Set: "name-space" locales Lattice, Distrib_lattice,
  5536 Linorder etc.  have disappeared; operations defined in terms of
  5537 fold_set now are named Inf_fin, Sup_fin.  INCOMPATIBILITY.
  5538 
  5539 * HOL/Nat: neq0_conv no longer declared as iff.  INCOMPATIBILITY.
  5540 
  5541 * HOL-Word: New extensive library and type for generic, fixed size
  5542 machine words, with arithmetic, bit-wise, shifting and rotating
  5543 operations, reflection into int, nat, and bool lists, automation for
  5544 linear arithmetic (by automatic reflection into nat or int), including
  5545 lemmas on overflow and monotonicity.  Instantiated to all appropriate
  5546 arithmetic type classes, supporting automatic simplification of
  5547 numerals on all operations.
  5548 
  5549 * Library/Boolean_Algebra: locales for abstract boolean algebras.
  5550 
  5551 * Library/Numeral_Type: numbers as types, e.g. TYPE(32).
  5552 
  5553 * Code generator library theories:
  5554   - Code_Integer represents HOL integers by big integer literals in target
  5555     languages.
  5556   - Code_Char represents HOL characters by character literals in target
  5557     languages.
  5558   - Code_Char_chr like Code_Char, but also offers treatment of character
  5559     codes; includes Code_Integer.
  5560   - Executable_Set allows to generate code for finite sets using lists.
  5561   - Executable_Rat implements rational numbers as triples (sign, enumerator,
  5562     denominator).
  5563   - Executable_Real implements a subset of real numbers, namly those
  5564     representable by rational numbers.
  5565   - Efficient_Nat implements natural numbers by integers, which in general will
  5566     result in higher efficency; pattern matching with 0/Suc is eliminated;
  5567     includes Code_Integer.
  5568   - Code_Index provides an additional datatype index which is mapped to
  5569     target-language built-in integers.
  5570   - Code_Message provides an additional datatype message_string which is isomorphic to
  5571     strings; messages are mapped to target-language strings.
  5572 
  5573 * New package for inductive predicates
  5574 
  5575   An n-ary predicate p with m parameters z_1, ..., z_m can now be defined via
  5576 
  5577     inductive
  5578       p :: "U_1 => ... => U_m => T_1 => ... => T_n => bool"
  5579       for z_1 :: U_1 and ... and z_n :: U_m
  5580     where
  5581       rule_1: "... ==> p z_1 ... z_m t_1_1 ... t_1_n"
  5582     | ...
  5583 
  5584   with full support for type-inference, rather than
  5585 
  5586     consts s :: "U_1 => ... => U_m => (T_1 * ... * T_n) set"
  5587 
  5588     abbreviation p :: "U_1 => ... => U_m => T_1 => ... => T_n => bool"
  5589     where "p z_1 ... z_m x_1 ... x_n == (x_1, ..., x_n) : s z_1 ... z_m"
  5590 
  5591     inductive "s z_1 ... z_m"
  5592     intros
  5593       rule_1: "... ==> (t_1_1, ..., t_1_n) : s z_1 ... z_m"
  5594       ...
  5595 
  5596   For backward compatibility, there is a wrapper allowing inductive
  5597   sets to be defined with the new package via
  5598 
  5599     inductive_set
  5600       s :: "U_1 => ... => U_m => (T_1 * ... * T_n) set"
  5601       for z_1 :: U_1 and ... and z_n :: U_m
  5602     where
  5603       rule_1: "... ==> (t_1_1, ..., t_1_n) : s z_1 ... z_m"
  5604     | ...
  5605 
  5606   or
  5607 
  5608     inductive_set
  5609       s :: "U_1 => ... => U_m => (T_1 * ... * T_n) set"
  5610       and p :: "U_1 => ... => U_m => T_1 => ... => T_n => bool"
  5611       for z_1 :: U_1 and ... and z_n :: U_m
  5612     where
  5613       "p z_1 ... z_m x_1 ... x_n == (x_1, ..., x_n) : s z_1 ... z_m"
  5614     | rule_1: "... ==> p z_1 ... z_m t_1_1 ... t_1_n"
  5615     | ...
  5616 
  5617   if the additional syntax "p ..." is required.
  5618 
  5619   Numerous examples can be found in the subdirectories src/HOL/Auth,
  5620   src/HOL/Bali, src/HOL/Induct, and src/HOL/MicroJava.
  5621 
  5622   INCOMPATIBILITIES:
  5623 
  5624   - Since declaration and definition of inductive sets or predicates
  5625     is no longer separated, abbreviations involving the newly
  5626     introduced sets or predicates must be specified together with the
  5627     introduction rules after the 'where' keyword (see above), rather
  5628     than before the actual inductive definition.
  5629 
  5630   - The variables in induction and elimination rules are now
  5631     quantified in the order of their occurrence in the introduction
  5632     rules, rather than in alphabetical order. Since this may break
  5633     some proofs, these proofs either have to be repaired, e.g. by
  5634     reordering the variables a_i_1 ... a_i_{k_i} in Isar 'case'
  5635     statements of the form
  5636 
  5637       case (rule_i a_i_1 ... a_i_{k_i})
  5638 
  5639     or the old order of quantification has to be restored by explicitly adding
  5640     meta-level quantifiers in the introduction rules, i.e.
  5641 
  5642       | rule_i: "!!a_i_1 ... a_i_{k_i}. ... ==> p z_1 ... z_m t_i_1 ... t_i_n"
  5643 
  5644   - The format of the elimination rules is now
  5645 
  5646       p z_1 ... z_m x_1 ... x_n ==>
  5647         (!!a_1_1 ... a_1_{k_1}. x_1 = t_1_1 ==> ... ==> x_n = t_1_n ==> ... ==> P)
  5648         ==> ... ==> P
  5649 
  5650     for predicates and
  5651 
  5652       (x_1, ..., x_n) : s z_1 ... z_m ==>
  5653         (!!a_1_1 ... a_1_{k_1}. x_1 = t_1_1 ==> ... ==> x_n = t_1_n ==> ... ==> P)
  5654         ==> ... ==> P
  5655 
  5656     for sets rather than
  5657 
  5658       x : s z_1 ... z_m ==>
  5659         (!!a_1_1 ... a_1_{k_1}. x = (t_1_1, ..., t_1_n) ==> ... ==> P)
  5660         ==> ... ==> P
  5661 
  5662     This may require terms in goals to be expanded to n-tuples
  5663     (e.g. using case_tac or simplification with the split_paired_all
  5664     rule) before the above elimination rule is applicable.
  5665 
  5666   - The elimination or case analysis rules for (mutually) inductive
  5667     sets or predicates are now called "p_1.cases" ... "p_k.cases". The
  5668     list of rules "p_1_..._p_k.elims" is no longer available.
  5669 
  5670 * New package "function"/"fun" for general recursive functions,
  5671 supporting mutual and nested recursion, definitions in local contexts,
  5672 more general pattern matching and partiality. See HOL/ex/Fundefs.thy
  5673 for small examples, and the separate tutorial on the function
  5674 package. The old recdef "package" is still available as before, but
  5675 users are encouraged to use the new package.
  5676 
  5677 * Method "lexicographic_order" automatically synthesizes termination
  5678 relations as lexicographic combinations of size measures. 
  5679 
  5680 * Case-expressions allow arbitrary constructor-patterns (including
  5681 "_") and take their order into account, like in functional
  5682 programming.  Internally, this is translated into nested
  5683 case-expressions; missing cases are added and mapped to the predefined
  5684 constant "undefined". In complicated cases printing may no longer show
  5685 the original input but the internal form. Lambda-abstractions allow
  5686 the same form of pattern matching: "% pat1 => e1 | ..." is an
  5687 abbreviation for "%x. case x of pat1 => e1 | ..." where x is a new
  5688 variable.
  5689 
  5690 * IntDef: The constant "int :: nat => int" has been removed; now "int"
  5691 is an abbreviation for "of_nat :: nat => int". The simplification
  5692 rules for "of_nat" have been changed to work like "int" did
  5693 previously.  Potential INCOMPATIBILITY:
  5694   - "of_nat (Suc m)" simplifies to "1 + of_nat m" instead of "of_nat m + 1"
  5695   - of_nat_diff and of_nat_mult are no longer default simp rules
  5696 
  5697 * Method "algebra" solves polynomial equations over (semi)rings using
  5698 Groebner bases. The (semi)ring structure is defined by locales and the
  5699 tool setup depends on that generic context. Installing the method for
  5700 a specific type involves instantiating the locale and possibly adding
  5701 declarations for computation on the coefficients.  The method is
  5702 already instantiated for natural numbers and for the axiomatic class
  5703 of idoms with numerals.  See also the paper by Chaieb and Wenzel at
  5704 CALCULEMUS 2007 for the general principles underlying this
  5705 architecture of context-aware proof-tools.
  5706 
  5707 * Method "ferrack" implements quantifier elimination over
  5708 special-purpose dense linear orders using locales (analogous to
  5709 "algebra"). The method is already installed for class
  5710 {ordered_field,recpower,number_ring} which subsumes real, hyperreal,
  5711 rat, etc.
  5712 
  5713 * Former constant "List.op @" now named "List.append".  Use ML
  5714 antiquotations @{const_name List.append} or @{term " ... @ ... "} to
  5715 circumvent possible incompatibilities when working on ML level.
  5716 
  5717 * primrec: missing cases mapped to "undefined" instead of "arbitrary".
  5718 
  5719 * New function listsum :: 'a list => 'a for arbitrary monoids.
  5720 Special syntax: "SUM x <- xs. f x" (and latex variants)
  5721 
  5722 * New syntax for Haskell-like list comprehension (input only), eg.
  5723 [(x,y). x <- xs, y <- ys, x ~= y], see also src/HOL/List.thy.
  5724 
  5725 * The special syntax for function "filter" has changed from [x :
  5726 xs. P] to [x <- xs. P] to avoid an ambiguity caused by list
  5727 comprehension syntax, and for uniformity.  INCOMPATIBILITY.
  5728 
  5729 * [a..b] is now defined for arbitrary linear orders.  It used to be
  5730 defined on nat only, as an abbreviation for [a..<Suc b]
  5731 INCOMPATIBILITY.
  5732 
  5733 * Renamed lemma "set_take_whileD"  to "set_takeWhileD".
  5734 
  5735 * New functions "sorted" and "sort" in src/HOL/List.thy.
  5736 
  5737 * New lemma collection field_simps (an extension of ring_simps) for
  5738 manipulating (in)equations involving division. Multiplies with all
  5739 denominators that can be proved to be non-zero (in equations) or
  5740 positive/negative (in inequations).
  5741 
  5742 * Lemma collections ring_eq_simps, group_eq_simps and ring_distrib
  5743 have been improved and renamed to ring_simps, group_simps and
  5744 ring_distribs.  Removed lemmas field_xyz in theory Ring_and_Field
  5745 because they were subsumed by lemmas xyz.  INCOMPATIBILITY.
  5746 
  5747 * Theory Library/Commutative_Ring: switched from recdef to function
  5748 package; constants add, mul, pow now curried.  Infix syntax for
  5749 algebraic operations.
  5750 
  5751 * Dropped redundant lemma def_imp_eq in favor of meta_eq_to_obj_eq.
  5752 INCOMPATIBILITY.
  5753 
  5754 * Dropped redundant lemma if_def2 in favor of if_bool_eq_conj.
  5755 INCOMPATIBILITY.
  5756 
  5757 * HOL/records: generalised field-update to take a function on the
  5758 field rather than the new value: r(|A := x|) is translated to A_update
  5759 (K x) r The K-combinator that is internally used is called K_record.
  5760 INCOMPATIBILITY: Usage of the plain update functions has to be
  5761 adapted.
  5762  
  5763 * Class "semiring_0" now contains annihilation axioms x * 0 = 0 and 0
  5764 * x = 0, which are required for a semiring.  Richer structures do not
  5765 inherit from semiring_0 anymore, because this property is a theorem
  5766 there, not an axiom.  INCOMPATIBILITY: In instances of semiring_0,
  5767 there is more to prove, but this is mostly trivial.
  5768 
  5769 * Class "recpower" is generalized to arbitrary monoids, not just
  5770 commutative semirings.  INCOMPATIBILITY: may need to incorporate
  5771 commutativity or semiring properties additionally.
  5772 
  5773 * Constant "List.list_all2" in List.thy now uses authentic syntax.
  5774 INCOMPATIBILITY: translations containing list_all2 may go wrong,
  5775 better use 'abbreviation'.
  5776 
  5777 * Renamed constant "List.op mem" to "List.member".  INCOMPATIBILITY.
  5778 
  5779 * Numeral syntax: type 'bin' which was a mere type copy of 'int' has
  5780 been abandoned in favour of plain 'int'.  INCOMPATIBILITY --
  5781 significant changes for setting up numeral syntax for types:
  5782   - New constants Numeral.pred and Numeral.succ instead
  5783       of former Numeral.bin_pred and Numeral.bin_succ.
  5784   - Use integer operations instead of bin_add, bin_mult and so on.
  5785   - Numeral simplification theorems named Numeral.numeral_simps instead of Bin_simps.
  5786   - ML structure Bin_Simprocs now named Int_Numeral_Base_Simprocs.
  5787 
  5788 See src/HOL/Integ/IntArith.thy for an example setup.
  5789 
  5790 * Command 'normal_form' computes the normal form of a term that may
  5791 contain free variables.  For example ``normal_form "rev [a, b, c]"''
  5792 produces ``[b, c, a]'' (without proof).  This command is suitable for
  5793 heavy-duty computations because the functions are compiled to ML
  5794 first.  Correspondingly, a method "normalization" is provided.  See
  5795 further src/HOL/ex/NormalForm.thy and src/Tools/nbe.ML.
  5796 
  5797 * Alternative iff syntax "A <-> B" for equality on bool (with priority
  5798 25 like -->); output depends on the "iff" print_mode, the default is
  5799 "A = B" (with priority 50).
  5800 
  5801 * Relations less (<) and less_eq (<=) are also available on type bool.
  5802 Modified syntax to disallow nesting without explicit parentheses,
  5803 e.g. "(x < y) < z" or "x < (y < z)", but NOT "x < y < z".  Potential
  5804 INCOMPATIBILITY.
  5805 
  5806 * "LEAST x:A. P" expands to "LEAST x. x:A & P" (input only).
  5807 
  5808 * Relation composition operator "op O" now has precedence 75 and binds
  5809 stronger than union and intersection. INCOMPATIBILITY.
  5810 
  5811 * The old set interval syntax "{m..n(}" (and relatives) has been
  5812 removed.  Use "{m..<n}" (and relatives) instead.
  5813 
  5814 * In the context of the assumption "~(s = t)" the Simplifier rewrites
  5815 "t = s" to False (by simproc "neq").  INCOMPATIBILITY, consider using
  5816 ``declare [[simproc del: neq]]''.
  5817 
  5818 * Simplifier: "m dvd n" where m and n are numbers is evaluated to
  5819 True/False.
  5820 
  5821 * Theorem Cons_eq_map_conv no longer declared as "simp".
  5822 
  5823 * Theorem setsum_mult renamed to setsum_right_distrib.
  5824 
  5825 * Prefer ex1I over ex_ex1I in single-step reasoning, e.g. by the
  5826 ``rule'' method.
  5827 
  5828 * Reimplemented methods "sat" and "satx", with several improvements:
  5829 goals no longer need to be stated as "<prems> ==> False", equivalences
  5830 (i.e. "=" on type bool) are handled, variable names of the form
  5831 "lit_<n>" are no longer reserved, significant speedup.
  5832 
  5833 * Methods "sat" and "satx" can now replay MiniSat proof traces.
  5834 zChaff is still supported as well.
  5835 
  5836 * 'inductive' and 'datatype': provide projections of mutual rules,
  5837 bundled as foo_bar.inducts;
  5838 
  5839 * Library: moved theories Parity, GCD, Binomial, Infinite_Set to
  5840 Library.
  5841 
  5842 * Library: moved theory Accessible_Part to main HOL.
  5843 
  5844 * Library: added theory Coinductive_List of potentially infinite lists
  5845 as greatest fixed-point.
  5846 
  5847 * Library: added theory AssocList which implements (finite) maps as
  5848 association lists.
  5849 
  5850 * Method "evaluation" solves goals (i.e. a boolean expression)
  5851 efficiently by compiling it to ML.  The goal is "proved" (via an
  5852 oracle) if it evaluates to True.
  5853 
  5854 * Linear arithmetic now splits certain operators (e.g. min, max, abs)
  5855 also when invoked by the simplifier.  This results in the Simplifier
  5856 being more powerful on arithmetic goals.  INCOMPATIBILITY.
  5857 Configuration option fast_arith_split_limit=0 recovers the old
  5858 behavior.
  5859 
  5860 * Support for hex (0x20) and binary (0b1001) numerals.
  5861 
  5862 * New method: reify eqs (t), where eqs are equations for an
  5863 interpretation I :: 'a list => 'b => 'c and t::'c is an optional
  5864 parameter, computes a term s::'b and a list xs::'a list and proves the
  5865 theorem I xs s = t. This is also known as reification or quoting. The
  5866 resulting theorem is applied to the subgoal to substitute t with I xs
  5867 s.  If t is omitted, the subgoal itself is reified.
  5868 
  5869 * New method: reflection corr_thm eqs (t). The parameters eqs and (t)
  5870 are as explained above. corr_thm is a theorem for I vs (f t) = I vs t,
  5871 where f is supposed to be a computable function (in the sense of code
  5872 generattion). The method uses reify to compute s and xs as above then
  5873 applies corr_thm and uses normalization by evaluation to "prove" f s =
  5874 r and finally gets the theorem t = r, which is again applied to the
  5875 subgoal. An Example is available in src/HOL/ex/ReflectionEx.thy.
  5876 
  5877 * Reflection: Automatic reification now handels binding, an example is
  5878 available in src/HOL/ex/ReflectionEx.thy
  5879 
  5880 * HOL-Statespace: ``State Spaces: The Locale Way'' introduces a
  5881 command 'statespace' that is similar to 'record', but introduces an
  5882 abstract specification based on the locale infrastructure instead of
  5883 HOL types.  This leads to extra flexibility in composing state spaces,
  5884 in particular multiple inheritance and renaming of components.
  5885 
  5886 
  5887 *** HOL-Complex ***
  5888 
  5889 * Hyperreal: Functions root and sqrt are now defined on negative real
  5890 inputs so that root n (- x) = - root n x and sqrt (- x) = - sqrt x.
  5891 Nonnegativity side conditions have been removed from many lemmas, so
  5892 that more subgoals may now be solved by simplification; potential
  5893 INCOMPATIBILITY.
  5894 
  5895 * Real: new type classes formalize real normed vector spaces and
  5896 algebras, using new overloaded constants scaleR :: real => 'a => 'a
  5897 and norm :: 'a => real.
  5898 
  5899 * Real: constant of_real :: real => 'a::real_algebra_1 injects from
  5900 reals into other types. The overloaded constant Reals :: 'a set is now
  5901 defined as range of_real; potential INCOMPATIBILITY.
  5902 
  5903 * Real: proper support for ML code generation, including 'quickcheck'.
  5904 Reals are implemented as arbitrary precision rationals.
  5905 
  5906 * Hyperreal: Several constants that previously worked only for the
  5907 reals have been generalized, so they now work over arbitrary vector
  5908 spaces. Type annotations may need to be added in some cases; potential
  5909 INCOMPATIBILITY.
  5910 
  5911   Infinitesimal  :: ('a::real_normed_vector) star set
  5912   HFinite        :: ('a::real_normed_vector) star set
  5913   HInfinite      :: ('a::real_normed_vector) star set
  5914   approx         :: ('a::real_normed_vector) star => 'a star => bool
  5915   monad          :: ('a::real_normed_vector) star => 'a star set
  5916   galaxy         :: ('a::real_normed_vector) star => 'a star set
  5917   (NS)LIMSEQ     :: [nat => 'a::real_normed_vector, 'a] => bool
  5918   (NS)convergent :: (nat => 'a::real_normed_vector) => bool
  5919   (NS)Bseq       :: (nat => 'a::real_normed_vector) => bool
  5920   (NS)Cauchy     :: (nat => 'a::real_normed_vector) => bool
  5921   (NS)LIM        :: ['a::real_normed_vector => 'b::real_normed_vector, 'a, 'b] => bool
  5922   is(NS)Cont     :: ['a::real_normed_vector => 'b::real_normed_vector, 'a] => bool
  5923   deriv          :: ['a::real_normed_field => 'a, 'a, 'a] => bool
  5924   sgn            :: 'a::real_normed_vector => 'a
  5925   exp            :: 'a::{recpower,real_normed_field,banach} => 'a
  5926 
  5927 * Complex: Some complex-specific constants are now abbreviations for
  5928 overloaded ones: complex_of_real = of_real, cmod = norm, hcmod =
  5929 hnorm.  Other constants have been entirely removed in favor of the
  5930 polymorphic versions (INCOMPATIBILITY):
  5931 
  5932   approx        <-- capprox
  5933   HFinite       <-- CFinite
  5934   HInfinite     <-- CInfinite
  5935   Infinitesimal <-- CInfinitesimal
  5936   monad         <-- cmonad
  5937   galaxy        <-- cgalaxy
  5938   (NS)LIM       <-- (NS)CLIM, (NS)CRLIM
  5939   is(NS)Cont    <-- is(NS)Contc, is(NS)contCR
  5940   (ns)deriv     <-- (ns)cderiv
  5941 
  5942 
  5943 *** HOL-Algebra ***
  5944 
  5945 * Formalisation of ideals and the quotient construction over rings.
  5946 
  5947 * Order and lattice theory no longer based on records.
  5948 INCOMPATIBILITY.
  5949 
  5950 * Renamed lemmas least_carrier -> least_closed and greatest_carrier ->
  5951 greatest_closed.  INCOMPATIBILITY.
  5952 
  5953 * Method algebra is now set up via an attribute.  For examples see
  5954 Ring.thy.  INCOMPATIBILITY: the method is now weaker on combinations
  5955 of algebraic structures.
  5956 
  5957 * Renamed theory CRing to Ring.
  5958 
  5959 
  5960 *** HOL-Nominal ***
  5961 
  5962 * Substantial, yet incomplete support for nominal datatypes (binding
  5963 structures) based on HOL-Nominal logic.  See src/HOL/Nominal and
  5964 src/HOL/Nominal/Examples.  Prospective users should consult
  5965 http://isabelle.in.tum.de/nominal/
  5966 
  5967 
  5968 *** ML ***
  5969 
  5970 * ML basics: just one true type int, which coincides with IntInf.int
  5971 (even on SML/NJ).
  5972 
  5973 * ML within Isar: antiquotations allow to embed statically-checked
  5974 formal entities in the source, referring to the context available at
  5975 compile-time.  For example:
  5976 
  5977 ML {* @{sort "{zero,one}"} *}
  5978 ML {* @{typ "'a => 'b"} *}
  5979 ML {* @{term "%x. x"} *}
  5980 ML {* @{prop "x == y"} *}
  5981 ML {* @{ctyp "'a => 'b"} *}
  5982 ML {* @{cterm "%x. x"} *}
  5983 ML {* @{cprop "x == y"} *}
  5984 ML {* @{thm asm_rl} *}
  5985 ML {* @{thms asm_rl} *}
  5986 ML {* @{type_name c} *}
  5987 ML {* @{type_syntax c} *}
  5988 ML {* @{const_name c} *}
  5989 ML {* @{const_syntax c} *}
  5990 ML {* @{context} *}
  5991 ML {* @{theory} *}
  5992 ML {* @{theory Pure} *}
  5993 ML {* @{theory_ref} *}
  5994 ML {* @{theory_ref Pure} *}
  5995 ML {* @{simpset} *}
  5996 ML {* @{claset} *}
  5997 ML {* @{clasimpset} *}
  5998 
  5999 The same works for sources being ``used'' within an Isar context.
  6000 
  6001 * ML in Isar: improved error reporting; extra verbosity with
  6002 ML_Context.trace enabled.
  6003 
  6004 * Pure/General/table.ML: the join operations now works via exceptions
  6005 DUP/SAME instead of type option. This is simpler in simple cases, and
  6006 admits slightly more efficient complex applications.
  6007 
  6008 * Pure: 'advanced' translation functions (parse_translation etc.) now
  6009 use Context.generic instead of just theory.
  6010 
  6011 * Pure: datatype Context.generic joins theory/Proof.context and
  6012 provides some facilities for code that works in either kind of
  6013 context, notably GenericDataFun for uniform theory and proof data.
  6014 
  6015 * Pure: simplified internal attribute type, which is now always
  6016 Context.generic * thm -> Context.generic * thm. Global (theory) vs.
  6017 local (Proof.context) attributes have been discontinued, while
  6018 minimizing code duplication. Thm.rule_attribute and
  6019 Thm.declaration_attribute build canonical attributes; see also structure
  6020 Context for further operations on Context.generic, notably
  6021 GenericDataFun. INCOMPATIBILITY, need to adapt attribute type
  6022 declarations and definitions.
  6023 
  6024 * Context data interfaces (Theory/Proof/GenericDataFun): removed
  6025 name/print, uninitialized data defaults to ad-hoc copy of empty value,
  6026 init only required for impure data. INCOMPATIBILITY: empty really need
  6027 to be empty (no dependencies on theory content!)
  6028 
  6029 * Pure/kernel: consts certification ignores sort constraints given in
  6030 signature declarations. (This information is not relevant to the
  6031 logic, but only for type inference.) SIGNIFICANT INTERNAL CHANGE,
  6032 potential INCOMPATIBILITY.
  6033 
  6034 * Pure: axiomatic type classes are now purely definitional, with
  6035 explicit proofs of class axioms and super class relations performed
  6036 internally. See Pure/axclass.ML for the main internal interfaces --
  6037 notably AxClass.define_class supercedes AxClass.add_axclass, and
  6038 AxClass.axiomatize_class/classrel/arity supersede
  6039 Sign.add_classes/classrel/arities.
  6040 
  6041 * Pure/Isar: Args/Attrib parsers operate on Context.generic --
  6042 global/local versions on theory vs. Proof.context have been
  6043 discontinued; Attrib.syntax and Method.syntax have been adapted
  6044 accordingly.  INCOMPATIBILITY, need to adapt parser expressions for
  6045 attributes, methods, etc.
  6046 
  6047 * Pure: several functions of signature "... -> theory -> theory * ..."
  6048 have been reoriented to "... -> theory -> ... * theory" in order to
  6049 allow natural usage in combination with the ||>, ||>>, |-> and
  6050 fold_map combinators.
  6051 
  6052 * Pure: official theorem names (closed derivations) and additional
  6053 comments (tags) are now strictly separate.  Name hints -- which are
  6054 maintained as tags -- may be attached any time without affecting the
  6055 derivation.
  6056 
  6057 * Pure: primitive rule lift_rule now takes goal cterm instead of an
  6058 actual goal state (thm).  Use Thm.lift_rule (Thm.cprem_of st i) to
  6059 achieve the old behaviour.
  6060 
  6061 * Pure: the "Goal" constant is now called "prop", supporting a
  6062 slightly more general idea of ``protecting'' meta-level rule
  6063 statements.
  6064 
  6065 * Pure: Logic.(un)varify only works in a global context, which is now
  6066 enforced instead of silently assumed.  INCOMPATIBILITY, may use
  6067 Logic.legacy_(un)varify as temporary workaround.
  6068 
  6069 * Pure: structure Name provides scalable operations for generating
  6070 internal variable names, notably Name.variants etc.  This replaces
  6071 some popular functions from term.ML:
  6072 
  6073   Term.variant		->  Name.variant
  6074   Term.variantlist	->  Name.variant_list
  6075   Term.invent_names	->  Name.invent_list
  6076 
  6077 Note that low-level renaming rarely occurs in new code -- operations
  6078 from structure Variable are used instead (see below).
  6079 
  6080 * Pure: structure Variable provides fundamental operations for proper
  6081 treatment of fixed/schematic variables in a context.  For example,
  6082 Variable.import introduces fixes for schematics of given facts and
  6083 Variable.export reverses the effect (up to renaming) -- this replaces
  6084 various freeze_thaw operations.
  6085 
  6086 * Pure: structure Goal provides simple interfaces for
  6087 init/conclude/finish and tactical prove operations (replacing former
  6088 Tactic.prove).  Goal.prove is the canonical way to prove results
  6089 within a given context; Goal.prove_global is a degraded version for
  6090 theory level goals, including a global Drule.standard.  Note that
  6091 OldGoals.prove_goalw_cterm has long been obsolete, since it is
  6092 ill-behaved in a local proof context (e.g. with local fixes/assumes or
  6093 in a locale context).
  6094 
  6095 * Pure/Syntax: generic interfaces for parsing (Syntax.parse_term etc.)
  6096 and type checking (Syntax.check_term etc.), with common combinations
  6097 (Syntax.read_term etc.). These supersede former Sign.read_term etc.
  6098 which are considered legacy and await removal.
  6099 
  6100 * Pure/Syntax: generic interfaces for type unchecking
  6101 (Syntax.uncheck_terms etc.) and unparsing (Syntax.unparse_term etc.),
  6102 with common combinations (Syntax.pretty_term, Syntax.string_of_term
  6103 etc.).  Former Sign.pretty_term, Sign.string_of_term etc. are still
  6104 available for convenience, but refer to the very same operations using
  6105 a mere theory instead of a full context.
  6106 
  6107 * Isar: simplified treatment of user-level errors, using exception
  6108 ERROR of string uniformly.  Function error now merely raises ERROR,
  6109 without any side effect on output channels.  The Isar toplevel takes
  6110 care of proper display of ERROR exceptions.  ML code may use plain
  6111 handle/can/try; cat_error may be used to concatenate errors like this:
  6112 
  6113   ... handle ERROR msg => cat_error msg "..."
  6114 
  6115 Toplevel ML code (run directly or through the Isar toplevel) may be
  6116 embedded into the Isar toplevel with exception display/debug like
  6117 this:
  6118 
  6119   Isar.toplevel (fn () => ...)
  6120 
  6121 INCOMPATIBILITY, removed special transform_error facilities, removed
  6122 obsolete variants of user-level exceptions (ERROR_MESSAGE,
  6123 Context.PROOF, ProofContext.CONTEXT, Proof.STATE, ProofHistory.FAIL)
  6124 -- use plain ERROR instead.
  6125 
  6126 * Isar: theory setup now has type (theory -> theory), instead of a
  6127 list.  INCOMPATIBILITY, may use #> to compose setup functions.
  6128 
  6129 * Isar: ML toplevel pretty printer for type Proof.context, subject to
  6130 ProofContext.debug/verbose flags.
  6131 
  6132 * Isar: Toplevel.theory_to_proof admits transactions that modify the
  6133 theory before entering a proof state.  Transactions now always see a
  6134 quasi-functional intermediate checkpoint, both in interactive and
  6135 batch mode.
  6136 
  6137 * Isar: simplified interfaces for outer syntax.  Renamed
  6138 OuterSyntax.add_keywords to OuterSyntax.keywords.  Removed
  6139 OuterSyntax.add_parsers -- this functionality is now included in
  6140 OuterSyntax.command etc.  INCOMPATIBILITY.
  6141 
  6142 * Simplifier: the simpset of a running simplification process now
  6143 contains a proof context (cf. Simplifier.the_context), which is the
  6144 very context that the initial simpset has been retrieved from (by
  6145 simpset_of/local_simpset_of).  Consequently, all plug-in components
  6146 (solver, looper etc.) may depend on arbitrary proof data.
  6147 
  6148 * Simplifier.inherit_context inherits the proof context (plus the
  6149 local bounds) of the current simplification process; any simproc
  6150 etc. that calls the Simplifier recursively should do this!  Removed
  6151 former Simplifier.inherit_bounds, which is already included here --
  6152 INCOMPATIBILITY.  Tools based on low-level rewriting may even have to
  6153 specify an explicit context using Simplifier.context/theory_context.
  6154 
  6155 * Simplifier/Classical Reasoner: more abstract interfaces
  6156 change_simpset/claset for modifying the simpset/claset reference of a
  6157 theory; raw versions simpset/claset_ref etc. have been discontinued --
  6158 INCOMPATIBILITY.
  6159 
  6160 * Provers: more generic wrt. syntax of object-logics, avoid hardwired
  6161 "Trueprop" etc.
  6162 
  6163 
  6164 *** System ***
  6165 
  6166 * settings: the default heap location within ISABELLE_HOME_USER now
  6167 includes ISABELLE_IDENTIFIER.  This simplifies use of multiple
  6168 Isabelle installations.
  6169 
  6170 * isabelle-process: option -S (secure mode) disables some critical
  6171 operations, notably runtime compilation and evaluation of ML source
  6172 code.
  6173 
  6174 * Basic Isabelle mode for jEdit, see Isabelle/lib/jedit/.
  6175 
  6176 * Support for parallel execution, using native multicore support of
  6177 Poly/ML 5.1.  The theory loader exploits parallelism when processing
  6178 independent theories, according to the given theory header
  6179 specifications. The maximum number of worker threads is specified via
  6180 usedir option -M or the "max-threads" setting in Proof General. A
  6181 speedup factor of 1.5--3.5 can be expected on a 4-core machine, and up
  6182 to 6 on a 8-core machine.  User-code needs to observe certain
  6183 guidelines for thread-safe programming, see appendix A in the Isar
  6184 Implementation manual.
  6185 
  6186 
  6187 
  6188 New in Isabelle2005 (October 2005)
  6189 ----------------------------------
  6190 
  6191 *** General ***
  6192 
  6193 * Theory headers: the new header syntax for Isar theories is
  6194 
  6195   theory <name>
  6196   imports <theory1> ... <theoryN>
  6197   uses <file1> ... <fileM>
  6198   begin
  6199 
  6200 where the 'uses' part is optional.  The previous syntax
  6201 
  6202   theory <name> = <theory1> + ... + <theoryN>:
  6203 
  6204 will disappear in the next release.  Use isatool fixheaders to convert
  6205 existing theory files.  Note that there is no change in ancient
  6206 non-Isar theories now, but these will disappear soon.
  6207 
  6208 * Theory loader: parent theories can now also be referred to via
  6209 relative and absolute paths.
  6210 
  6211 * Command 'find_theorems' searches for a list of criteria instead of a
  6212 list of constants. Known criteria are: intro, elim, dest, name:string,
  6213 simp:term, and any term. Criteria can be preceded by '-' to select
  6214 theorems that do not match. Intro, elim, dest select theorems that
  6215 match the current goal, name:s selects theorems whose fully qualified
  6216 name contain s, and simp:term selects all simplification rules whose
  6217 lhs match term.  Any other term is interpreted as pattern and selects
  6218 all theorems matching the pattern. Available in ProofGeneral under
  6219 'ProofGeneral -> Find Theorems' or C-c C-f.  Example:
  6220 
  6221   C-c C-f (100) "(_::nat) + _ + _" intro -name: "HOL."
  6222 
  6223 prints the last 100 theorems matching the pattern "(_::nat) + _ + _",
  6224 matching the current goal as introduction rule and not having "HOL."
  6225 in their name (i.e. not being defined in theory HOL).
  6226 
  6227 * Command 'thms_containing' has been discontinued in favour of
  6228 'find_theorems'; INCOMPATIBILITY.
  6229 
  6230 * Communication with Proof General is now 8bit clean, which means that
  6231 Unicode text in UTF-8 encoding may be used within theory texts (both
  6232 formal and informal parts).  Cf. option -U of the Isabelle Proof
  6233 General interface.  Here are some simple examples (cf. src/HOL/ex):
  6234 
  6235   http://isabelle.in.tum.de/library/HOL/ex/Hebrew.html
  6236   http://isabelle.in.tum.de/library/HOL/ex/Chinese.html
  6237 
  6238 * Improved efficiency of the Simplifier and, to a lesser degree, the
  6239 Classical Reasoner.  Typical big applications run around 2 times
  6240 faster.