NEWS
author wenzelm
Sat Jul 28 13:11:58 2012 +0200 (2012-07-28)
changeset 48574 4af9f3122138
parent 48431 6efff142bb54
child 48585 a82910dd2270
permissions -rw-r--r--
discontinued special treatment of Proof General;
     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 * Code generation for Haskell: restrict unqualified imports from
    19 Haskell Prelude to a small set of fundamental operations.
    20 
    21 * Command "export_code": relative file names are interpreted
    22 relatively to master directory of current theory rather than
    23 the rather arbitrary current working directory.
    24 INCOMPATIBILITY.
    25 
    26 * Discontinued obsolete attribute "COMP".  Potential INCOMPATIBILITY,
    27 use regular rule composition via "OF" / "THEN", or explicit proof
    28 structure instead.  Note that Isabelle/ML provides a variety of
    29 operators like COMP, INCR_COMP, COMP_INCR, which need to be applied
    30 with some care where this is really required.
    31 
    32 
    33 *** HOL ***
    34 
    35 * Library/Debug.thy and Library/Parallel.thy: debugging and parallel
    36 execution for code generated towards Isabelle/ML.
    37 
    38 * Simproc "finite_Collect" rewrites set comprehensions into pointfree
    39 expressions.
    40 
    41 * Quickcheck:
    42 
    43   - added an optimisation for equality premises.
    44     It is switched on by default, and can be switched off by setting
    45     the configuration quickcheck_optimise_equality to false.    
    46 
    47 * The SMT solver Z3 has now by default a restricted set of directly
    48 supported features. For the full set of features (div/mod, nonlinear
    49 arithmetic, datatypes/records) with potential proof reconstruction
    50 failures, enable the configuration option "z3_with_extensions".
    51 Minor INCOMPATIBILITY.
    52 
    53 * Sledgehammer:
    54 
    55   - Added MaSh relevance filter based on machine-learning; see the
    56     Sledgehammer manual for details.
    57   - Rationalized type encodings ("type_enc" option).
    58   - Renamed options:
    59       max_relevant ~> max_facts
    60       relevance_thresholds ~> fact_thresholds
    61 
    62 
    63 *** Document preparation ***
    64 
    65 * Default for \<euro> is now based on eurosym package, instead of
    66 slightly exotic babel/greek.
    67 
    68 
    69 *** System ***
    70 
    71 * Discontinued support for Poly/ML 5.2.1, which was the last version
    72 without exception positions and advanced ML compiler/toplevel
    73 configuration.
    74 
    75 * Discontinued special treatment of Proof General -- no longer guess
    76 PROOFGENERAL_HOME based on accidental file-system layout.  Minor
    77 INCOMPATIBILITY: provide PROOFGENERAL_HOME and PROOFGENERAL_OPTIONS
    78 settings manually, or use a Proof General version that has been
    79 bundled as Isabelle component.
    80 
    81 
    82 New in Isabelle2012 (May 2012)
    83 ------------------------------
    84 
    85 *** General ***
    86 
    87 * Prover IDE (PIDE) improvements:
    88 
    89   - more robust Sledgehammer integration (as before the sledgehammer
    90     command-line needs to be typed into the source buffer)
    91   - markup for bound variables
    92   - markup for types of term variables (displayed as tooltips)
    93   - support for user-defined Isar commands within the running session
    94   - improved support for Unicode outside original 16bit range
    95     e.g. glyph for \<A> (thanks to jEdit 4.5.1)
    96 
    97 * Forward declaration of outer syntax keywords within the theory
    98 header -- minor INCOMPATIBILITY for user-defined commands.  Allow new
    99 commands to be used in the same theory where defined.
   100 
   101 * Auxiliary contexts indicate block structure for specifications with
   102 additional parameters and assumptions.  Such unnamed contexts may be
   103 nested within other targets, like 'theory', 'locale', 'class',
   104 'instantiation' etc.  Results from the local context are generalized
   105 accordingly and applied to the enclosing target context.  Example:
   106 
   107   context
   108     fixes x y z :: 'a
   109     assumes xy: "x = y" and yz: "y = z"
   110   begin
   111 
   112   lemma my_trans: "x = z" using xy yz by simp
   113 
   114   end
   115 
   116   thm my_trans
   117 
   118 The most basic application is to factor-out context elements of
   119 several fixes/assumes/shows theorem statements, e.g. see
   120 ~~/src/HOL/Isar_Examples/Group_Context.thy
   121 
   122 Any other local theory specification element works within the "context
   123 ... begin ... end" block as well.
   124 
   125 * Bundled declarations associate attributed fact expressions with a
   126 given name in the context.  These may be later included in other
   127 contexts.  This allows to manage context extensions casually, without
   128 the logical dependencies of locales and locale interpretation.  See
   129 commands 'bundle', 'include', 'including' etc. in the isar-ref manual.
   130 
   131 * Commands 'lemmas' and 'theorems' allow local variables using 'for'
   132 declaration, and results are standardized before being stored.  Thus
   133 old-style "standard" after instantiation or composition of facts
   134 becomes obsolete.  Minor INCOMPATIBILITY, due to potential change of
   135 indices of schematic variables.
   136 
   137 * Rule attributes in local theory declarations (e.g. locale or class)
   138 are now statically evaluated: the resulting theorem is stored instead
   139 of the original expression.  INCOMPATIBILITY in rare situations, where
   140 the historic accident of dynamic re-evaluation in interpretations
   141 etc. was exploited.
   142 
   143 * New tutorial "Programming and Proving in Isabelle/HOL"
   144 ("prog-prove").  It completely supersedes "A Tutorial Introduction to
   145 Structured Isar Proofs" ("isar-overview"), which has been removed.  It
   146 also supersedes "Isabelle/HOL, A Proof Assistant for Higher-Order
   147 Logic" as the recommended beginners tutorial, but does not cover all
   148 of the material of that old tutorial.
   149 
   150 * Updated and extended reference manuals: "isar-ref",
   151 "implementation", "system"; reduced remaining material in old "ref"
   152 manual.
   153 
   154 
   155 *** Pure ***
   156 
   157 * Command 'definition' no longer exports the foundational "raw_def"
   158 into the user context.  Minor INCOMPATIBILITY, may use the regular
   159 "def" result with attribute "abs_def" to imitate the old version.
   160 
   161 * Attribute "abs_def" turns an equation of the form "f x y == t" into
   162 "f == %x y. t", which ensures that "simp" or "unfold" steps always
   163 expand it.  This also works for object-logic equality.  (Formerly
   164 undocumented feature.)
   165 
   166 * Sort constraints are now propagated in simultaneous statements, just
   167 like type constraints.  INCOMPATIBILITY in rare situations, where
   168 distinct sorts used to be assigned accidentally.  For example:
   169 
   170   lemma "P (x::'a::foo)" and "Q (y::'a::bar)"  -- "now illegal"
   171 
   172   lemma "P (x::'a)" and "Q (y::'a::bar)"
   173     -- "now uniform 'a::bar instead of default sort for first occurrence (!)"
   174 
   175 * Rule composition via attribute "OF" (or ML functions OF/MRS) is more
   176 tolerant against multiple unifiers, as long as the final result is
   177 unique.  (As before, rules are composed in canonical right-to-left
   178 order to accommodate newly introduced premises.)
   179 
   180 * Renamed some inner syntax categories:
   181 
   182     num ~> num_token
   183     xnum ~> xnum_token
   184     xstr ~> str_token
   185 
   186 Minor INCOMPATIBILITY.  Note that in practice "num_const" or
   187 "num_position" etc. are mainly used instead (which also include
   188 position information via constraints).
   189 
   190 * Simplified configuration options for syntax ambiguity: see
   191 "syntax_ambiguity_warning" and "syntax_ambiguity_limit" in isar-ref
   192 manual.  Minor INCOMPATIBILITY.
   193 
   194 * Discontinued configuration option "syntax_positions": atomic terms
   195 in parse trees are always annotated by position constraints.
   196 
   197 * Old code generator for SML and its commands 'code_module',
   198 'code_library', 'consts_code', 'types_code' have been discontinued.
   199 Use commands of the generic code generator instead.  INCOMPATIBILITY.
   200 
   201 * Redundant attribute "code_inline" has been discontinued. Use
   202 "code_unfold" instead.  INCOMPATIBILITY.
   203 
   204 * Dropped attribute "code_unfold_post" in favor of the its dual
   205 "code_abbrev", which yields a common pattern in definitions like
   206 
   207   definition [code_abbrev]: "f = t"
   208 
   209 INCOMPATIBILITY.
   210 
   211 * Obsolete 'types' command has been discontinued.  Use 'type_synonym'
   212 instead.  INCOMPATIBILITY.
   213 
   214 * Discontinued old "prems" fact, which used to refer to the accidental
   215 collection of foundational premises in the context (already marked as
   216 legacy since Isabelle2011).
   217 
   218 
   219 *** HOL ***
   220 
   221 * Type 'a set is now a proper type constructor (just as before
   222 Isabelle2008).  Definitions mem_def and Collect_def have disappeared.
   223 Non-trivial INCOMPATIBILITY.  For developments keeping predicates and
   224 sets separate, it is often sufficient to rephrase some set S that has
   225 been accidentally used as predicates by "%x. x : S", and some
   226 predicate P that has been accidentally used as set by "{x. P x}".
   227 Corresponding proofs in a first step should be pruned from any
   228 tinkering with former theorems mem_def and Collect_def as far as
   229 possible.
   230 
   231 For developments which deliberately mix predicates and sets, a
   232 planning step is necessary to determine what should become a predicate
   233 and what a set.  It can be helpful to carry out that step in
   234 Isabelle2011-1 before jumping right into the current release.
   235 
   236 * Code generation by default implements sets as container type rather
   237 than predicates.  INCOMPATIBILITY.
   238 
   239 * New type synonym 'a rel = ('a * 'a) set
   240 
   241 * The representation of numerals has changed.  Datatype "num"
   242 represents strictly positive binary numerals, along with functions
   243 "numeral :: num => 'a" and "neg_numeral :: num => 'a" to represent
   244 positive and negated numeric literals, respectively.  See also
   245 definitions in ~~/src/HOL/Num.thy.  Potential INCOMPATIBILITY, some
   246 user theories may require adaptations as follows:
   247 
   248   - Theorems with number_ring or number_semiring constraints: These
   249     classes are gone; use comm_ring_1 or comm_semiring_1 instead.
   250 
   251   - Theories defining numeric types: Remove number, number_semiring,
   252     and number_ring instances. Defer all theorems about numerals until
   253     after classes one and semigroup_add have been instantiated.
   254 
   255   - Numeral-only simp rules: Replace each rule having a "number_of v"
   256     pattern with two copies, one for numeral and one for neg_numeral.
   257 
   258   - Theorems about subclasses of semiring_1 or ring_1: These classes
   259     automatically support numerals now, so more simp rules and
   260     simprocs may now apply within the proof.
   261 
   262   - Definitions and theorems using old constructors Pls/Min/Bit0/Bit1:
   263     Redefine using other integer operations.
   264 
   265 * Transfer: New package intended to generalize the existing
   266 "descending" method and related theorem attributes from the Quotient
   267 package.  (Not all functionality is implemented yet, but future
   268 development will focus on Transfer as an eventual replacement for the
   269 corresponding parts of the Quotient package.)
   270 
   271   - transfer_rule attribute: Maintains a collection of transfer rules,
   272     which relate constants at two different types. Transfer rules may
   273     relate different type instances of the same polymorphic constant,
   274     or they may relate an operation on a raw type to a corresponding
   275     operation on an abstract type (quotient or subtype). For example:
   276 
   277     ((A ===> B) ===> list_all2 A ===> list_all2 B) map map
   278     (cr_int ===> cr_int ===> cr_int) (%(x,y) (u,v). (x+u, y+v)) plus_int
   279 
   280   - transfer method: Replaces a subgoal on abstract types with an
   281     equivalent subgoal on the corresponding raw types. Constants are
   282     replaced with corresponding ones according to the transfer rules.
   283     Goals are generalized over all free variables by default; this is
   284     necessary for variables whose types change, but can be overridden
   285     for specific variables with e.g. "transfer fixing: x y z".  The
   286     variant transfer' method allows replacing a subgoal with one that
   287     is logically stronger (rather than equivalent).
   288 
   289   - relator_eq attribute: Collects identity laws for relators of
   290     various type constructors, e.g. "list_all2 (op =) = (op =)".  The
   291     transfer method uses these lemmas to infer transfer rules for
   292     non-polymorphic constants on the fly.
   293 
   294   - transfer_prover method: Assists with proving a transfer rule for a
   295     new constant, provided the constant is defined in terms of other
   296     constants that already have transfer rules. It should be applied
   297     after unfolding the constant definitions.
   298 
   299   - HOL/ex/Transfer_Int_Nat.thy: Example theory demonstrating transfer
   300     from type nat to type int.
   301 
   302 * Lifting: New package intended to generalize the quotient_definition
   303 facility of the Quotient package; designed to work with Transfer.
   304 
   305   - lift_definition command: Defines operations on an abstract type in
   306     terms of a corresponding operation on a representation
   307     type.  Example syntax:
   308 
   309     lift_definition dlist_insert :: "'a => 'a dlist => 'a dlist"
   310       is List.insert
   311 
   312     Users must discharge a respectfulness proof obligation when each
   313     constant is defined. (For a type copy, i.e. a typedef with UNIV,
   314     the proof is discharged automatically.) The obligation is
   315     presented in a user-friendly, readable form; a respectfulness
   316     theorem in the standard format and a transfer rule are generated
   317     by the package.
   318 
   319   - Integration with code_abstype: For typedefs (e.g. subtypes
   320     corresponding to a datatype invariant, such as dlist),
   321     lift_definition generates a code certificate theorem and sets up
   322     code generation for each constant.
   323 
   324   - setup_lifting command: Sets up the Lifting package to work with a
   325     user-defined type. The user must provide either a quotient theorem
   326     or a type_definition theorem.  The package configures transfer
   327     rules for equality and quantifiers on the type, and sets up the
   328     lift_definition command to work with the type.
   329 
   330   - Usage examples: See Quotient_Examples/Lift_DList.thy,
   331     Quotient_Examples/Lift_RBT.thy, Quotient_Examples/Lift_FSet.thy,
   332     Word/Word.thy and Library/Float.thy.
   333 
   334 * Quotient package:
   335 
   336   - The 'quotient_type' command now supports a 'morphisms' option with
   337     rep and abs functions, similar to typedef.
   338 
   339   - 'quotient_type' sets up new types to work with the Lifting and
   340     Transfer packages, as with 'setup_lifting'.
   341 
   342   - The 'quotient_definition' command now requires the user to prove a
   343     respectfulness property at the point where the constant is
   344     defined, similar to lift_definition; INCOMPATIBILITY.
   345 
   346   - Renamed predicate 'Quotient' to 'Quotient3', and renamed theorems
   347     accordingly, INCOMPATIBILITY.
   348 
   349 * New diagnostic command 'find_unused_assms' to find potentially
   350 superfluous assumptions in theorems using Quickcheck.
   351 
   352 * Quickcheck:
   353 
   354   - Quickcheck returns variable assignments as counterexamples, which
   355     allows to reveal the underspecification of functions under test.
   356     For example, refuting "hd xs = x", it presents the variable
   357     assignment xs = [] and x = a1 as a counterexample, assuming that
   358     any property is false whenever "hd []" occurs in it.
   359 
   360     These counterexample are marked as potentially spurious, as
   361     Quickcheck also returns "xs = []" as a counterexample to the
   362     obvious theorem "hd xs = hd xs".
   363 
   364     After finding a potentially spurious counterexample, Quickcheck
   365     continues searching for genuine ones.
   366 
   367     By default, Quickcheck shows potentially spurious and genuine
   368     counterexamples. The option "genuine_only" sets quickcheck to only
   369     show genuine counterexamples.
   370 
   371   - The command 'quickcheck_generator' creates random and exhaustive
   372     value generators for a given type and operations.
   373 
   374     It generates values by using the operations as if they were
   375     constructors of that type.
   376 
   377   - Support for multisets.
   378 
   379   - Added "use_subtype" options.
   380 
   381   - Added "quickcheck_locale" configuration to specify how to process
   382     conjectures in a locale context.
   383 
   384 * Nitpick: Fixed infinite loop caused by the 'peephole_optim' option
   385 and affecting 'rat' and 'real'.
   386 
   387 * Sledgehammer:
   388   - Integrated more tightly with SPASS, as described in the ITP 2012
   389     paper "More SPASS with Isabelle".
   390   - Made it try "smt" as a fallback if "metis" fails or times out.
   391   - Added support for the following provers: Alt-Ergo (via Why3 and
   392     TFF1), iProver, iProver-Eq.
   393   - Sped up the minimizer.
   394   - Added "lam_trans", "uncurry_aliases", and "minimize" options.
   395   - Renamed "slicing" ("no_slicing") option to "slice" ("dont_slice").
   396   - Renamed "sound" option to "strict".
   397 
   398 * Metis: Added possibility to specify lambda translations scheme as a
   399 parenthesized argument (e.g., "by (metis (lifting) ...)").
   400 
   401 * SMT: Renamed "smt_fixed" option to "smt_read_only_certificates".
   402 
   403 * Command 'try0': Renamed from 'try_methods'. INCOMPATIBILITY.
   404 
   405 * New "case_product" attribute to generate a case rule doing multiple
   406 case distinctions at the same time.  E.g.
   407 
   408   list.exhaust [case_product nat.exhaust]
   409 
   410 produces a rule which can be used to perform case distinction on both
   411 a list and a nat.
   412 
   413 * New "eventually_elim" method as a generalized variant of the
   414 eventually_elim* rules.  Supports structured proofs.
   415 
   416 * Typedef with implicit set definition is considered legacy.  Use
   417 "typedef (open)" form instead, which will eventually become the
   418 default.
   419 
   420 * Record: code generation can be switched off manually with
   421 
   422   declare [[record_coden = false]]  -- "default true"
   423 
   424 * Datatype: type parameters allow explicit sort constraints.
   425 
   426 * Concrete syntax for case expressions includes constraints for source
   427 positions, and thus produces Prover IDE markup for its bindings.
   428 INCOMPATIBILITY for old-style syntax translations that augment the
   429 pattern notation; e.g. see src/HOL/HOLCF/One.thy for translations of
   430 one_case.
   431 
   432 * Clarified attribute "mono_set": pure declaration without modifying
   433 the result of the fact expression.
   434 
   435 * More default pred/set conversions on a couple of relation operations
   436 and predicates.  Added powers of predicate relations.  Consolidation
   437 of some relation theorems:
   438 
   439   converse_def ~> converse_unfold
   440   rel_comp_def ~> relcomp_unfold
   441   symp_def ~> (modified, use symp_def and sym_def instead)
   442   transp_def ~> transp_trans
   443   Domain_def ~> Domain_unfold
   444   Range_def ~> Domain_converse [symmetric]
   445 
   446 Generalized theorems INF_INT_eq, INF_INT_eq2, SUP_UN_eq, SUP_UN_eq2.
   447 
   448 See theory "Relation" for examples for making use of pred/set
   449 conversions by means of attributes "to_set" and "to_pred".
   450 
   451 INCOMPATIBILITY.
   452 
   453 * Renamed facts about the power operation on relations, i.e., relpow
   454 to match the constant's name:
   455 
   456   rel_pow_1 ~> relpow_1
   457   rel_pow_0_I ~> relpow_0_I
   458   rel_pow_Suc_I ~> relpow_Suc_I
   459   rel_pow_Suc_I2 ~> relpow_Suc_I2
   460   rel_pow_0_E ~> relpow_0_E
   461   rel_pow_Suc_E ~> relpow_Suc_E
   462   rel_pow_E ~> relpow_E
   463   rel_pow_Suc_D2 ~> relpow_Suc_D2
   464   rel_pow_Suc_E2 ~> relpow_Suc_E2
   465   rel_pow_Suc_D2' ~> relpow_Suc_D2'
   466   rel_pow_E2 ~> relpow_E2
   467   rel_pow_add ~> relpow_add
   468   rel_pow_commute ~> relpow
   469   rel_pow_empty ~> relpow_empty:
   470   rtrancl_imp_UN_rel_pow ~> rtrancl_imp_UN_relpow
   471   rel_pow_imp_rtrancl ~> relpow_imp_rtrancl
   472   rtrancl_is_UN_rel_pow ~> rtrancl_is_UN_relpow
   473   rtrancl_imp_rel_pow ~> rtrancl_imp_relpow
   474   rel_pow_fun_conv ~> relpow_fun_conv
   475   rel_pow_finite_bounded1 ~> relpow_finite_bounded1
   476   rel_pow_finite_bounded ~> relpow_finite_bounded
   477   rtrancl_finite_eq_rel_pow ~> rtrancl_finite_eq_relpow
   478   trancl_finite_eq_rel_pow ~> trancl_finite_eq_relpow
   479   single_valued_rel_pow ~> single_valued_relpow
   480 
   481 INCOMPATIBILITY.
   482 
   483 * Theory Relation: Consolidated constant name for relation composition
   484 and corresponding theorem names:
   485 
   486   - Renamed constant rel_comp to relcomp.
   487 
   488   - Dropped abbreviation pred_comp. Use relcompp instead.
   489 
   490   - Renamed theorems:
   491 
   492     rel_compI ~> relcompI
   493     rel_compEpair ~> relcompEpair
   494     rel_compE ~> relcompE
   495     pred_comp_rel_comp_eq ~> relcompp_relcomp_eq
   496     rel_comp_empty1 ~> relcomp_empty1
   497     rel_comp_mono ~> relcomp_mono
   498     rel_comp_subset_Sigma ~> relcomp_subset_Sigma
   499     rel_comp_distrib ~> relcomp_distrib
   500     rel_comp_distrib2 ~> relcomp_distrib2
   501     rel_comp_UNION_distrib ~> relcomp_UNION_distrib
   502     rel_comp_UNION_distrib2 ~> relcomp_UNION_distrib2
   503     single_valued_rel_comp ~> single_valued_relcomp
   504     rel_comp_def ~> relcomp_unfold
   505     converse_rel_comp ~> converse_relcomp
   506     pred_compI ~> relcomppI
   507     pred_compE ~> relcomppE
   508     pred_comp_bot1 ~> relcompp_bot1
   509     pred_comp_bot2 ~> relcompp_bot2
   510     transp_pred_comp_less_eq ~> transp_relcompp_less_eq
   511     pred_comp_mono ~> relcompp_mono
   512     pred_comp_distrib ~> relcompp_distrib
   513     pred_comp_distrib2 ~> relcompp_distrib2
   514     converse_pred_comp ~> converse_relcompp
   515 
   516     finite_rel_comp ~> finite_relcomp
   517 
   518     set_rel_comp ~> set_relcomp
   519 
   520 INCOMPATIBILITY.
   521 
   522 * Theory Divides: Discontinued redundant theorems about div and mod.
   523 INCOMPATIBILITY, use the corresponding generic theorems instead.
   524 
   525   DIVISION_BY_ZERO ~> div_by_0, mod_by_0
   526   zdiv_self ~> div_self
   527   zmod_self ~> mod_self
   528   zdiv_zero ~> div_0
   529   zmod_zero ~> mod_0
   530   zdiv_zmod_equality ~> div_mod_equality2
   531   zdiv_zmod_equality2 ~> div_mod_equality
   532   zmod_zdiv_trivial ~> mod_div_trivial
   533   zdiv_zminus_zminus ~> div_minus_minus
   534   zmod_zminus_zminus ~> mod_minus_minus
   535   zdiv_zminus2 ~> div_minus_right
   536   zmod_zminus2 ~> mod_minus_right
   537   zdiv_minus1_right ~> div_minus1_right
   538   zmod_minus1_right ~> mod_minus1_right
   539   zdvd_mult_div_cancel ~> dvd_mult_div_cancel
   540   zmod_zmult1_eq ~> mod_mult_right_eq
   541   zpower_zmod ~> power_mod
   542   zdvd_zmod ~> dvd_mod
   543   zdvd_zmod_imp_zdvd ~> dvd_mod_imp_dvd
   544   mod_mult_distrib ~> mult_mod_left
   545   mod_mult_distrib2 ~> mult_mod_right
   546 
   547 * Removed redundant theorems nat_mult_2 and nat_mult_2_right; use
   548 generic mult_2 and mult_2_right instead. INCOMPATIBILITY.
   549 
   550 * Finite_Set.fold now qualified.  INCOMPATIBILITY.
   551 
   552 * Consolidated theorem names concerning fold combinators:
   553 
   554   inf_INFI_fold_inf ~> inf_INF_fold_inf
   555   sup_SUPR_fold_sup ~> sup_SUP_fold_sup
   556   INFI_fold_inf ~> INF_fold_inf
   557   SUPR_fold_sup ~> SUP_fold_sup
   558   union_set ~> union_set_fold
   559   minus_set ~> minus_set_fold
   560   INFI_set_fold ~> INF_set_fold
   561   SUPR_set_fold ~> SUP_set_fold
   562   INF_code ~> INF_set_foldr
   563   SUP_code ~> SUP_set_foldr
   564   foldr.simps ~> foldr.simps (in point-free formulation)
   565   foldr_fold_rev ~> foldr_conv_fold
   566   foldl_fold ~> foldl_conv_fold
   567   foldr_foldr ~> foldr_conv_foldl
   568   foldl_foldr ~> foldl_conv_foldr
   569   fold_set_remdups ~> fold_set_fold_remdups
   570   fold_set ~> fold_set_fold
   571   fold1_set ~> fold1_set_fold
   572 
   573 INCOMPATIBILITY.
   574 
   575 * Dropped rarely useful theorems concerning fold combinators:
   576 foldl_apply, foldl_fun_comm, foldl_rev, fold_weak_invariant,
   577 rev_foldl_cons, fold_set_remdups, fold_set, fold_set1,
   578 concat_conv_foldl, foldl_weak_invariant, foldl_invariant,
   579 foldr_invariant, foldl_absorb0, foldl_foldr1_lemma, foldl_foldr1,
   580 listsum_conv_fold, listsum_foldl, sort_foldl_insort, foldl_assoc,
   581 foldr_conv_foldl, start_le_sum, elem_le_sum, sum_eq_0_conv.
   582 INCOMPATIBILITY.  For the common phrases "%xs. List.foldr plus xs 0"
   583 and "List.foldl plus 0", prefer "List.listsum".  Otherwise it can be
   584 useful to boil down "List.foldr" and "List.foldl" to "List.fold" by
   585 unfolding "foldr_conv_fold" and "foldl_conv_fold".
   586 
   587 * Dropped lemmas minus_set_foldr, union_set_foldr, union_coset_foldr,
   588 inter_coset_foldr, Inf_fin_set_foldr, Sup_fin_set_foldr,
   589 Min_fin_set_foldr, Max_fin_set_foldr, Inf_set_foldr, Sup_set_foldr,
   590 INF_set_foldr, SUP_set_foldr.  INCOMPATIBILITY.  Prefer corresponding
   591 lemmas over fold rather than foldr, or make use of lemmas
   592 fold_conv_foldr and fold_rev.
   593 
   594 * Congruence rules Option.map_cong and Option.bind_cong for recursion
   595 through option types.
   596 
   597 * "Transitive_Closure.ntrancl": bounded transitive closure on
   598 relations.
   599 
   600 * Constant "Set.not_member" now qualified.  INCOMPATIBILITY.
   601 
   602 * Theory Int: Discontinued many legacy theorems specific to type int.
   603 INCOMPATIBILITY, use the corresponding generic theorems instead.
   604 
   605   zminus_zminus ~> minus_minus
   606   zminus_0 ~> minus_zero
   607   zminus_zadd_distrib ~> minus_add_distrib
   608   zadd_commute ~> add_commute
   609   zadd_assoc ~> add_assoc
   610   zadd_left_commute ~> add_left_commute
   611   zadd_ac ~> add_ac
   612   zmult_ac ~> mult_ac
   613   zadd_0 ~> add_0_left
   614   zadd_0_right ~> add_0_right
   615   zadd_zminus_inverse2 ~> left_minus
   616   zmult_zminus ~> mult_minus_left
   617   zmult_commute ~> mult_commute
   618   zmult_assoc ~> mult_assoc
   619   zadd_zmult_distrib ~> left_distrib
   620   zadd_zmult_distrib2 ~> right_distrib
   621   zdiff_zmult_distrib ~> left_diff_distrib
   622   zdiff_zmult_distrib2 ~> right_diff_distrib
   623   zmult_1 ~> mult_1_left
   624   zmult_1_right ~> mult_1_right
   625   zle_refl ~> order_refl
   626   zle_trans ~> order_trans
   627   zle_antisym ~> order_antisym
   628   zle_linear ~> linorder_linear
   629   zless_linear ~> linorder_less_linear
   630   zadd_left_mono ~> add_left_mono
   631   zadd_strict_right_mono ~> add_strict_right_mono
   632   zadd_zless_mono ~> add_less_le_mono
   633   int_0_less_1 ~> zero_less_one
   634   int_0_neq_1 ~> zero_neq_one
   635   zless_le ~> less_le
   636   zpower_zadd_distrib ~> power_add
   637   zero_less_zpower_abs_iff ~> zero_less_power_abs_iff
   638   zero_le_zpower_abs ~> zero_le_power_abs
   639 
   640 * Theory Deriv: Renamed
   641 
   642   DERIV_nonneg_imp_nonincreasing ~> DERIV_nonneg_imp_nondecreasing
   643 
   644 * Theory Library/Multiset: Improved code generation of multisets.
   645 
   646 * Theory HOL/Library/Set_Algebras: Addition and multiplication on sets
   647 are expressed via type classes again. The special syntax
   648 \<oplus>/\<otimes> has been replaced by plain +/*. Removed constant
   649 setsum_set, which is now subsumed by Big_Operators.setsum.
   650 INCOMPATIBILITY.
   651 
   652 * Theory HOL/Library/Diagonalize has been removed. INCOMPATIBILITY,
   653 use theory HOL/Library/Nat_Bijection instead.
   654 
   655 * Theory HOL/Library/RBT_Impl: Backing implementation of red-black
   656 trees is now inside a type class context.  Names of affected
   657 operations and lemmas have been prefixed by rbt_.  INCOMPATIBILITY for
   658 theories working directly with raw red-black trees, adapt the names as
   659 follows:
   660 
   661   Operations:
   662   bulkload -> rbt_bulkload
   663   del_from_left -> rbt_del_from_left
   664   del_from_right -> rbt_del_from_right
   665   del -> rbt_del
   666   delete -> rbt_delete
   667   ins -> rbt_ins
   668   insert -> rbt_insert
   669   insertw -> rbt_insert_with
   670   insert_with_key -> rbt_insert_with_key
   671   map_entry -> rbt_map_entry
   672   lookup -> rbt_lookup
   673   sorted -> rbt_sorted
   674   tree_greater -> rbt_greater
   675   tree_less -> rbt_less
   676   tree_less_symbol -> rbt_less_symbol
   677   union -> rbt_union
   678   union_with -> rbt_union_with
   679   union_with_key -> rbt_union_with_key
   680 
   681   Lemmas:
   682   balance_left_sorted -> balance_left_rbt_sorted
   683   balance_left_tree_greater -> balance_left_rbt_greater
   684   balance_left_tree_less -> balance_left_rbt_less
   685   balance_right_sorted -> balance_right_rbt_sorted
   686   balance_right_tree_greater -> balance_right_rbt_greater
   687   balance_right_tree_less -> balance_right_rbt_less
   688   balance_sorted -> balance_rbt_sorted
   689   balance_tree_greater -> balance_rbt_greater
   690   balance_tree_less -> balance_rbt_less
   691   bulkload_is_rbt -> rbt_bulkload_is_rbt
   692   combine_sorted -> combine_rbt_sorted
   693   combine_tree_greater -> combine_rbt_greater
   694   combine_tree_less -> combine_rbt_less
   695   delete_in_tree -> rbt_delete_in_tree
   696   delete_is_rbt -> rbt_delete_is_rbt
   697   del_from_left_tree_greater -> rbt_del_from_left_rbt_greater
   698   del_from_left_tree_less -> rbt_del_from_left_rbt_less
   699   del_from_right_tree_greater -> rbt_del_from_right_rbt_greater
   700   del_from_right_tree_less -> rbt_del_from_right_rbt_less
   701   del_in_tree -> rbt_del_in_tree
   702   del_inv1_inv2 -> rbt_del_inv1_inv2
   703   del_sorted -> rbt_del_rbt_sorted
   704   del_tree_greater -> rbt_del_rbt_greater
   705   del_tree_less -> rbt_del_rbt_less
   706   dom_lookup_Branch -> dom_rbt_lookup_Branch
   707   entries_lookup -> entries_rbt_lookup
   708   finite_dom_lookup -> finite_dom_rbt_lookup
   709   insert_sorted -> rbt_insert_rbt_sorted
   710   insertw_is_rbt -> rbt_insertw_is_rbt
   711   insertwk_is_rbt -> rbt_insertwk_is_rbt
   712   insertwk_sorted -> rbt_insertwk_rbt_sorted
   713   insertw_sorted -> rbt_insertw_rbt_sorted
   714   ins_sorted -> ins_rbt_sorted
   715   ins_tree_greater -> ins_rbt_greater
   716   ins_tree_less -> ins_rbt_less
   717   is_rbt_sorted -> is_rbt_rbt_sorted
   718   lookup_balance -> rbt_lookup_balance
   719   lookup_bulkload -> rbt_lookup_rbt_bulkload
   720   lookup_delete -> rbt_lookup_rbt_delete
   721   lookup_Empty -> rbt_lookup_Empty
   722   lookup_from_in_tree -> rbt_lookup_from_in_tree
   723   lookup_in_tree -> rbt_lookup_in_tree
   724   lookup_ins -> rbt_lookup_ins
   725   lookup_insert -> rbt_lookup_rbt_insert
   726   lookup_insertw -> rbt_lookup_rbt_insertw
   727   lookup_insertwk -> rbt_lookup_rbt_insertwk
   728   lookup_keys -> rbt_lookup_keys
   729   lookup_map -> rbt_lookup_map
   730   lookup_map_entry -> rbt_lookup_rbt_map_entry
   731   lookup_tree_greater -> rbt_lookup_rbt_greater
   732   lookup_tree_less -> rbt_lookup_rbt_less
   733   lookup_union -> rbt_lookup_rbt_union
   734   map_entry_color_of -> rbt_map_entry_color_of
   735   map_entry_inv1 -> rbt_map_entry_inv1
   736   map_entry_inv2 -> rbt_map_entry_inv2
   737   map_entry_is_rbt -> rbt_map_entry_is_rbt
   738   map_entry_sorted -> rbt_map_entry_rbt_sorted
   739   map_entry_tree_greater -> rbt_map_entry_rbt_greater
   740   map_entry_tree_less -> rbt_map_entry_rbt_less
   741   map_tree_greater -> map_rbt_greater
   742   map_tree_less -> map_rbt_less
   743   map_sorted -> map_rbt_sorted
   744   paint_sorted -> paint_rbt_sorted
   745   paint_lookup -> paint_rbt_lookup
   746   paint_tree_greater -> paint_rbt_greater
   747   paint_tree_less -> paint_rbt_less
   748   sorted_entries -> rbt_sorted_entries
   749   tree_greater_eq_trans -> rbt_greater_eq_trans
   750   tree_greater_nit -> rbt_greater_nit
   751   tree_greater_prop -> rbt_greater_prop
   752   tree_greater_simps -> rbt_greater_simps
   753   tree_greater_trans -> rbt_greater_trans
   754   tree_less_eq_trans -> rbt_less_eq_trans
   755   tree_less_nit -> rbt_less_nit
   756   tree_less_prop -> rbt_less_prop
   757   tree_less_simps -> rbt_less_simps
   758   tree_less_trans -> rbt_less_trans
   759   tree_ord_props -> rbt_ord_props
   760   union_Branch -> rbt_union_Branch
   761   union_is_rbt -> rbt_union_is_rbt
   762   unionw_is_rbt -> rbt_unionw_is_rbt
   763   unionwk_is_rbt -> rbt_unionwk_is_rbt
   764   unionwk_sorted -> rbt_unionwk_rbt_sorted
   765 
   766 * Theory HOL/Library/Float: Floating point numbers are now defined as
   767 a subset of the real numbers.  All operations are defined using the
   768 lifing-framework and proofs use the transfer method.  INCOMPATIBILITY.
   769 
   770   Changed Operations:
   771   float_abs -> abs
   772   float_nprt -> nprt
   773   float_pprt -> pprt
   774   pow2 -> use powr
   775   round_down -> float_round_down
   776   round_up -> float_round_up
   777   scale -> exponent
   778 
   779   Removed Operations:
   780   ceiling_fl, lb_mult, lb_mod, ub_mult, ub_mod
   781 
   782   Renamed Lemmas:
   783   abs_float_def -> Float.compute_float_abs
   784   bitlen_ge0 -> bitlen_nonneg
   785   bitlen.simps -> Float.compute_bitlen
   786   float_components -> Float_mantissa_exponent
   787   float_divl.simps -> Float.compute_float_divl
   788   float_divr.simps -> Float.compute_float_divr
   789   float_eq_odd -> mult_powr_eq_mult_powr_iff
   790   float_power -> real_of_float_power
   791   lapprox_posrat_def -> Float.compute_lapprox_posrat
   792   lapprox_rat.simps -> Float.compute_lapprox_rat
   793   le_float_def' -> Float.compute_float_le
   794   le_float_def -> less_eq_float.rep_eq
   795   less_float_def' -> Float.compute_float_less
   796   less_float_def -> less_float.rep_eq
   797   normfloat_def -> Float.compute_normfloat
   798   normfloat_imp_odd_or_zero -> mantissa_not_dvd and mantissa_noteq_0
   799   normfloat -> normfloat_def
   800   normfloat_unique -> use normfloat_def
   801   number_of_float_Float -> Float.compute_float_numeral, Float.compute_float_neg_numeral
   802   one_float_def -> Float.compute_float_one
   803   plus_float_def -> Float.compute_float_plus
   804   rapprox_posrat_def -> Float.compute_rapprox_posrat
   805   rapprox_rat.simps -> Float.compute_rapprox_rat
   806   real_of_float_0 -> zero_float.rep_eq
   807   real_of_float_1 -> one_float.rep_eq
   808   real_of_float_abs -> abs_float.rep_eq
   809   real_of_float_add -> plus_float.rep_eq
   810   real_of_float_minus -> uminus_float.rep_eq
   811   real_of_float_mult -> times_float.rep_eq
   812   real_of_float_simp -> Float.rep_eq
   813   real_of_float_sub -> minus_float.rep_eq
   814   round_down.simps -> Float.compute_float_round_down
   815   round_up.simps -> Float.compute_float_round_up
   816   times_float_def -> Float.compute_float_times
   817   uminus_float_def -> Float.compute_float_uminus
   818   zero_float_def -> Float.compute_float_zero
   819 
   820   Lemmas not necessary anymore, use the transfer method:
   821   bitlen_B0, bitlen_B1, bitlen_ge1, bitlen_Min, bitlen_Pls, float_divl,
   822   float_divr, float_le_simp, float_less1_mantissa_bound,
   823   float_less_simp, float_less_zero, float_le_zero,
   824   float_pos_less1_e_neg, float_pos_m_pos, float_split, float_split2,
   825   floor_pos_exp, lapprox_posrat, lapprox_posrat_bottom, lapprox_rat,
   826   lapprox_rat_bottom, normalized_float, rapprox_posrat,
   827   rapprox_posrat_le1, rapprox_rat, real_of_float_ge0_exp,
   828   real_of_float_neg_exp, real_of_float_nge0_exp, round_down floor_fl,
   829   round_up, zero_le_float, zero_less_float
   830 
   831 * New theory HOL/Library/DAList provides an abstract type for
   832 association lists with distinct keys.
   833 
   834 * Session HOL/IMP: Added new theory of abstract interpretation of
   835 annotated commands.
   836 
   837 * Session HOL-Import: Re-implementation from scratch is faster,
   838 simpler, and more scalable.  Requires a proof bundle, which is
   839 available as an external component.  Discontinued old (and mostly
   840 dead) Importer for HOL4 and HOL Light.  INCOMPATIBILITY.
   841 
   842 * Session HOL-Word: Discontinued many redundant theorems specific to
   843 type 'a word. INCOMPATIBILITY, use the corresponding generic theorems
   844 instead.
   845 
   846   word_sub_alt ~> word_sub_wi
   847   word_add_alt ~> word_add_def
   848   word_mult_alt ~> word_mult_def
   849   word_minus_alt ~> word_minus_def
   850   word_0_alt ~> word_0_wi
   851   word_1_alt ~> word_1_wi
   852   word_add_0 ~> add_0_left
   853   word_add_0_right ~> add_0_right
   854   word_mult_1 ~> mult_1_left
   855   word_mult_1_right ~> mult_1_right
   856   word_add_commute ~> add_commute
   857   word_add_assoc ~> add_assoc
   858   word_add_left_commute ~> add_left_commute
   859   word_mult_commute ~> mult_commute
   860   word_mult_assoc ~> mult_assoc
   861   word_mult_left_commute ~> mult_left_commute
   862   word_left_distrib ~> left_distrib
   863   word_right_distrib ~> right_distrib
   864   word_left_minus ~> left_minus
   865   word_diff_0_right ~> diff_0_right
   866   word_diff_self ~> diff_self
   867   word_sub_def ~> diff_minus
   868   word_diff_minus ~> diff_minus
   869   word_add_ac ~> add_ac
   870   word_mult_ac ~> mult_ac
   871   word_plus_ac0 ~> add_0_left add_0_right add_ac
   872   word_times_ac1 ~> mult_1_left mult_1_right mult_ac
   873   word_order_trans ~> order_trans
   874   word_order_refl ~> order_refl
   875   word_order_antisym ~> order_antisym
   876   word_order_linear ~> linorder_linear
   877   lenw1_zero_neq_one ~> zero_neq_one
   878   word_number_of_eq ~> number_of_eq
   879   word_of_int_add_hom ~> wi_hom_add
   880   word_of_int_sub_hom ~> wi_hom_sub
   881   word_of_int_mult_hom ~> wi_hom_mult
   882   word_of_int_minus_hom ~> wi_hom_neg
   883   word_of_int_succ_hom ~> wi_hom_succ
   884   word_of_int_pred_hom ~> wi_hom_pred
   885   word_of_int_0_hom ~> word_0_wi
   886   word_of_int_1_hom ~> word_1_wi
   887 
   888 * Session HOL-Word: New proof method "word_bitwise" for splitting
   889 machine word equalities and inequalities into logical circuits,
   890 defined in HOL/Word/WordBitwise.thy.  Supports addition, subtraction,
   891 multiplication, shifting by constants, bitwise operators and numeric
   892 constants.  Requires fixed-length word types, not 'a word.  Solves
   893 many standard word identities outright and converts more into first
   894 order problems amenable to blast or similar.  See also examples in
   895 HOL/Word/Examples/WordExamples.thy.
   896 
   897 * Session HOL-Probability: Introduced the type "'a measure" to
   898 represent measures, this replaces the records 'a algebra and 'a
   899 measure_space.  The locales based on subset_class now have two
   900 locale-parameters the space \<Omega> and the set of measurable sets M.
   901 The product of probability spaces uses now the same constant as the
   902 finite product of sigma-finite measure spaces "PiM :: ('i => 'a)
   903 measure".  Most constants are defined now outside of locales and gain
   904 an additional parameter, like null_sets, almost_eventually or \<mu>'.
   905 Measure space constructions for distributions and densities now got
   906 their own constants distr and density.  Instead of using locales to
   907 describe measure spaces with a finite space, the measure count_space
   908 and point_measure is introduced.  INCOMPATIBILITY.
   909 
   910   Renamed constants:
   911   measure -> emeasure
   912   finite_measure.\<mu>' -> measure
   913   product_algebra_generator -> prod_algebra
   914   product_prob_space.emb -> prod_emb
   915   product_prob_space.infprod_algebra -> PiM
   916 
   917   Removed locales:
   918   completeable_measure_space
   919   finite_measure_space
   920   finite_prob_space
   921   finite_product_finite_prob_space
   922   finite_product_sigma_algebra
   923   finite_sigma_algebra
   924   measure_space
   925   pair_finite_prob_space
   926   pair_finite_sigma_algebra
   927   pair_finite_space
   928   pair_sigma_algebra
   929   product_sigma_algebra
   930 
   931   Removed constants:
   932   conditional_space
   933   distribution -> use distr measure, or distributed predicate
   934   image_space
   935   joint_distribution -> use distr measure, or distributed predicate
   936   pair_measure_generator
   937   product_prob_space.infprod_algebra -> use PiM
   938   subvimage
   939 
   940   Replacement theorems:
   941   finite_additivity_sufficient -> ring_of_sets.countably_additiveI_finite
   942   finite_measure.empty_measure -> measure_empty
   943   finite_measure.finite_continuity_from_above -> finite_measure.finite_Lim_measure_decseq
   944   finite_measure.finite_continuity_from_below -> finite_measure.finite_Lim_measure_incseq
   945   finite_measure.finite_measure_countably_subadditive -> finite_measure.finite_measure_subadditive_countably
   946   finite_measure.finite_measure_eq -> finite_measure.emeasure_eq_measure
   947   finite_measure.finite_measure -> finite_measure.emeasure_finite
   948   finite_measure.finite_measure_finite_singleton -> finite_measure.finite_measure_eq_setsum_singleton
   949   finite_measure.positive_measure' -> measure_nonneg
   950   finite_measure.real_measure -> finite_measure.emeasure_real
   951   finite_product_prob_space.finite_measure_times -> finite_product_prob_space.finite_measure_PiM_emb
   952   finite_product_sigma_algebra.in_P -> sets_PiM_I_finite
   953   finite_product_sigma_algebra.P_empty -> space_PiM_empty, sets_PiM_empty
   954   information_space.conditional_entropy_eq -> information_space.conditional_entropy_simple_distributed
   955   information_space.conditional_entropy_positive -> information_space.conditional_entropy_nonneg_simple
   956   information_space.conditional_mutual_information_eq_mutual_information -> information_space.conditional_mutual_information_eq_mutual_information_simple
   957   information_space.conditional_mutual_information_generic_positive -> information_space.conditional_mutual_information_nonneg_simple
   958   information_space.conditional_mutual_information_positive -> information_space.conditional_mutual_information_nonneg_simple
   959   information_space.entropy_commute -> information_space.entropy_commute_simple
   960   information_space.entropy_eq -> information_space.entropy_simple_distributed
   961   information_space.entropy_generic_eq -> information_space.entropy_simple_distributed
   962   information_space.entropy_positive -> information_space.entropy_nonneg_simple
   963   information_space.entropy_uniform_max -> information_space.entropy_uniform
   964   information_space.KL_eq_0_imp -> information_space.KL_eq_0_iff_eq
   965   information_space.KL_eq_0 -> information_space.KL_same_eq_0
   966   information_space.KL_ge_0 -> information_space.KL_nonneg
   967   information_space.mutual_information_eq -> information_space.mutual_information_simple_distributed
   968   information_space.mutual_information_positive -> information_space.mutual_information_nonneg_simple
   969   Int_stable_cuboids -> Int_stable_atLeastAtMost
   970   Int_stable_product_algebra_generator -> positive_integral
   971   measure_preserving -> equality "distr M N f = N" "f : measurable M N"
   972   measure_space.additive -> emeasure_additive
   973   measure_space.AE_iff_null_set -> AE_iff_null
   974   measure_space.almost_everywhere_def -> eventually_ae_filter
   975   measure_space.almost_everywhere_vimage -> AE_distrD
   976   measure_space.continuity_from_above -> INF_emeasure_decseq
   977   measure_space.continuity_from_above_Lim -> Lim_emeasure_decseq
   978   measure_space.continuity_from_below_Lim -> Lim_emeasure_incseq
   979   measure_space.continuity_from_below -> SUP_emeasure_incseq
   980   measure_space_density -> emeasure_density
   981   measure_space.density_is_absolutely_continuous -> absolutely_continuousI_density
   982   measure_space.integrable_vimage -> integrable_distr
   983   measure_space.integral_translated_density -> integral_density
   984   measure_space.integral_vimage -> integral_distr
   985   measure_space.measure_additive -> plus_emeasure
   986   measure_space.measure_compl -> emeasure_compl
   987   measure_space.measure_countable_increasing -> emeasure_countable_increasing
   988   measure_space.measure_countably_subadditive -> emeasure_subadditive_countably
   989   measure_space.measure_decseq -> decseq_emeasure
   990   measure_space.measure_Diff -> emeasure_Diff
   991   measure_space.measure_Diff_null_set -> emeasure_Diff_null_set
   992   measure_space.measure_eq_0 -> emeasure_eq_0
   993   measure_space.measure_finitely_subadditive -> emeasure_subadditive_finite
   994   measure_space.measure_finite_singleton -> emeasure_eq_setsum_singleton
   995   measure_space.measure_incseq -> incseq_emeasure
   996   measure_space.measure_insert -> emeasure_insert
   997   measure_space.measure_mono -> emeasure_mono
   998   measure_space.measure_not_negative -> emeasure_not_MInf
   999   measure_space.measure_preserving_Int_stable -> measure_eqI_generator_eq
  1000   measure_space.measure_setsum -> setsum_emeasure
  1001   measure_space.measure_setsum_split -> setsum_emeasure_cover
  1002   measure_space.measure_space_vimage -> emeasure_distr
  1003   measure_space.measure_subadditive_finite -> emeasure_subadditive_finite
  1004   measure_space.measure_subadditive -> subadditive
  1005   measure_space.measure_top -> emeasure_space
  1006   measure_space.measure_UN_eq_0 -> emeasure_UN_eq_0
  1007   measure_space.measure_Un_null_set -> emeasure_Un_null_set
  1008   measure_space.positive_integral_translated_density -> positive_integral_density
  1009   measure_space.positive_integral_vimage -> positive_integral_distr
  1010   measure_space.real_continuity_from_above -> Lim_measure_decseq
  1011   measure_space.real_continuity_from_below -> Lim_measure_incseq
  1012   measure_space.real_measure_countably_subadditive -> measure_subadditive_countably
  1013   measure_space.real_measure_Diff -> measure_Diff
  1014   measure_space.real_measure_finite_Union -> measure_finite_Union
  1015   measure_space.real_measure_setsum_singleton -> measure_eq_setsum_singleton
  1016   measure_space.real_measure_subadditive -> measure_subadditive
  1017   measure_space.real_measure_Union -> measure_Union
  1018   measure_space.real_measure_UNION -> measure_UNION
  1019   measure_space.simple_function_vimage -> simple_function_comp
  1020   measure_space.simple_integral_vimage -> simple_integral_distr
  1021   measure_space.simple_integral_vimage -> simple_integral_distr
  1022   measure_unique_Int_stable -> measure_eqI_generator_eq
  1023   measure_unique_Int_stable_vimage -> measure_eqI_generator_eq
  1024   pair_sigma_algebra.measurable_cut_fst -> sets_Pair1
  1025   pair_sigma_algebra.measurable_cut_snd -> sets_Pair2
  1026   pair_sigma_algebra.measurable_pair_image_fst -> measurable_Pair1
  1027   pair_sigma_algebra.measurable_pair_image_snd -> measurable_Pair2
  1028   pair_sigma_algebra.measurable_product_swap -> measurable_pair_swap_iff
  1029   pair_sigma_algebra.pair_sigma_algebra_measurable -> measurable_pair_swap
  1030   pair_sigma_algebra.pair_sigma_algebra_swap_measurable -> measurable_pair_swap'
  1031   pair_sigma_algebra.sets_swap -> sets_pair_swap
  1032   pair_sigma_finite.measure_cut_measurable_fst -> pair_sigma_finite.measurable_emeasure_Pair1
  1033   pair_sigma_finite.measure_cut_measurable_snd -> pair_sigma_finite.measurable_emeasure_Pair2
  1034   pair_sigma_finite.measure_preserving_swap -> pair_sigma_finite.distr_pair_swap
  1035   pair_sigma_finite.pair_measure_alt2 -> pair_sigma_finite.emeasure_pair_measure_alt2
  1036   pair_sigma_finite.pair_measure_alt -> pair_sigma_finite.emeasure_pair_measure_alt
  1037   pair_sigma_finite.pair_measure_times -> pair_sigma_finite.emeasure_pair_measure_Times
  1038   prob_space.indep_distribution_eq_measure -> prob_space.indep_vars_iff_distr_eq_PiM
  1039   prob_space.indep_var_distributionD -> prob_space.indep_var_distribution_eq
  1040   prob_space.measure_space_1 -> prob_space.emeasure_space_1
  1041   prob_space.prob_space_vimage -> prob_space_distr
  1042   prob_space.random_variable_restrict -> measurable_restrict
  1043   prob_space_unique_Int_stable -> measure_eqI_prob_space
  1044   product_algebraE -> prod_algebraE_all
  1045   product_algebra_generator_der -> prod_algebra_eq_finite
  1046   product_algebra_generator_into_space -> prod_algebra_sets_into_space
  1047   product_algebraI -> sets_PiM_I_finite
  1048   product_measure_exists -> product_sigma_finite.sigma_finite
  1049   product_prob_space.finite_index_eq_finite_product -> product_prob_space.sets_PiM_generator
  1050   product_prob_space.finite_measure_infprod_emb_Pi -> product_prob_space.measure_PiM_emb
  1051   product_prob_space.infprod_spec -> product_prob_space.emeasure_PiM_emb_not_empty
  1052   product_prob_space.measurable_component -> measurable_component_singleton
  1053   product_prob_space.measurable_emb -> measurable_prod_emb
  1054   product_prob_space.measurable_into_infprod_algebra -> measurable_PiM_single
  1055   product_prob_space.measurable_singleton_infprod -> measurable_component_singleton
  1056   product_prob_space.measure_emb -> emeasure_prod_emb
  1057   product_prob_space.measure_preserving_restrict -> product_prob_space.distr_restrict
  1058   product_sigma_algebra.product_algebra_into_space -> space_closed
  1059   product_sigma_finite.measure_fold -> product_sigma_finite.distr_merge
  1060   product_sigma_finite.measure_preserving_component_singelton -> product_sigma_finite.distr_singleton
  1061   product_sigma_finite.measure_preserving_merge -> product_sigma_finite.distr_merge
  1062   sequence_space.measure_infprod -> sequence_space.measure_PiM_countable
  1063   sets_product_algebra -> sets_PiM
  1064   sigma_algebra.measurable_sigma -> measurable_measure_of
  1065   sigma_finite_measure.disjoint_sigma_finite -> sigma_finite_disjoint
  1066   sigma_finite_measure.RN_deriv_vimage -> sigma_finite_measure.RN_deriv_distr
  1067   sigma_product_algebra_sigma_eq -> sigma_prod_algebra_sigma_eq
  1068   space_product_algebra -> space_PiM
  1069 
  1070 * Session HOL-TPTP: support to parse and import TPTP problems (all
  1071 languages) into Isabelle/HOL.
  1072 
  1073 
  1074 *** FOL ***
  1075 
  1076 * New "case_product" attribute (see HOL).
  1077 
  1078 
  1079 *** ZF ***
  1080 
  1081 * Greater support for structured proofs involving induction or case
  1082 analysis.
  1083 
  1084 * Much greater use of mathematical symbols.
  1085 
  1086 * Removal of many ML theorem bindings.  INCOMPATIBILITY.
  1087 
  1088 
  1089 *** ML ***
  1090 
  1091 * Antiquotation @{keyword "name"} produces a parser for outer syntax
  1092 from a minor keyword introduced via theory header declaration.
  1093 
  1094 * Antiquotation @{command_spec "name"} produces the
  1095 Outer_Syntax.command_spec from a major keyword introduced via theory
  1096 header declaration; it can be passed to Outer_Syntax.command etc.
  1097 
  1098 * Local_Theory.define no longer hard-wires default theorem name
  1099 "foo_def", but retains the binding as given.  If that is Binding.empty
  1100 / Attrib.empty_binding, the result is not registered as user-level
  1101 fact.  The Local_Theory.define_internal variant allows to specify a
  1102 non-empty name (used for the foundation in the background theory),
  1103 while omitting the fact binding in the user-context.  Potential
  1104 INCOMPATIBILITY for derived definitional packages: need to specify
  1105 naming policy for primitive definitions more explicitly.
  1106 
  1107 * Renamed Thm.capply to Thm.apply, and Thm.cabs to Thm.lambda in
  1108 conformance with similar operations in structure Term and Logic.
  1109 
  1110 * Antiquotation @{attributes [...]} embeds attribute source
  1111 representation into the ML text, which is particularly useful with
  1112 declarations like Local_Theory.note.
  1113 
  1114 * Structure Proof_Context follows standard naming scheme.  Old
  1115 ProofContext has been discontinued.  INCOMPATIBILITY.
  1116 
  1117 * Refined Local_Theory.declaration {syntax, pervasive}, with subtle
  1118 change of semantics: update is applied to auxiliary local theory
  1119 context as well.
  1120 
  1121 * Modernized some old-style infix operations:
  1122 
  1123   addeqcongs    ~> Simplifier.add_eqcong
  1124   deleqcongs    ~> Simplifier.del_eqcong
  1125   addcongs      ~> Simplifier.add_cong
  1126   delcongs      ~> Simplifier.del_cong
  1127   setmksimps    ~> Simplifier.set_mksimps
  1128   setmkcong     ~> Simplifier.set_mkcong
  1129   setmksym      ~> Simplifier.set_mksym
  1130   setmkeqTrue   ~> Simplifier.set_mkeqTrue
  1131   settermless   ~> Simplifier.set_termless
  1132   setsubgoaler  ~> Simplifier.set_subgoaler
  1133   addsplits     ~> Splitter.add_split
  1134   delsplits     ~> Splitter.del_split
  1135 
  1136 
  1137 *** System ***
  1138 
  1139 * USER_HOME settings variable points to cross-platform user home
  1140 directory, which coincides with HOME on POSIX systems only.  Likewise,
  1141 the Isabelle path specification "~" now expands to $USER_HOME, instead
  1142 of former $HOME.  A different default for USER_HOME may be set
  1143 explicitly in shell environment, before Isabelle settings are
  1144 evaluated.  Minor INCOMPATIBILITY: need to adapt Isabelle path where
  1145 the generic user home was intended.
  1146 
  1147 * ISABELLE_HOME_WINDOWS refers to ISABELLE_HOME in windows file name
  1148 notation, which is useful for the jEdit file browser, for example.
  1149 
  1150 * ISABELLE_JDK_HOME settings variable points to JDK with javac and jar
  1151 (not just JRE).
  1152 
  1153 
  1154 
  1155 New in Isabelle2011-1 (October 2011)
  1156 ------------------------------------
  1157 
  1158 *** General ***
  1159 
  1160 * Improved Isabelle/jEdit Prover IDE (PIDE), which can be invoked as
  1161 "isabelle jedit" or "ISABELLE_HOME/Isabelle" on the command line.
  1162 
  1163   - Management of multiple theory files directly from the editor
  1164     buffer store -- bypassing the file-system (no requirement to save
  1165     files for checking).
  1166 
  1167   - Markup of formal entities within the text buffer, with semantic
  1168     highlighting, tooltips and hyperlinks to jump to defining source
  1169     positions.
  1170 
  1171   - Improved text rendering, with sub/superscripts in the source
  1172     buffer (including support for copy/paste wrt. output panel, HTML
  1173     theory output and other non-Isabelle text boxes).
  1174 
  1175   - Refined scheduling of proof checking and printing of results,
  1176     based on interactive editor view.  (Note: jEdit folding and
  1177     narrowing allows to restrict buffer perspectives explicitly.)
  1178 
  1179   - Reduced CPU performance requirements, usable on machines with few
  1180     cores.
  1181 
  1182   - Reduced memory requirements due to pruning of unused document
  1183     versions (garbage collection).
  1184 
  1185 See also ~~/src/Tools/jEdit/README.html for further information,
  1186 including some remaining limitations.
  1187 
  1188 * Theory loader: source files are exclusively located via the master
  1189 directory of each theory node (where the .thy file itself resides).
  1190 The global load path (such as src/HOL/Library) has been discontinued.
  1191 Note that the path element ~~ may be used to reference theories in the
  1192 Isabelle home folder -- for instance, "~~/src/HOL/Library/FuncSet".
  1193 INCOMPATIBILITY.
  1194 
  1195 * Theory loader: source files are identified by content via SHA1
  1196 digests.  Discontinued former path/modtime identification and optional
  1197 ISABELLE_FILE_IDENT plugin scripts.
  1198 
  1199 * Parallelization of nested Isar proofs is subject to
  1200 Goal.parallel_proofs_threshold (default 100).  See also isabelle
  1201 usedir option -Q.
  1202 
  1203 * Name space: former unsynchronized references are now proper
  1204 configuration options, with more conventional names:
  1205 
  1206   long_names   ~> names_long
  1207   short_names  ~> names_short
  1208   unique_names ~> names_unique
  1209 
  1210 Minor INCOMPATIBILITY, need to declare options in context like this:
  1211 
  1212   declare [[names_unique = false]]
  1213 
  1214 * Literal facts `prop` may contain dummy patterns, e.g. `_ = _`.  Note
  1215 that the result needs to be unique, which means fact specifications
  1216 may have to be refined after enriching a proof context.
  1217 
  1218 * Attribute "case_names" has been refined: the assumptions in each case
  1219 can be named now by following the case name with [name1 name2 ...].
  1220 
  1221 * Isabelle/Isar reference manual has been updated and extended:
  1222   - "Synopsis" provides a catalog of main Isar language concepts.
  1223   - Formal references in syntax diagrams, via @{rail} antiquotation.
  1224   - Updated material from classic "ref" manual, notably about
  1225     "Classical Reasoner".
  1226 
  1227 
  1228 *** HOL ***
  1229 
  1230 * Class bot and top require underlying partial order rather than
  1231 preorder: uniqueness of bot and top is guaranteed.  INCOMPATIBILITY.
  1232 
  1233 * Class complete_lattice: generalized a couple of lemmas from sets;
  1234 generalized theorems INF_cong and SUP_cong.  New type classes for
  1235 complete boolean algebras and complete linear orders.  Lemmas
  1236 Inf_less_iff, less_Sup_iff, INF_less_iff, less_SUP_iff now reside in
  1237 class complete_linorder.
  1238 
  1239 Changed proposition of lemmas Inf_bool_def, Sup_bool_def, Inf_fun_def,
  1240 Sup_fun_def, Inf_apply, Sup_apply.
  1241 
  1242 Removed redundant lemmas (the right hand side gives hints how to
  1243 replace them for (metis ...), or (simp only: ...) proofs):
  1244 
  1245   Inf_singleton ~> Inf_insert [where A="{}", unfolded Inf_empty inf_top_right]
  1246   Sup_singleton ~> Sup_insert [where A="{}", unfolded Sup_empty sup_bot_right]
  1247   Inf_binary ~> Inf_insert, Inf_empty, and inf_top_right
  1248   Sup_binary ~> Sup_insert, Sup_empty, and sup_bot_right
  1249   Int_eq_Inter ~> Inf_insert, Inf_empty, and inf_top_right
  1250   Un_eq_Union ~> Sup_insert, Sup_empty, and sup_bot_right
  1251   Inter_def ~> INF_def, image_def
  1252   Union_def ~> SUP_def, image_def
  1253   INT_eq ~> INF_def, and image_def
  1254   UN_eq ~> SUP_def, and image_def
  1255   INF_subset ~> INF_superset_mono [OF _ order_refl]
  1256 
  1257 More consistent and comprehensive names:
  1258 
  1259   INTER_eq_Inter_image ~> INF_def
  1260   UNION_eq_Union_image ~> SUP_def
  1261   INFI_def ~> INF_def
  1262   SUPR_def ~> SUP_def
  1263   INF_leI ~> INF_lower
  1264   INF_leI2 ~> INF_lower2
  1265   le_INFI ~> INF_greatest
  1266   le_SUPI ~> SUP_upper
  1267   le_SUPI2 ~> SUP_upper2
  1268   SUP_leI ~> SUP_least
  1269   INFI_bool_eq ~> INF_bool_eq
  1270   SUPR_bool_eq ~> SUP_bool_eq
  1271   INFI_apply ~> INF_apply
  1272   SUPR_apply ~> SUP_apply
  1273   INTER_def ~> INTER_eq
  1274   UNION_def ~> UNION_eq
  1275 
  1276 INCOMPATIBILITY.
  1277 
  1278 * Renamed theory Complete_Lattice to Complete_Lattices.
  1279 INCOMPATIBILITY.
  1280 
  1281 * Theory Complete_Lattices: lemmas Inf_eq_top_iff, INF_eq_top_iff,
  1282 INF_image, Inf_insert, INF_top, Inf_top_conv, INF_top_conv, SUP_bot,
  1283 Sup_bot_conv, SUP_bot_conv, Sup_eq_top_iff, SUP_eq_top_iff, SUP_image,
  1284 Sup_insert are now declared as [simp].  INCOMPATIBILITY.
  1285 
  1286 * Theory Lattice: lemmas compl_inf_bot, compl_le_comp_iff,
  1287 compl_sup_top, inf_idem, inf_left_idem, inf_sup_absorb, sup_idem,
  1288 sup_inf_absob, sup_left_idem are now declared as [simp].  Minor
  1289 INCOMPATIBILITY.
  1290 
  1291 * Added syntactic classes "inf" and "sup" for the respective
  1292 constants.  INCOMPATIBILITY: Changes in the argument order of the
  1293 (mostly internal) locale predicates for some derived classes.
  1294 
  1295 * Theorem collections ball_simps and bex_simps do not contain theorems
  1296 referring to UNION any longer; these have been moved to collection
  1297 UN_ball_bex_simps.  INCOMPATIBILITY.
  1298 
  1299 * Theory Archimedean_Field: floor now is defined as parameter of a
  1300 separate type class floor_ceiling.
  1301 
  1302 * Theory Finite_Set: more coherent development of fold_set locales:
  1303 
  1304     locale fun_left_comm ~> locale comp_fun_commute
  1305     locale fun_left_comm_idem ~> locale comp_fun_idem
  1306 
  1307 Both use point-free characterization; interpretation proofs may need
  1308 adjustment.  INCOMPATIBILITY.
  1309 
  1310 * Theory Limits: Type "'a net" has been renamed to "'a filter", in
  1311 accordance with standard mathematical terminology. INCOMPATIBILITY.
  1312 
  1313 * Theory Complex_Main: The locale interpretations for the
  1314 bounded_linear and bounded_bilinear locales have been removed, in
  1315 order to reduce the number of duplicate lemmas. Users must use the
  1316 original names for distributivity theorems, potential INCOMPATIBILITY.
  1317 
  1318   divide.add ~> add_divide_distrib
  1319   divide.diff ~> diff_divide_distrib
  1320   divide.setsum ~> setsum_divide_distrib
  1321   mult.add_right ~> right_distrib
  1322   mult.diff_right ~> right_diff_distrib
  1323   mult_right.setsum ~> setsum_right_distrib
  1324   mult_left.diff ~> left_diff_distrib
  1325 
  1326 * Theory Complex_Main: Several redundant theorems have been removed or
  1327 replaced by more general versions. INCOMPATIBILITY.
  1328 
  1329   real_diff_def ~> minus_real_def
  1330   real_divide_def ~> divide_real_def
  1331   real_less_def ~> less_le
  1332   real_abs_def ~> abs_real_def
  1333   real_sgn_def ~> sgn_real_def
  1334   real_mult_commute ~> mult_commute
  1335   real_mult_assoc ~> mult_assoc
  1336   real_mult_1 ~> mult_1_left
  1337   real_add_mult_distrib ~> left_distrib
  1338   real_zero_not_eq_one ~> zero_neq_one
  1339   real_mult_inverse_left ~> left_inverse
  1340   INVERSE_ZERO ~> inverse_zero
  1341   real_le_refl ~> order_refl
  1342   real_le_antisym ~> order_antisym
  1343   real_le_trans ~> order_trans
  1344   real_le_linear ~> linear
  1345   real_le_eq_diff ~> le_iff_diff_le_0
  1346   real_add_left_mono ~> add_left_mono
  1347   real_mult_order ~> mult_pos_pos
  1348   real_mult_less_mono2 ~> mult_strict_left_mono
  1349   real_of_int_real_of_nat ~> real_of_int_of_nat_eq
  1350   real_0_le_divide_iff ~> zero_le_divide_iff
  1351   realpow_two_disj ~> power2_eq_iff
  1352   real_squared_diff_one_factored ~> square_diff_one_factored
  1353   realpow_two_diff ~> square_diff_square_factored
  1354   reals_complete2 ~> complete_real
  1355   real_sum_squared_expand ~> power2_sum
  1356   exp_ln_eq ~> ln_unique
  1357   expi_add ~> exp_add
  1358   expi_zero ~> exp_zero
  1359   lemma_DERIV_subst ~> DERIV_cong
  1360   LIMSEQ_Zfun_iff ~> tendsto_Zfun_iff
  1361   LIMSEQ_const ~> tendsto_const
  1362   LIMSEQ_norm ~> tendsto_norm
  1363   LIMSEQ_add ~> tendsto_add
  1364   LIMSEQ_minus ~> tendsto_minus
  1365   LIMSEQ_minus_cancel ~> tendsto_minus_cancel
  1366   LIMSEQ_diff ~> tendsto_diff
  1367   bounded_linear.LIMSEQ ~> bounded_linear.tendsto
  1368   bounded_bilinear.LIMSEQ ~> bounded_bilinear.tendsto
  1369   LIMSEQ_mult ~> tendsto_mult
  1370   LIMSEQ_inverse ~> tendsto_inverse
  1371   LIMSEQ_divide ~> tendsto_divide
  1372   LIMSEQ_pow ~> tendsto_power
  1373   LIMSEQ_setsum ~> tendsto_setsum
  1374   LIMSEQ_setprod ~> tendsto_setprod
  1375   LIMSEQ_norm_zero ~> tendsto_norm_zero_iff
  1376   LIMSEQ_rabs_zero ~> tendsto_rabs_zero_iff
  1377   LIMSEQ_imp_rabs ~> tendsto_rabs
  1378   LIMSEQ_add_minus ~> tendsto_add [OF _ tendsto_minus]
  1379   LIMSEQ_add_const ~> tendsto_add [OF _ tendsto_const]
  1380   LIMSEQ_diff_const ~> tendsto_diff [OF _ tendsto_const]
  1381   LIMSEQ_Complex ~> tendsto_Complex
  1382   LIM_ident ~> tendsto_ident_at
  1383   LIM_const ~> tendsto_const
  1384   LIM_add ~> tendsto_add
  1385   LIM_add_zero ~> tendsto_add_zero
  1386   LIM_minus ~> tendsto_minus
  1387   LIM_diff ~> tendsto_diff
  1388   LIM_norm ~> tendsto_norm
  1389   LIM_norm_zero ~> tendsto_norm_zero
  1390   LIM_norm_zero_cancel ~> tendsto_norm_zero_cancel
  1391   LIM_norm_zero_iff ~> tendsto_norm_zero_iff
  1392   LIM_rabs ~> tendsto_rabs
  1393   LIM_rabs_zero ~> tendsto_rabs_zero
  1394   LIM_rabs_zero_cancel ~> tendsto_rabs_zero_cancel
  1395   LIM_rabs_zero_iff ~> tendsto_rabs_zero_iff
  1396   LIM_compose ~> tendsto_compose
  1397   LIM_mult ~> tendsto_mult
  1398   LIM_scaleR ~> tendsto_scaleR
  1399   LIM_of_real ~> tendsto_of_real
  1400   LIM_power ~> tendsto_power
  1401   LIM_inverse ~> tendsto_inverse
  1402   LIM_sgn ~> tendsto_sgn
  1403   isCont_LIM_compose ~> isCont_tendsto_compose
  1404   bounded_linear.LIM ~> bounded_linear.tendsto
  1405   bounded_linear.LIM_zero ~> bounded_linear.tendsto_zero
  1406   bounded_bilinear.LIM ~> bounded_bilinear.tendsto
  1407   bounded_bilinear.LIM_prod_zero ~> bounded_bilinear.tendsto_zero
  1408   bounded_bilinear.LIM_left_zero ~> bounded_bilinear.tendsto_left_zero
  1409   bounded_bilinear.LIM_right_zero ~> bounded_bilinear.tendsto_right_zero
  1410   LIM_inverse_fun ~> tendsto_inverse [OF tendsto_ident_at]
  1411 
  1412 * Theory Complex_Main: The definition of infinite series was
  1413 generalized.  Now it is defined on the type class {topological_space,
  1414 comm_monoid_add}.  Hence it is useable also for extended real numbers.
  1415 
  1416 * Theory Complex_Main: The complex exponential function "expi" is now
  1417 a type-constrained abbreviation for "exp :: complex => complex"; thus
  1418 several polymorphic lemmas about "exp" are now applicable to "expi".
  1419 
  1420 * Code generation:
  1421 
  1422   - Theory Library/Code_Char_ord provides native ordering of
  1423     characters in the target language.
  1424 
  1425   - Commands code_module and code_library are legacy, use export_code
  1426     instead.
  1427 
  1428   - Method "evaluation" is legacy, use method "eval" instead.
  1429 
  1430   - Legacy evaluator "SML" is deactivated by default.  May be
  1431     reactivated by the following theory command:
  1432 
  1433       setup {* Value.add_evaluator ("SML", Codegen.eval_term) *}
  1434 
  1435 * Declare ext [intro] by default.  Rare INCOMPATIBILITY.
  1436 
  1437 * New proof method "induction" that gives induction hypotheses the
  1438 name "IH", thus distinguishing them from further hypotheses that come
  1439 from rule induction.  The latter are still called "hyps".  Method
  1440 "induction" is a thin wrapper around "induct" and follows the same
  1441 syntax.
  1442 
  1443 * Method "fastsimp" has been renamed to "fastforce", but "fastsimp" is
  1444 still available as a legacy feature for some time.
  1445 
  1446 * Nitpick:
  1447   - Added "need" and "total_consts" options.
  1448   - Reintroduced "show_skolems" option by popular demand.
  1449   - Renamed attribute: nitpick_def ~> nitpick_unfold.
  1450     INCOMPATIBILITY.
  1451 
  1452 * Sledgehammer:
  1453   - Use quasi-sound (and efficient) translations by default.
  1454   - Added support for the following provers: E-ToFoF, LEO-II,
  1455     Satallax, SNARK, Waldmeister, and Z3 with TPTP syntax.
  1456   - Automatically preplay and minimize proofs before showing them if
  1457     this can be done within reasonable time.
  1458   - sledgehammer available_provers ~> sledgehammer supported_provers.
  1459     INCOMPATIBILITY.
  1460   - Added "preplay_timeout", "slicing", "type_enc", "sound",
  1461     "max_mono_iters", and "max_new_mono_instances" options.
  1462   - Removed "explicit_apply" and "full_types" options as well as "Full
  1463     Types" Proof General menu item. INCOMPATIBILITY.
  1464 
  1465 * Metis:
  1466   - Removed "metisF" -- use "metis" instead. INCOMPATIBILITY.
  1467   - Obsoleted "metisFT" -- use "metis (full_types)" instead.
  1468     INCOMPATIBILITY.
  1469 
  1470 * Command 'try':
  1471   - Renamed 'try_methods' and added "simp:", "intro:", "dest:", and
  1472     "elim:" options. INCOMPATIBILITY.
  1473   - Introduced 'try' that not only runs 'try_methods' but also
  1474     'solve_direct', 'sledgehammer', 'quickcheck', and 'nitpick'.
  1475 
  1476 * Quickcheck:
  1477   - Added "eval" option to evaluate terms for the found counterexample
  1478     (currently only supported by the default (exhaustive) tester).
  1479   - Added post-processing of terms to obtain readable counterexamples
  1480     (currently only supported by the default (exhaustive) tester).
  1481   - New counterexample generator quickcheck[narrowing] enables
  1482     narrowing-based testing.  Requires the Glasgow Haskell compiler
  1483     with its installation location defined in the Isabelle settings
  1484     environment as ISABELLE_GHC.
  1485   - Removed quickcheck tester "SML" based on the SML code generator
  1486     (formly in HOL/Library).
  1487 
  1488 * Function package: discontinued option "tailrec".  INCOMPATIBILITY,
  1489 use 'partial_function' instead.
  1490 
  1491 * Theory Library/Extended_Reals replaces now the positive extended
  1492 reals found in probability theory. This file is extended by
  1493 Multivariate_Analysis/Extended_Real_Limits.
  1494 
  1495 * Theory Library/Old_Recdef: old 'recdef' package has been moved here,
  1496 from where it must be imported explicitly if it is really required.
  1497 INCOMPATIBILITY.
  1498 
  1499 * Theory Library/Wfrec: well-founded recursion combinator "wfrec" has
  1500 been moved here.  INCOMPATIBILITY.
  1501 
  1502 * Theory Library/Saturated provides type of numbers with saturated
  1503 arithmetic.
  1504 
  1505 * Theory Library/Product_Lattice defines a pointwise ordering for the
  1506 product type 'a * 'b, and provides instance proofs for various order
  1507 and lattice type classes.
  1508 
  1509 * Theory Library/Countable now provides the "countable_datatype" proof
  1510 method for proving "countable" class instances for datatypes.
  1511 
  1512 * Theory Library/Cset_Monad allows do notation for computable sets
  1513 (cset) via the generic monad ad-hoc overloading facility.
  1514 
  1515 * Library: Theories of common data structures are split into theories
  1516 for implementation, an invariant-ensuring type, and connection to an
  1517 abstract type. INCOMPATIBILITY.
  1518 
  1519   - RBT is split into RBT and RBT_Mapping.
  1520   - AssocList is split and renamed into AList and AList_Mapping.
  1521   - DList is split into DList_Impl, DList, and DList_Cset.
  1522   - Cset is split into Cset and List_Cset.
  1523 
  1524 * Theory Library/Nat_Infinity has been renamed to
  1525 Library/Extended_Nat, with name changes of the following types and
  1526 constants:
  1527 
  1528   type inat   ~> type enat
  1529   Fin         ~> enat
  1530   Infty       ~> infinity (overloaded)
  1531   iSuc        ~> eSuc
  1532   the_Fin     ~> the_enat
  1533 
  1534 Every theorem name containing "inat", "Fin", "Infty", or "iSuc" has
  1535 been renamed accordingly. INCOMPATIBILITY.
  1536 
  1537 * Session Multivariate_Analysis: The euclidean_space type class now
  1538 fixes a constant "Basis :: 'a set" consisting of the standard
  1539 orthonormal basis for the type. Users now have the option of
  1540 quantifying over this set instead of using the "basis" function, e.g.
  1541 "ALL x:Basis. P x" vs "ALL i<DIM('a). P (basis i)".
  1542 
  1543 * Session Multivariate_Analysis: Type "('a, 'b) cart" has been renamed
  1544 to "('a, 'b) vec" (the syntax "'a ^ 'b" remains unaffected). Constants
  1545 "Cart_nth" and "Cart_lambda" have been respectively renamed to
  1546 "vec_nth" and "vec_lambda"; theorems mentioning those names have
  1547 changed to match. Definition theorems for overloaded constants now use
  1548 the standard "foo_vec_def" naming scheme. A few other theorems have
  1549 been renamed as follows (INCOMPATIBILITY):
  1550 
  1551   Cart_eq          ~> vec_eq_iff
  1552   dist_nth_le_cart ~> dist_vec_nth_le
  1553   tendsto_vector   ~> vec_tendstoI
  1554   Cauchy_vector    ~> vec_CauchyI
  1555 
  1556 * Session Multivariate_Analysis: Several duplicate theorems have been
  1557 removed, and other theorems have been renamed or replaced with more
  1558 general versions. INCOMPATIBILITY.
  1559 
  1560   finite_choice ~> finite_set_choice
  1561   eventually_conjI ~> eventually_conj
  1562   eventually_and ~> eventually_conj_iff
  1563   eventually_false ~> eventually_False
  1564   setsum_norm ~> norm_setsum
  1565   Lim_sequentially ~> LIMSEQ_def
  1566   Lim_ident_at ~> LIM_ident
  1567   Lim_const ~> tendsto_const
  1568   Lim_cmul ~> tendsto_scaleR [OF tendsto_const]
  1569   Lim_neg ~> tendsto_minus
  1570   Lim_add ~> tendsto_add
  1571   Lim_sub ~> tendsto_diff
  1572   Lim_mul ~> tendsto_scaleR
  1573   Lim_vmul ~> tendsto_scaleR [OF _ tendsto_const]
  1574   Lim_null_norm ~> tendsto_norm_zero_iff [symmetric]
  1575   Lim_linear ~> bounded_linear.tendsto
  1576   Lim_component ~> tendsto_euclidean_component
  1577   Lim_component_cart ~> tendsto_vec_nth
  1578   Lim_inner ~> tendsto_inner [OF tendsto_const]
  1579   dot_lsum ~> inner_setsum_left
  1580   dot_rsum ~> inner_setsum_right
  1581   continuous_cmul ~> continuous_scaleR [OF continuous_const]
  1582   continuous_neg ~> continuous_minus
  1583   continuous_sub ~> continuous_diff
  1584   continuous_vmul ~> continuous_scaleR [OF _ continuous_const]
  1585   continuous_mul ~> continuous_scaleR
  1586   continuous_inv ~> continuous_inverse
  1587   continuous_at_within_inv ~> continuous_at_within_inverse
  1588   continuous_at_inv ~> continuous_at_inverse
  1589   continuous_at_norm ~> continuous_norm [OF continuous_at_id]
  1590   continuous_at_infnorm ~> continuous_infnorm [OF continuous_at_id]
  1591   continuous_at_component ~> continuous_component [OF continuous_at_id]
  1592   continuous_on_neg ~> continuous_on_minus
  1593   continuous_on_sub ~> continuous_on_diff
  1594   continuous_on_cmul ~> continuous_on_scaleR [OF continuous_on_const]
  1595   continuous_on_vmul ~> continuous_on_scaleR [OF _ continuous_on_const]
  1596   continuous_on_mul ~> continuous_on_scaleR
  1597   continuous_on_mul_real ~> continuous_on_mult
  1598   continuous_on_inner ~> continuous_on_inner [OF continuous_on_const]
  1599   continuous_on_norm ~> continuous_on_norm [OF continuous_on_id]
  1600   continuous_on_inverse ~> continuous_on_inv
  1601   uniformly_continuous_on_neg ~> uniformly_continuous_on_minus
  1602   uniformly_continuous_on_sub ~> uniformly_continuous_on_diff
  1603   subset_interior ~> interior_mono
  1604   subset_closure ~> closure_mono
  1605   closure_univ ~> closure_UNIV
  1606   real_arch_lt ~> reals_Archimedean2
  1607   real_arch ~> reals_Archimedean3
  1608   real_abs_norm ~> abs_norm_cancel
  1609   real_abs_sub_norm ~> norm_triangle_ineq3
  1610   norm_cauchy_schwarz_abs ~> Cauchy_Schwarz_ineq2
  1611 
  1612 * Session HOL-Probability:
  1613   - Caratheodory's extension lemma is now proved for ring_of_sets.
  1614   - Infinite products of probability measures are now available.
  1615   - Sigma closure is independent, if the generator is independent
  1616   - Use extended reals instead of positive extended
  1617     reals. INCOMPATIBILITY.
  1618 
  1619 * Session HOLCF: Discontinued legacy theorem names, INCOMPATIBILITY.
  1620 
  1621   expand_fun_below ~> fun_below_iff
  1622   below_fun_ext ~> fun_belowI
  1623   expand_cfun_eq ~> cfun_eq_iff
  1624   ext_cfun ~> cfun_eqI
  1625   expand_cfun_below ~> cfun_below_iff
  1626   below_cfun_ext ~> cfun_belowI
  1627   monofun_fun_fun ~> fun_belowD
  1628   monofun_fun_arg ~> monofunE
  1629   monofun_lub_fun ~> adm_monofun [THEN admD]
  1630   cont_lub_fun ~> adm_cont [THEN admD]
  1631   cont2cont_Rep_CFun ~> cont2cont_APP
  1632   cont_Rep_CFun_app ~> cont_APP_app
  1633   cont_Rep_CFun_app_app ~> cont_APP_app_app
  1634   cont_cfun_fun ~> cont_Rep_cfun1 [THEN contE]
  1635   cont_cfun_arg ~> cont_Rep_cfun2 [THEN contE]
  1636   contlub_cfun ~> lub_APP [symmetric]
  1637   contlub_LAM ~> lub_LAM [symmetric]
  1638   thelubI ~> lub_eqI
  1639   UU_I ~> bottomI
  1640   lift_distinct1 ~> lift.distinct(1)
  1641   lift_distinct2 ~> lift.distinct(2)
  1642   Def_not_UU ~> lift.distinct(2)
  1643   Def_inject ~> lift.inject
  1644   below_UU_iff ~> below_bottom_iff
  1645   eq_UU_iff ~> eq_bottom_iff
  1646 
  1647 
  1648 *** Document preparation ***
  1649 
  1650 * Antiquotation @{rail} layouts railroad syntax diagrams, see also
  1651 isar-ref manual, both for description and actual application of the
  1652 same.
  1653 
  1654 * Antiquotation @{value} evaluates the given term and presents its
  1655 result.
  1656 
  1657 * Antiquotations: term style "isub" provides ad-hoc conversion of
  1658 variables x1, y23 into subscripted form x\<^isub>1,
  1659 y\<^isub>2\<^isub>3.
  1660 
  1661 * Predefined LaTeX macros for Isabelle symbols \<bind> and \<then>
  1662 (e.g. see ~~/src/HOL/Library/Monad_Syntax.thy).
  1663 
  1664 * Localized \isabellestyle switch can be used within blocks or groups
  1665 like this:
  1666 
  1667   \isabellestyle{it}  %preferred default
  1668   {\isabellestylett @{text "typewriter stuff"}}
  1669 
  1670 * Discontinued special treatment of hard tabulators.  Implicit
  1671 tab-width is now defined as 1.  Potential INCOMPATIBILITY for visual
  1672 layouts.
  1673 
  1674 
  1675 *** ML ***
  1676 
  1677 * The inner syntax of sort/type/term/prop supports inlined YXML
  1678 representations within quoted string tokens.  By encoding logical
  1679 entities via Term_XML (in ML or Scala) concrete syntax can be
  1680 bypassed, which is particularly useful for producing bits of text
  1681 under external program control.
  1682 
  1683 * Antiquotations for ML and document preparation are managed as theory
  1684 data, which requires explicit setup.
  1685 
  1686 * Isabelle_Process.is_active allows tools to check if the official
  1687 process wrapper is running (Isabelle/Scala/jEdit) or the old TTY loop
  1688 (better known as Proof General).
  1689 
  1690 * Structure Proof_Context follows standard naming scheme.  Old
  1691 ProofContext is still available for some time as legacy alias.
  1692 
  1693 * Structure Timing provides various operations for timing; supersedes
  1694 former start_timing/end_timing etc.
  1695 
  1696 * Path.print is the official way to show file-system paths to users
  1697 (including quotes etc.).
  1698 
  1699 * Inner syntax: identifiers in parse trees of generic categories
  1700 "logic", "aprop", "idt" etc. carry position information (disguised as
  1701 type constraints).  Occasional INCOMPATIBILITY with non-compliant
  1702 translations that choke on unexpected type constraints.  Positions can
  1703 be stripped in ML translations via Syntax.strip_positions /
  1704 Syntax.strip_positions_ast, or via the syntax constant
  1705 "_strip_positions" within parse trees.  As last resort, positions can
  1706 be disabled via the configuration option Syntax.positions, which is
  1707 called "syntax_positions" in Isar attribute syntax.
  1708 
  1709 * Discontinued special status of various ML structures that contribute
  1710 to structure Syntax (Ast, Lexicon, Mixfix, Parser, Printer etc.): less
  1711 pervasive content, no inclusion in structure Syntax.  INCOMPATIBILITY,
  1712 refer directly to Ast.Constant, Lexicon.is_identifier,
  1713 Syntax_Trans.mk_binder_tr etc.
  1714 
  1715 * Typed print translation: discontinued show_sorts argument, which is
  1716 already available via context of "advanced" translation.
  1717 
  1718 * Refined PARALLEL_GOALS tactical: degrades gracefully for schematic
  1719 goal states; body tactic needs to address all subgoals uniformly.
  1720 
  1721 * Slightly more special eq_list/eq_set, with shortcut involving
  1722 pointer equality (assumes that eq relation is reflexive).
  1723 
  1724 * Classical tactics use proper Proof.context instead of historic types
  1725 claset/clasimpset.  Old-style declarations like addIs, addEs, addDs
  1726 operate directly on Proof.context.  Raw type claset retains its use as
  1727 snapshot of the classical context, which can be recovered via
  1728 (put_claset HOL_cs) etc.  Type clasimpset has been discontinued.
  1729 INCOMPATIBILITY, classical tactics and derived proof methods require
  1730 proper Proof.context.
  1731 
  1732 
  1733 *** System ***
  1734 
  1735 * Discontinued support for Poly/ML 5.2, which was the last version
  1736 without proper multithreading and TimeLimit implementation.
  1737 
  1738 * Discontinued old lib/scripts/polyml-platform, which has been
  1739 obsolete since Isabelle2009-2.
  1740 
  1741 * Various optional external tools are referenced more robustly and
  1742 uniformly by explicit Isabelle settings as follows:
  1743 
  1744   ISABELLE_CSDP   (formerly CSDP_EXE)
  1745   ISABELLE_GHC    (formerly EXEC_GHC or GHC_PATH)
  1746   ISABELLE_OCAML  (formerly EXEC_OCAML)
  1747   ISABELLE_SWIPL  (formerly EXEC_SWIPL)
  1748   ISABELLE_YAP    (formerly EXEC_YAP)
  1749 
  1750 Note that automated detection from the file-system or search path has
  1751 been discontinued.  INCOMPATIBILITY.
  1752 
  1753 * Scala layer provides JVM method invocation service for static
  1754 methods of type (String)String, see Invoke_Scala.method in ML.  For
  1755 example:
  1756 
  1757   Invoke_Scala.method "java.lang.System.getProperty" "java.home"
  1758 
  1759 Together with YXML.string_of_body/parse_body and XML.Encode/Decode
  1760 this allows to pass structured values between ML and Scala.
  1761 
  1762 * The IsabelleText fonts includes some further glyphs to support the
  1763 Prover IDE.  Potential INCOMPATIBILITY: users who happen to have
  1764 installed a local copy (which is normally *not* required) need to
  1765 delete or update it from ~~/lib/fonts/.
  1766 
  1767 
  1768 
  1769 New in Isabelle2011 (January 2011)
  1770 ----------------------------------
  1771 
  1772 *** General ***
  1773 
  1774 * Experimental Prover IDE based on Isabelle/Scala and jEdit (see
  1775 src/Tools/jEdit).  This also serves as IDE for Isabelle/ML, with
  1776 useful tooltips and hyperlinks produced from its static analysis.  The
  1777 bundled component provides an executable Isabelle tool that can be run
  1778 like this:
  1779 
  1780   Isabelle2011/bin/isabelle jedit
  1781 
  1782 * Significantly improved Isabelle/Isar implementation manual.
  1783 
  1784 * System settings: ISABELLE_HOME_USER now includes ISABELLE_IDENTIFIER
  1785 (and thus refers to something like $HOME/.isabelle/Isabelle2011),
  1786 while the default heap location within that directory lacks that extra
  1787 suffix.  This isolates multiple Isabelle installations from each
  1788 other, avoiding problems with old settings in new versions.
  1789 INCOMPATIBILITY, need to copy/upgrade old user settings manually.
  1790 
  1791 * Source files are always encoded as UTF-8, instead of old-fashioned
  1792 ISO-Latin-1.  INCOMPATIBILITY.  Isabelle LaTeX documents might require
  1793 the following package declarations:
  1794 
  1795   \usepackage[utf8]{inputenc}
  1796   \usepackage{textcomp}
  1797 
  1798 * Explicit treatment of UTF-8 sequences as Isabelle symbols, such that
  1799 a Unicode character is treated as a single symbol, not a sequence of
  1800 non-ASCII bytes as before.  Since Isabelle/ML string literals may
  1801 contain symbols without further backslash escapes, Unicode can now be
  1802 used here as well.  Recall that Symbol.explode in ML provides a
  1803 consistent view on symbols, while raw explode (or String.explode)
  1804 merely give a byte-oriented representation.
  1805 
  1806 * Theory loader: source files are primarily located via the master
  1807 directory of each theory node (where the .thy file itself resides).
  1808 The global load path is still partially available as legacy feature.
  1809 Minor INCOMPATIBILITY due to subtle change in file lookup: use
  1810 explicit paths, relatively to the theory.
  1811 
  1812 * Special treatment of ML file names has been discontinued.
  1813 Historically, optional extensions .ML or .sml were added on demand --
  1814 at the cost of clarity of file dependencies.  Recall that Isabelle/ML
  1815 files exclusively use the .ML extension.  Minor INCOMPATIBILTY.
  1816 
  1817 * Various options that affect pretty printing etc. are now properly
  1818 handled within the context via configuration options, instead of
  1819 unsynchronized references or print modes.  There are both ML Config.T
  1820 entities and Isar declaration attributes to access these.
  1821 
  1822   ML (Config.T)                 Isar (attribute)
  1823 
  1824   eta_contract                  eta_contract
  1825   show_brackets                 show_brackets
  1826   show_sorts                    show_sorts
  1827   show_types                    show_types
  1828   show_question_marks           show_question_marks
  1829   show_consts                   show_consts
  1830   show_abbrevs                  show_abbrevs
  1831 
  1832   Syntax.ast_trace              syntax_ast_trace
  1833   Syntax.ast_stat               syntax_ast_stat
  1834   Syntax.ambiguity_level        syntax_ambiguity_level
  1835 
  1836   Goal_Display.goals_limit      goals_limit
  1837   Goal_Display.show_main_goal   show_main_goal
  1838 
  1839   Method.rule_trace             rule_trace
  1840 
  1841   Thy_Output.display            thy_output_display
  1842   Thy_Output.quotes             thy_output_quotes
  1843   Thy_Output.indent             thy_output_indent
  1844   Thy_Output.source             thy_output_source
  1845   Thy_Output.break              thy_output_break
  1846 
  1847 Note that corresponding "..._default" references in ML may only be
  1848 changed globally at the ROOT session setup, but *not* within a theory.
  1849 The option "show_abbrevs" supersedes the former print mode
  1850 "no_abbrevs" with inverted meaning.
  1851 
  1852 * More systematic naming of some configuration options.
  1853 INCOMPATIBILITY.
  1854 
  1855   trace_simp  ~>  simp_trace
  1856   debug_simp  ~>  simp_debug
  1857 
  1858 * Support for real valued configuration options, using simplistic
  1859 floating-point notation that coincides with the inner syntax for
  1860 float_token.
  1861 
  1862 * Support for real valued preferences (with approximative PGIP type):
  1863 front-ends need to accept "pgint" values in float notation.
  1864 INCOMPATIBILITY.
  1865 
  1866 * The IsabelleText font now includes Cyrillic, Hebrew, Arabic from
  1867 DejaVu Sans.
  1868 
  1869 * Discontinued support for Poly/ML 5.0 and 5.1 versions.
  1870 
  1871 
  1872 *** Pure ***
  1873 
  1874 * Command 'type_synonym' (with single argument) replaces somewhat
  1875 outdated 'types', which is still available as legacy feature for some
  1876 time.
  1877 
  1878 * Command 'nonterminal' (with 'and' separated list of arguments)
  1879 replaces somewhat outdated 'nonterminals'.  INCOMPATIBILITY.
  1880 
  1881 * Command 'notepad' replaces former 'example_proof' for
  1882 experimentation in Isar without any result.  INCOMPATIBILITY.
  1883 
  1884 * Locale interpretation commands 'interpret' and 'sublocale' accept
  1885 lists of equations to map definitions in a locale to appropriate
  1886 entities in the context of the interpretation.  The 'interpretation'
  1887 command already provided this functionality.
  1888 
  1889 * Diagnostic command 'print_dependencies' prints the locale instances
  1890 that would be activated if the specified expression was interpreted in
  1891 the current context.  Variant "print_dependencies!" assumes a context
  1892 without interpretations.
  1893 
  1894 * Diagnostic command 'print_interps' prints interpretations in proofs
  1895 in addition to interpretations in theories.
  1896 
  1897 * Discontinued obsolete 'global' and 'local' commands to manipulate
  1898 the theory name space.  Rare INCOMPATIBILITY.  The ML functions
  1899 Sign.root_path and Sign.local_path may be applied directly where this
  1900 feature is still required for historical reasons.
  1901 
  1902 * Discontinued obsolete 'constdefs' command.  INCOMPATIBILITY, use
  1903 'definition' instead.
  1904 
  1905 * The "prems" fact, which refers to the accidental collection of
  1906 foundational premises in the context, is now explicitly marked as
  1907 legacy feature and will be discontinued soon.  Consider using "assms"
  1908 of the head statement or reference facts by explicit names.
  1909 
  1910 * Document antiquotations @{class} and @{type} print classes and type
  1911 constructors.
  1912 
  1913 * Document antiquotation @{file} checks file/directory entries within
  1914 the local file system.
  1915 
  1916 
  1917 *** HOL ***
  1918 
  1919 * Coercive subtyping: functions can be declared as coercions and type
  1920 inference will add them as necessary upon input of a term.  Theory
  1921 Complex_Main declares real :: nat => real and real :: int => real as
  1922 coercions. A coercion function f is declared like this:
  1923 
  1924   declare [[coercion f]]
  1925 
  1926 To lift coercions through type constructors (e.g. from nat => real to
  1927 nat list => real list), map functions can be declared, e.g.
  1928 
  1929   declare [[coercion_map map]]
  1930 
  1931 Currently coercion inference is activated only in theories including
  1932 real numbers, i.e. descendants of Complex_Main.  This is controlled by
  1933 the configuration option "coercion_enabled", e.g. it can be enabled in
  1934 other theories like this:
  1935 
  1936   declare [[coercion_enabled]]
  1937 
  1938 * Command 'partial_function' provides basic support for recursive
  1939 function definitions over complete partial orders.  Concrete instances
  1940 are provided for i) the option type, ii) tail recursion on arbitrary
  1941 types, and iii) the heap monad of Imperative_HOL.  See
  1942 src/HOL/ex/Fundefs.thy and src/HOL/Imperative_HOL/ex/Linked_Lists.thy
  1943 for examples.
  1944 
  1945 * Function package: f.psimps rules are no longer implicitly declared
  1946 as [simp].  INCOMPATIBILITY.
  1947 
  1948 * Datatype package: theorems generated for executable equality (class
  1949 "eq") carry proper names and are treated as default code equations.
  1950 
  1951 * Inductive package: now offers command 'inductive_simps' to
  1952 automatically derive instantiated and simplified equations for
  1953 inductive predicates, similar to 'inductive_cases'.
  1954 
  1955 * Command 'enriched_type' allows to register properties of the
  1956 functorial structure of types.
  1957 
  1958 * Improved infrastructure for term evaluation using code generator
  1959 techniques, in particular static evaluation conversions.
  1960 
  1961 * Code generator: Scala (2.8 or higher) has been added to the target
  1962 languages.
  1963 
  1964 * Code generator: globbing constant expressions "*" and "Theory.*"
  1965 have been replaced by the more idiomatic "_" and "Theory._".
  1966 INCOMPATIBILITY.
  1967 
  1968 * Code generator: export_code without explicit file declaration prints
  1969 to standard output.  INCOMPATIBILITY.
  1970 
  1971 * Code generator: do not print function definitions for case
  1972 combinators any longer.
  1973 
  1974 * Code generator: simplification with rules determined with
  1975 src/Tools/Code/code_simp.ML and method "code_simp".
  1976 
  1977 * Code generator for records: more idiomatic representation of record
  1978 types.  Warning: records are not covered by ancient SML code
  1979 generation any longer.  INCOMPATIBILITY.  In cases of need, a suitable
  1980 rep_datatype declaration helps to succeed then:
  1981 
  1982   record 'a foo = ...
  1983   ...
  1984   rep_datatype foo_ext ...
  1985 
  1986 * Records: logical foundation type for records does not carry a
  1987 '_type' suffix any longer (obsolete due to authentic syntax).
  1988 INCOMPATIBILITY.
  1989 
  1990 * Quickcheck now by default uses exhaustive testing instead of random
  1991 testing.  Random testing can be invoked by "quickcheck [random]",
  1992 exhaustive testing by "quickcheck [exhaustive]".
  1993 
  1994 * Quickcheck instantiates polymorphic types with small finite
  1995 datatypes by default. This enables a simple execution mechanism to
  1996 handle quantifiers and function equality over the finite datatypes.
  1997 
  1998 * Quickcheck random generator has been renamed from "code" to
  1999 "random".  INCOMPATIBILITY.
  2000 
  2001 * Quickcheck now has a configurable time limit which is set to 30
  2002 seconds by default. This can be changed by adding [timeout = n] to the
  2003 quickcheck command. The time limit for Auto Quickcheck is still set
  2004 independently.
  2005 
  2006 * Quickcheck in locales considers interpretations of that locale for
  2007 counter example search.
  2008 
  2009 * Sledgehammer:
  2010   - Added "smt" and "remote_smt" provers based on the "smt" proof
  2011     method. See the Sledgehammer manual for details ("isabelle doc
  2012     sledgehammer").
  2013   - Renamed commands:
  2014     sledgehammer atp_info ~> sledgehammer running_provers
  2015     sledgehammer atp_kill ~> sledgehammer kill_provers
  2016     sledgehammer available_atps ~> sledgehammer available_provers
  2017     INCOMPATIBILITY.
  2018   - Renamed options:
  2019     sledgehammer [atps = ...] ~> sledgehammer [provers = ...]
  2020     sledgehammer [atp = ...] ~> sledgehammer [prover = ...]
  2021     sledgehammer [timeout = 77 s] ~> sledgehammer [timeout = 77]
  2022     (and "ms" and "min" are no longer supported)
  2023     INCOMPATIBILITY.
  2024 
  2025 * Nitpick:
  2026   - Renamed options:
  2027     nitpick [timeout = 77 s] ~> nitpick [timeout = 77]
  2028     nitpick [tac_timeout = 777 ms] ~> nitpick [tac_timeout = 0.777]
  2029     INCOMPATIBILITY.
  2030   - Added support for partial quotient types.
  2031   - Added local versions of the "Nitpick.register_xxx" functions.
  2032   - Added "whack" option.
  2033   - Allow registration of quotient types as codatatypes.
  2034   - Improved "merge_type_vars" option to merge more types.
  2035   - Removed unsound "fast_descrs" option.
  2036   - Added custom symmetry breaking for datatypes, making it possible to reach
  2037     higher cardinalities.
  2038   - Prevent the expansion of too large definitions.
  2039 
  2040 * Proof methods "metis" and "meson" now have configuration options
  2041 "meson_trace", "metis_trace", and "metis_verbose" that can be enabled
  2042 to diagnose these tools. E.g.
  2043 
  2044     using [[metis_trace = true]]
  2045 
  2046 * Auto Solve: Renamed "Auto Solve Direct".  The tool is now available
  2047 manually as command 'solve_direct'.
  2048 
  2049 * The default SMT solver Z3 must be enabled explicitly (due to
  2050 licensing issues) by setting the environment variable
  2051 Z3_NON_COMMERCIAL in etc/settings of the component, for example.  For
  2052 commercial applications, the SMT solver CVC3 is provided as fall-back;
  2053 changing the SMT solver is done via the configuration option
  2054 "smt_solver".
  2055 
  2056 * Remote SMT solvers need to be referred to by the "remote_" prefix,
  2057 i.e. "remote_cvc3" and "remote_z3".
  2058 
  2059 * Added basic SMT support for datatypes, records, and typedefs using
  2060 the oracle mode (no proofs).  Direct support of pairs has been dropped
  2061 in exchange (pass theorems fst_conv snd_conv pair_collapse to the SMT
  2062 support for a similar behavior).  Minor INCOMPATIBILITY.
  2063 
  2064 * Changed SMT configuration options:
  2065   - Renamed:
  2066     z3_proofs ~> smt_oracle (with inverted meaning)
  2067     z3_trace_assms ~> smt_trace_used_facts
  2068     INCOMPATIBILITY.
  2069   - Added:
  2070     smt_verbose
  2071     smt_random_seed
  2072     smt_datatypes
  2073     smt_infer_triggers
  2074     smt_monomorph_limit
  2075     cvc3_options
  2076     remote_cvc3_options
  2077     remote_z3_options
  2078     yices_options
  2079 
  2080 * Boogie output files (.b2i files) need to be declared in the theory
  2081 header.
  2082 
  2083 * Simplification procedure "list_to_set_comprehension" rewrites list
  2084 comprehensions applied to List.set to set comprehensions.  Occasional
  2085 INCOMPATIBILITY, may be deactivated like this:
  2086 
  2087   declare [[simproc del: list_to_set_comprehension]]
  2088 
  2089 * Removed old version of primrec package.  INCOMPATIBILITY.
  2090 
  2091 * Removed simplifier congruence rule of "prod_case", as has for long
  2092 been the case with "split".  INCOMPATIBILITY.
  2093 
  2094 * String.literal is a type, but not a datatype.  INCOMPATIBILITY.
  2095 
  2096 * Removed [split_format ... and ... and ...] version of
  2097 [split_format].  Potential INCOMPATIBILITY.
  2098 
  2099 * Predicate "sorted" now defined inductively, with nice induction
  2100 rules.  INCOMPATIBILITY: former sorted.simps now named sorted_simps.
  2101 
  2102 * Constant "contents" renamed to "the_elem", to free the generic name
  2103 contents for other uses.  INCOMPATIBILITY.
  2104 
  2105 * Renamed class eq and constant eq (for code generation) to class
  2106 equal and constant equal, plus renaming of related facts and various
  2107 tuning.  INCOMPATIBILITY.
  2108 
  2109 * Dropped type classes mult_mono and mult_mono1.  INCOMPATIBILITY.
  2110 
  2111 * Removed output syntax "'a ~=> 'b" for "'a => 'b option".
  2112 INCOMPATIBILITY.
  2113 
  2114 * Renamed theory Fset to Cset, type Fset.fset to Cset.set, in order to
  2115 avoid confusion with finite sets.  INCOMPATIBILITY.
  2116 
  2117 * Abandoned locales equiv, congruent and congruent2 for equivalence
  2118 relations.  INCOMPATIBILITY: use equivI rather than equiv_intro (same
  2119 for congruent(2)).
  2120 
  2121 * Some previously unqualified names have been qualified:
  2122 
  2123   types
  2124     bool ~> HOL.bool
  2125     nat ~> Nat.nat
  2126 
  2127   constants
  2128     Trueprop ~> HOL.Trueprop
  2129     True ~> HOL.True
  2130     False ~> HOL.False
  2131     op & ~> HOL.conj
  2132     op | ~> HOL.disj
  2133     op --> ~> HOL.implies
  2134     op = ~> HOL.eq
  2135     Not ~> HOL.Not
  2136     The ~> HOL.The
  2137     All ~> HOL.All
  2138     Ex ~> HOL.Ex
  2139     Ex1 ~> HOL.Ex1
  2140     Let ~> HOL.Let
  2141     If ~> HOL.If
  2142     Ball ~> Set.Ball
  2143     Bex ~> Set.Bex
  2144     Suc ~> Nat.Suc
  2145     Pair ~> Product_Type.Pair
  2146     fst ~> Product_Type.fst
  2147     snd ~> Product_Type.snd
  2148     curry ~> Product_Type.curry
  2149     op : ~> Set.member
  2150     Collect ~> Set.Collect
  2151 
  2152 INCOMPATIBILITY.
  2153 
  2154 * More canonical naming convention for some fundamental definitions:
  2155 
  2156     bot_bool_eq ~> bot_bool_def
  2157     top_bool_eq ~> top_bool_def
  2158     inf_bool_eq ~> inf_bool_def
  2159     sup_bool_eq ~> sup_bool_def
  2160     bot_fun_eq  ~> bot_fun_def
  2161     top_fun_eq  ~> top_fun_def
  2162     inf_fun_eq  ~> inf_fun_def
  2163     sup_fun_eq  ~> sup_fun_def
  2164 
  2165 INCOMPATIBILITY.
  2166 
  2167 * More stylized fact names:
  2168 
  2169   expand_fun_eq ~> fun_eq_iff
  2170   expand_set_eq ~> set_eq_iff
  2171   set_ext       ~> set_eqI
  2172   nat_number    ~> eval_nat_numeral
  2173 
  2174 INCOMPATIBILITY.
  2175 
  2176 * Refactoring of code-generation specific operations in theory List:
  2177 
  2178   constants
  2179     null ~> List.null
  2180 
  2181   facts
  2182     mem_iff ~> member_def
  2183     null_empty ~> null_def
  2184 
  2185 INCOMPATIBILITY.  Note that these were not supposed to be used
  2186 regularly unless for striking reasons; their main purpose was code
  2187 generation.
  2188 
  2189 Various operations from the Haskell prelude are used for generating
  2190 Haskell code.
  2191 
  2192 * Term "bij f" is now an abbreviation of "bij_betw f UNIV UNIV".  Term
  2193 "surj f" is now an abbreviation of "range f = UNIV".  The theorems
  2194 bij_def and surj_def are unchanged.  INCOMPATIBILITY.
  2195 
  2196 * Abolished some non-alphabetic type names: "prod" and "sum" replace
  2197 "*" and "+" respectively.  INCOMPATIBILITY.
  2198 
  2199 * Name "Plus" of disjoint sum operator "<+>" is now hidden.  Write
  2200 "Sum_Type.Plus" instead.
  2201 
  2202 * Constant "split" has been merged with constant "prod_case"; names of
  2203 ML functions, facts etc. involving split have been retained so far,
  2204 though.  INCOMPATIBILITY.
  2205 
  2206 * Dropped old infix syntax "_ mem _" for List.member; use "_ : set _"
  2207 instead.  INCOMPATIBILITY.
  2208 
  2209 * Removed lemma "Option.is_none_none" which duplicates "is_none_def".
  2210 INCOMPATIBILITY.
  2211 
  2212 * Former theory Library/Enum is now part of the HOL-Main image.
  2213 INCOMPATIBILITY: all constants of the Enum theory now have to be
  2214 referred to by its qualified name.
  2215 
  2216   enum    ~>  Enum.enum
  2217   nlists  ~>  Enum.nlists
  2218   product ~>  Enum.product
  2219 
  2220 * Theory Library/Monad_Syntax provides do-syntax for monad types.
  2221 Syntax in Library/State_Monad has been changed to avoid ambiguities.
  2222 INCOMPATIBILITY.
  2223 
  2224 * Theory Library/SetsAndFunctions has been split into
  2225 Library/Function_Algebras and Library/Set_Algebras; canonical names
  2226 for instance definitions for functions; various improvements.
  2227 INCOMPATIBILITY.
  2228 
  2229 * Theory Library/Multiset provides stable quicksort implementation of
  2230 sort_key.
  2231 
  2232 * Theory Library/Multiset: renamed empty_idemp ~> empty_neutral.
  2233 INCOMPATIBILITY.
  2234 
  2235 * Session Multivariate_Analysis: introduced a type class for euclidean
  2236 space.  Most theorems are now stated in terms of euclidean spaces
  2237 instead of finite cartesian products.
  2238 
  2239   types
  2240     real ^ 'n ~>  'a::real_vector
  2241               ~>  'a::euclidean_space
  2242               ~>  'a::ordered_euclidean_space
  2243         (depends on your needs)
  2244 
  2245   constants
  2246      _ $ _        ~> _ $$ _
  2247      \<chi> x. _  ~> \<chi>\<chi> x. _
  2248      CARD('n)     ~> DIM('a)
  2249 
  2250 Also note that the indices are now natural numbers and not from some
  2251 finite type. Finite cartesian products of euclidean spaces, products
  2252 of euclidean spaces the real and complex numbers are instantiated to
  2253 be euclidean_spaces.  INCOMPATIBILITY.
  2254 
  2255 * Session Probability: introduced pextreal as positive extended real
  2256 numbers.  Use pextreal as value for measures.  Introduce the
  2257 Radon-Nikodym derivative, product spaces and Fubini's theorem for
  2258 arbitrary sigma finite measures.  Introduces Lebesgue measure based on
  2259 the integral in Multivariate Analysis.  INCOMPATIBILITY.
  2260 
  2261 * Session Imperative_HOL: revamped, corrected dozens of inadequacies.
  2262 INCOMPATIBILITY.
  2263 
  2264 * Session SPARK (with image HOL-SPARK) provides commands to load and
  2265 prove verification conditions generated by the SPARK Ada program
  2266 verifier.  See also src/HOL/SPARK and src/HOL/SPARK/Examples.
  2267 
  2268 
  2269 *** HOL-Algebra ***
  2270 
  2271 * Theorems for additive ring operations (locale abelian_monoid and
  2272 descendants) are generated by interpretation from their multiplicative
  2273 counterparts.  Names (in particular theorem names) have the mandatory
  2274 qualifier 'add'.  Previous theorem names are redeclared for
  2275 compatibility.
  2276 
  2277 * Structure "int_ring" is now an abbreviation (previously a
  2278 definition).  This fits more natural with advanced interpretations.
  2279 
  2280 
  2281 *** HOLCF ***
  2282 
  2283 * The domain package now runs in definitional mode by default: The
  2284 former command 'new_domain' is now called 'domain'.  To use the domain
  2285 package in its original axiomatic mode, use 'domain (unsafe)'.
  2286 INCOMPATIBILITY.
  2287 
  2288 * The new class "domain" is now the default sort.  Class "predomain"
  2289 is an unpointed version of "domain". Theories can be updated by
  2290 replacing sort annotations as shown below.  INCOMPATIBILITY.
  2291 
  2292   'a::type ~> 'a::countable
  2293   'a::cpo  ~> 'a::predomain
  2294   'a::pcpo ~> 'a::domain
  2295 
  2296 * The old type class "rep" has been superseded by class "domain".
  2297 Accordingly, users of the definitional package must remove any
  2298 "default_sort rep" declarations.  INCOMPATIBILITY.
  2299 
  2300 * The domain package (definitional mode) now supports unpointed
  2301 predomain argument types, as long as they are marked 'lazy'. (Strict
  2302 arguments must be in class "domain".) For example, the following
  2303 domain definition now works:
  2304 
  2305   domain natlist = nil | cons (lazy "nat discr") (lazy "natlist")
  2306 
  2307 * Theory HOLCF/Library/HOL_Cpo provides cpo and predomain class
  2308 instances for types from main HOL: bool, nat, int, char, 'a + 'b,
  2309 'a option, and 'a list.  Additionally, it configures fixrec and the
  2310 domain package to work with these types.  For example:
  2311 
  2312   fixrec isInl :: "('a + 'b) u -> tr"
  2313     where "isInl$(up$(Inl x)) = TT" | "isInl$(up$(Inr y)) = FF"
  2314 
  2315   domain V = VFun (lazy "V -> V") | VCon (lazy "nat") (lazy "V list")
  2316 
  2317 * The "(permissive)" option of fixrec has been replaced with a
  2318 per-equation "(unchecked)" option. See
  2319 src/HOL/HOLCF/Tutorial/Fixrec_ex.thy for examples. INCOMPATIBILITY.
  2320 
  2321 * The "bifinite" class no longer fixes a constant "approx"; the class
  2322 now just asserts that such a function exists.  INCOMPATIBILITY.
  2323 
  2324 * Former type "alg_defl" has been renamed to "defl".  HOLCF no longer
  2325 defines an embedding of type 'a defl into udom by default; instances
  2326 of "bifinite" and "domain" classes are available in
  2327 src/HOL/HOLCF/Library/Defl_Bifinite.thy.
  2328 
  2329 * The syntax "REP('a)" has been replaced with "DEFL('a)".
  2330 
  2331 * The predicate "directed" has been removed.  INCOMPATIBILITY.
  2332 
  2333 * The type class "finite_po" has been removed.  INCOMPATIBILITY.
  2334 
  2335 * The function "cprod_map" has been renamed to "prod_map".
  2336 INCOMPATIBILITY.
  2337 
  2338 * The monadic bind operator on each powerdomain has new binder syntax
  2339 similar to sets, e.g. "\<Union>\<sharp>x\<in>xs. t" represents
  2340 "upper_bind\<cdot>xs\<cdot>(\<Lambda> x. t)".
  2341 
  2342 * The infix syntax for binary union on each powerdomain has changed
  2343 from e.g. "+\<sharp>" to "\<union>\<sharp>", for consistency with set
  2344 syntax.  INCOMPATIBILITY.
  2345 
  2346 * The constant "UU" has been renamed to "bottom".  The syntax "UU" is
  2347 still supported as an input translation.
  2348 
  2349 * Renamed some theorems (the original names are also still available).
  2350 
  2351   expand_fun_below   ~> fun_below_iff
  2352   below_fun_ext      ~> fun_belowI
  2353   expand_cfun_eq     ~> cfun_eq_iff
  2354   ext_cfun           ~> cfun_eqI
  2355   expand_cfun_below  ~> cfun_below_iff
  2356   below_cfun_ext     ~> cfun_belowI
  2357   cont2cont_Rep_CFun ~> cont2cont_APP
  2358 
  2359 * The Abs and Rep functions for various types have changed names.
  2360 Related theorem names have also changed to match. INCOMPATIBILITY.
  2361 
  2362   Rep_CFun  ~> Rep_cfun
  2363   Abs_CFun  ~> Abs_cfun
  2364   Rep_Sprod ~> Rep_sprod
  2365   Abs_Sprod ~> Abs_sprod
  2366   Rep_Ssum  ~> Rep_ssum
  2367   Abs_Ssum  ~> Abs_ssum
  2368 
  2369 * Lemmas with names of the form *_defined_iff or *_strict_iff have
  2370 been renamed to *_bottom_iff.  INCOMPATIBILITY.
  2371 
  2372 * Various changes to bisimulation/coinduction with domain package:
  2373 
  2374   - Definitions of "bisim" constants no longer mention definedness.
  2375   - With mutual recursion, "bisim" predicate is now curried.
  2376   - With mutual recursion, each type gets a separate coind theorem.
  2377   - Variable names in bisim_def and coinduct rules have changed.
  2378 
  2379 INCOMPATIBILITY.
  2380 
  2381 * Case combinators generated by the domain package for type "foo" are
  2382 now named "foo_case" instead of "foo_when".  INCOMPATIBILITY.
  2383 
  2384 * Several theorems have been renamed to more accurately reflect the
  2385 names of constants and types involved.  INCOMPATIBILITY.
  2386 
  2387   thelub_const    ~> lub_const
  2388   lub_const       ~> is_lub_const
  2389   thelubI         ~> lub_eqI
  2390   is_lub_lub      ~> is_lubD2
  2391   lubI            ~> is_lub_lub
  2392   unique_lub      ~> is_lub_unique
  2393   is_ub_lub       ~> is_lub_rangeD1
  2394   lub_bin_chain   ~> is_lub_bin_chain
  2395   lub_fun         ~> is_lub_fun
  2396   thelub_fun      ~> lub_fun
  2397   thelub_cfun     ~> lub_cfun
  2398   thelub_Pair     ~> lub_Pair
  2399   lub_cprod       ~> is_lub_prod
  2400   thelub_cprod    ~> lub_prod
  2401   minimal_cprod   ~> minimal_prod
  2402   inst_cprod_pcpo ~> inst_prod_pcpo
  2403   UU_I            ~> bottomI
  2404   compact_UU      ~> compact_bottom
  2405   deflation_UU    ~> deflation_bottom
  2406   finite_deflation_UU ~> finite_deflation_bottom
  2407 
  2408 * Many legacy theorem names have been discontinued.  INCOMPATIBILITY.
  2409 
  2410   sq_ord_less_eq_trans ~> below_eq_trans
  2411   sq_ord_eq_less_trans ~> eq_below_trans
  2412   refl_less            ~> below_refl
  2413   trans_less           ~> below_trans
  2414   antisym_less         ~> below_antisym
  2415   antisym_less_inverse ~> po_eq_conv [THEN iffD1]
  2416   box_less             ~> box_below
  2417   rev_trans_less       ~> rev_below_trans
  2418   not_less2not_eq      ~> not_below2not_eq
  2419   less_UU_iff          ~> below_UU_iff
  2420   flat_less_iff        ~> flat_below_iff
  2421   adm_less             ~> adm_below
  2422   adm_not_less         ~> adm_not_below
  2423   adm_compact_not_less ~> adm_compact_not_below
  2424   less_fun_def         ~> below_fun_def
  2425   expand_fun_less      ~> fun_below_iff
  2426   less_fun_ext         ~> fun_belowI
  2427   less_discr_def       ~> below_discr_def
  2428   discr_less_eq        ~> discr_below_eq
  2429   less_unit_def        ~> below_unit_def
  2430   less_cprod_def       ~> below_prod_def
  2431   prod_lessI           ~> prod_belowI
  2432   Pair_less_iff        ~> Pair_below_iff
  2433   fst_less_iff         ~> fst_below_iff
  2434   snd_less_iff         ~> snd_below_iff
  2435   expand_cfun_less     ~> cfun_below_iff
  2436   less_cfun_ext        ~> cfun_belowI
  2437   injection_less       ~> injection_below
  2438   less_up_def          ~> below_up_def
  2439   not_Iup_less         ~> not_Iup_below
  2440   Iup_less             ~> Iup_below
  2441   up_less              ~> up_below
  2442   Def_inject_less_eq   ~> Def_below_Def
  2443   Def_less_is_eq       ~> Def_below_iff
  2444   spair_less_iff       ~> spair_below_iff
  2445   less_sprod           ~> below_sprod
  2446   spair_less           ~> spair_below
  2447   sfst_less_iff        ~> sfst_below_iff
  2448   ssnd_less_iff        ~> ssnd_below_iff
  2449   fix_least_less       ~> fix_least_below
  2450   dist_less_one        ~> dist_below_one
  2451   less_ONE             ~> below_ONE
  2452   ONE_less_iff         ~> ONE_below_iff
  2453   less_sinlD           ~> below_sinlD
  2454   less_sinrD           ~> below_sinrD
  2455 
  2456 
  2457 *** FOL and ZF ***
  2458 
  2459 * All constant names are now qualified internally and use proper
  2460 identifiers, e.g. "IFOL.eq" instead of "op =".  INCOMPATIBILITY.
  2461 
  2462 
  2463 *** ML ***
  2464 
  2465 * Antiquotation @{assert} inlines a function bool -> unit that raises
  2466 Fail if the argument is false.  Due to inlining the source position of
  2467 failed assertions is included in the error output.
  2468 
  2469 * Discontinued antiquotation @{theory_ref}, which is obsolete since ML
  2470 text is in practice always evaluated with a stable theory checkpoint.
  2471 Minor INCOMPATIBILITY, use (Theory.check_thy @{theory}) instead.
  2472 
  2473 * Antiquotation @{theory A} refers to theory A from the ancestry of
  2474 the current context, not any accidental theory loader state as before.
  2475 Potential INCOMPATIBILITY, subtle change in semantics.
  2476 
  2477 * Syntax.pretty_priority (default 0) configures the required priority
  2478 of pretty-printed output and thus affects insertion of parentheses.
  2479 
  2480 * Syntax.default_root (default "any") configures the inner syntax
  2481 category (nonterminal symbol) for parsing of terms.
  2482 
  2483 * Former exception Library.UnequalLengths now coincides with
  2484 ListPair.UnequalLengths.
  2485 
  2486 * Renamed structure MetaSimplifier to Raw_Simplifier.  Note that the
  2487 main functionality is provided by structure Simplifier.
  2488 
  2489 * Renamed raw "explode" function to "raw_explode" to emphasize its
  2490 meaning.  Note that internally to Isabelle, Symbol.explode is used in
  2491 almost all situations.
  2492 
  2493 * Discontinued obsolete function sys_error and exception SYS_ERROR.
  2494 See implementation manual for further details on exceptions in
  2495 Isabelle/ML.
  2496 
  2497 * Renamed setmp_noncritical to Unsynchronized.setmp to emphasize its
  2498 meaning.
  2499 
  2500 * Renamed structure PureThy to Pure_Thy and moved most of its
  2501 operations to structure Global_Theory, to emphasize that this is
  2502 rarely-used global-only stuff.
  2503 
  2504 * Discontinued Output.debug.  Minor INCOMPATIBILITY, use plain writeln
  2505 instead (or tracing for high-volume output).
  2506 
  2507 * Configuration option show_question_marks only affects regular pretty
  2508 printing of types and terms, not raw Term.string_of_vname.
  2509 
  2510 * ML_Context.thm and ML_Context.thms are no longer pervasive.  Rare
  2511 INCOMPATIBILITY, superseded by static antiquotations @{thm} and
  2512 @{thms} for most purposes.
  2513 
  2514 * ML structure Unsynchronized is never opened, not even in Isar
  2515 interaction mode as before.  Old Unsynchronized.set etc. have been
  2516 discontinued -- use plain := instead.  This should be *rare* anyway,
  2517 since modern tools always work via official context data, notably
  2518 configuration options.
  2519 
  2520 * Parallel and asynchronous execution requires special care concerning
  2521 interrupts.  Structure Exn provides some convenience functions that
  2522 avoid working directly with raw Interrupt.  User code must not absorb
  2523 interrupts -- intermediate handling (for cleanup etc.) needs to be
  2524 followed by re-raising of the original exception.  Another common
  2525 source of mistakes are "handle _" patterns, which make the meaning of
  2526 the program subject to physical effects of the environment.
  2527 
  2528 
  2529 
  2530 New in Isabelle2009-2 (June 2010)
  2531 ---------------------------------
  2532 
  2533 *** General ***
  2534 
  2535 * Authentic syntax for *all* logical entities (type classes, type
  2536 constructors, term constants): provides simple and robust
  2537 correspondence between formal entities and concrete syntax.  Within
  2538 the parse tree / AST representations, "constants" are decorated by
  2539 their category (class, type, const) and spelled out explicitly with
  2540 their full internal name.
  2541 
  2542 Substantial INCOMPATIBILITY concerning low-level syntax declarations
  2543 and translations (translation rules and translation functions in ML).
  2544 Some hints on upgrading:
  2545 
  2546   - Many existing uses of 'syntax' and 'translations' can be replaced
  2547     by more modern 'type_notation', 'notation' and 'abbreviation',
  2548     which are independent of this issue.
  2549 
  2550   - 'translations' require markup within the AST; the term syntax
  2551     provides the following special forms:
  2552 
  2553       CONST c   -- produces syntax version of constant c from context
  2554       XCONST c  -- literally c, checked as constant from context
  2555       c         -- literally c, if declared by 'syntax'
  2556 
  2557     Plain identifiers are treated as AST variables -- occasionally the
  2558     system indicates accidental variables via the error "rhs contains
  2559     extra variables".
  2560 
  2561     Type classes and type constructors are marked according to their
  2562     concrete syntax.  Some old translations rules need to be written
  2563     for the "type" category, using type constructor application
  2564     instead of pseudo-term application of the default category
  2565     "logic".
  2566 
  2567   - 'parse_translation' etc. in ML may use the following
  2568     antiquotations:
  2569 
  2570       @{class_syntax c}   -- type class c within parse tree / AST
  2571       @{term_syntax c}    -- type constructor c within parse tree / AST
  2572       @{const_syntax c}   -- ML version of "CONST c" above
  2573       @{syntax_const c}   -- literally c (checked wrt. 'syntax' declarations)
  2574 
  2575   - Literal types within 'typed_print_translations', i.e. those *not*
  2576     represented as pseudo-terms are represented verbatim.  Use @{class
  2577     c} or @{type_name c} here instead of the above syntax
  2578     antiquotations.
  2579 
  2580 Note that old non-authentic syntax was based on unqualified base
  2581 names, so all of the above "constant" names would coincide.  Recall
  2582 that 'print_syntax' and ML_command "set Syntax.trace_ast" help to
  2583 diagnose syntax problems.
  2584 
  2585 * Type constructors admit general mixfix syntax, not just infix.
  2586 
  2587 * Concrete syntax may be attached to local entities without a proof
  2588 body, too.  This works via regular mixfix annotations for 'fix',
  2589 'def', 'obtain' etc. or via the explicit 'write' command, which is
  2590 similar to the 'notation' command in theory specifications.
  2591 
  2592 * Discontinued unnamed infix syntax (legacy feature for many years) --
  2593 need to specify constant name and syntax separately.  Internal ML
  2594 datatype constructors have been renamed from InfixName to Infix etc.
  2595 Minor INCOMPATIBILITY.
  2596 
  2597 * Schematic theorem statements need to be explicitly markup as such,
  2598 via commands 'schematic_lemma', 'schematic_theorem',
  2599 'schematic_corollary'.  Thus the relevance of the proof is made
  2600 syntactically clear, which impacts performance in a parallel or
  2601 asynchronous interactive environment.  Minor INCOMPATIBILITY.
  2602 
  2603 * Use of cumulative prems via "!" in some proof methods has been
  2604 discontinued (old legacy feature).
  2605 
  2606 * References 'trace_simp' and 'debug_simp' have been replaced by
  2607 configuration options stored in the context. Enabling tracing (the
  2608 case of debugging is similar) in proofs works via
  2609 
  2610   using [[trace_simp = true]]
  2611 
  2612 Tracing is then active for all invocations of the simplifier in
  2613 subsequent goal refinement steps. Tracing may also still be enabled or
  2614 disabled via the ProofGeneral settings menu.
  2615 
  2616 * Separate commands 'hide_class', 'hide_type', 'hide_const',
  2617 'hide_fact' replace the former 'hide' KIND command.  Minor
  2618 INCOMPATIBILITY.
  2619 
  2620 * Improved parallelism of proof term normalization: usedir -p2 -q0 is
  2621 more efficient than combinations with -q1 or -q2.
  2622 
  2623 
  2624 *** Pure ***
  2625 
  2626 * Proofterms record type-class reasoning explicitly, using the
  2627 "unconstrain" operation internally.  This eliminates all sort
  2628 constraints from a theorem and proof, introducing explicit
  2629 OFCLASS-premises.  On the proof term level, this operation is
  2630 automatically applied at theorem boundaries, such that closed proofs
  2631 are always free of sort constraints.  INCOMPATIBILITY for tools that
  2632 inspect proof terms.
  2633 
  2634 * Local theory specifications may depend on extra type variables that
  2635 are not present in the result type -- arguments TYPE('a) :: 'a itself
  2636 are added internally.  For example:
  2637 
  2638   definition unitary :: bool where "unitary = (ALL (x::'a) y. x = y)"
  2639 
  2640 * Predicates of locales introduced by classes carry a mandatory
  2641 "class" prefix.  INCOMPATIBILITY.
  2642 
  2643 * Vacuous class specifications observe default sort.  INCOMPATIBILITY.
  2644 
  2645 * Old 'axclass' command has been discontinued.  INCOMPATIBILITY, use
  2646 'class' instead.
  2647 
  2648 * Command 'code_reflect' allows to incorporate generated ML code into
  2649 runtime environment; replaces immature code_datatype antiquotation.
  2650 INCOMPATIBILITY.
  2651 
  2652 * Code generator: simple concept for abstract datatypes obeying
  2653 invariants.
  2654 
  2655 * Code generator: details of internal data cache have no impact on the
  2656 user space functionality any longer.
  2657 
  2658 * Methods "unfold_locales" and "intro_locales" ignore non-locale
  2659 subgoals.  This is more appropriate for interpretations with 'where'.
  2660 INCOMPATIBILITY.
  2661 
  2662 * Command 'example_proof' opens an empty proof body.  This allows to
  2663 experiment with Isar, without producing any persistent result.
  2664 
  2665 * Commands 'type_notation' and 'no_type_notation' declare type syntax
  2666 within a local theory context, with explicit checking of the
  2667 constructors involved (in contrast to the raw 'syntax' versions).
  2668 
  2669 * Commands 'types' and 'typedecl' now work within a local theory
  2670 context -- without introducing dependencies on parameters or
  2671 assumptions, which is not possible in Isabelle/Pure.
  2672 
  2673 * Command 'defaultsort' has been renamed to 'default_sort', it works
  2674 within a local theory context.  Minor INCOMPATIBILITY.
  2675 
  2676 
  2677 *** HOL ***
  2678 
  2679 * Command 'typedef' now works within a local theory context -- without
  2680 introducing dependencies on parameters or assumptions, which is not
  2681 possible in Isabelle/Pure/HOL.  Note that the logical environment may
  2682 contain multiple interpretations of local typedefs (with different
  2683 non-emptiness proofs), even in a global theory context.
  2684 
  2685 * New package for quotient types.  Commands 'quotient_type' and
  2686 'quotient_definition' may be used for defining types and constants by
  2687 quotient constructions.  An example is the type of integers created by
  2688 quotienting pairs of natural numbers:
  2689 
  2690   fun
  2691     intrel :: "(nat * nat) => (nat * nat) => bool"
  2692   where
  2693     "intrel (x, y) (u, v) = (x + v = u + y)"
  2694 
  2695   quotient_type int = "nat * nat" / intrel
  2696     by (auto simp add: equivp_def expand_fun_eq)
  2697 
  2698   quotient_definition
  2699     "0::int" is "(0::nat, 0::nat)"
  2700 
  2701 The method "lifting" can be used to lift of theorems from the
  2702 underlying "raw" type to the quotient type.  The example
  2703 src/HOL/Quotient_Examples/FSet.thy includes such a quotient
  2704 construction and provides a reasoning infrastructure for finite sets.
  2705 
  2706 * Renamed Library/Quotient.thy to Library/Quotient_Type.thy to avoid
  2707 clash with new theory Quotient in Main HOL.
  2708 
  2709 * Moved the SMT binding into the main HOL session, eliminating
  2710 separate HOL-SMT session.
  2711 
  2712 * List membership infix mem operation is only an input abbreviation.
  2713 INCOMPATIBILITY.
  2714 
  2715 * Theory Library/Word.thy has been removed.  Use library Word/Word.thy
  2716 for future developements; former Library/Word.thy is still present in
  2717 the AFP entry RSAPPS.
  2718 
  2719 * Theorem Int.int_induct renamed to Int.int_of_nat_induct and is no
  2720 longer shadowed.  INCOMPATIBILITY.
  2721 
  2722 * Dropped theorem duplicate comp_arith; use semiring_norm instead.
  2723 INCOMPATIBILITY.
  2724 
  2725 * Dropped theorem RealPow.real_sq_order; use power2_le_imp_le instead.
  2726 INCOMPATIBILITY.
  2727 
  2728 * Dropped normalizing_semiring etc; use the facts in semiring classes
  2729 instead.  INCOMPATIBILITY.
  2730 
  2731 * Dropped several real-specific versions of lemmas about floor and
  2732 ceiling; use the generic lemmas from theory "Archimedean_Field"
  2733 instead.  INCOMPATIBILITY.
  2734 
  2735   floor_number_of_eq         ~> floor_number_of
  2736   le_floor_eq_number_of      ~> number_of_le_floor
  2737   le_floor_eq_zero           ~> zero_le_floor
  2738   le_floor_eq_one            ~> one_le_floor
  2739   floor_less_eq_number_of    ~> floor_less_number_of
  2740   floor_less_eq_zero         ~> floor_less_zero
  2741   floor_less_eq_one          ~> floor_less_one
  2742   less_floor_eq_number_of    ~> number_of_less_floor
  2743   less_floor_eq_zero         ~> zero_less_floor
  2744   less_floor_eq_one          ~> one_less_floor
  2745   floor_le_eq_number_of      ~> floor_le_number_of
  2746   floor_le_eq_zero           ~> floor_le_zero
  2747   floor_le_eq_one            ~> floor_le_one
  2748   floor_subtract_number_of   ~> floor_diff_number_of
  2749   floor_subtract_one         ~> floor_diff_one
  2750   ceiling_number_of_eq       ~> ceiling_number_of
  2751   ceiling_le_eq_number_of    ~> ceiling_le_number_of
  2752   ceiling_le_zero_eq         ~> ceiling_le_zero
  2753   ceiling_le_eq_one          ~> ceiling_le_one
  2754   less_ceiling_eq_number_of  ~> number_of_less_ceiling
  2755   less_ceiling_eq_zero       ~> zero_less_ceiling
  2756   less_ceiling_eq_one        ~> one_less_ceiling
  2757   ceiling_less_eq_number_of  ~> ceiling_less_number_of
  2758   ceiling_less_eq_zero       ~> ceiling_less_zero
  2759   ceiling_less_eq_one        ~> ceiling_less_one
  2760   le_ceiling_eq_number_of    ~> number_of_le_ceiling
  2761   le_ceiling_eq_zero         ~> zero_le_ceiling
  2762   le_ceiling_eq_one          ~> one_le_ceiling
  2763   ceiling_subtract_number_of ~> ceiling_diff_number_of
  2764   ceiling_subtract_one       ~> ceiling_diff_one
  2765 
  2766 * Theory "Finite_Set": various folding_XXX locales facilitate the
  2767 application of the various fold combinators on finite sets.
  2768 
  2769 * Library theory "RBT" renamed to "RBT_Impl"; new library theory "RBT"
  2770 provides abstract red-black tree type which is backed by "RBT_Impl" as
  2771 implementation.  INCOMPATIBILTY.
  2772 
  2773 * Theory Library/Coinductive_List has been removed -- superseded by
  2774 AFP/thys/Coinductive.
  2775 
  2776 * Theory PReal, including the type "preal" and related operations, has
  2777 been removed.  INCOMPATIBILITY.
  2778 
  2779 * Real: new development using Cauchy Sequences.
  2780 
  2781 * Split off theory "Big_Operators" containing setsum, setprod,
  2782 Inf_fin, Sup_fin, Min, Max from theory Finite_Set.  INCOMPATIBILITY.
  2783 
  2784 * Theory "Rational" renamed to "Rat", for consistency with "Nat",
  2785 "Int" etc.  INCOMPATIBILITY.
  2786 
  2787 * Constant Rat.normalize needs to be qualified.  INCOMPATIBILITY.
  2788 
  2789 * New set of rules "ac_simps" provides combined assoc / commute
  2790 rewrites for all interpretations of the appropriate generic locales.
  2791 
  2792 * Renamed theory "OrderedGroup" to "Groups" and split theory
  2793 "Ring_and_Field" into theories "Rings" and "Fields"; for more
  2794 appropriate and more consistent names suitable for name prefixes
  2795 within the HOL theories.  INCOMPATIBILITY.
  2796 
  2797 * Some generic constants have been put to appropriate theories:
  2798   - less_eq, less: Orderings
  2799   - zero, one, plus, minus, uminus, times, abs, sgn: Groups
  2800   - inverse, divide: Rings
  2801 INCOMPATIBILITY.
  2802 
  2803 * More consistent naming of type classes involving orderings (and
  2804 lattices):
  2805 
  2806     lower_semilattice                   ~> semilattice_inf
  2807     upper_semilattice                   ~> semilattice_sup
  2808 
  2809     dense_linear_order                  ~> dense_linorder
  2810 
  2811     pordered_ab_group_add               ~> ordered_ab_group_add
  2812     pordered_ab_group_add_abs           ~> ordered_ab_group_add_abs
  2813     pordered_ab_semigroup_add           ~> ordered_ab_semigroup_add
  2814     pordered_ab_semigroup_add_imp_le    ~> ordered_ab_semigroup_add_imp_le
  2815     pordered_cancel_ab_semigroup_add    ~> ordered_cancel_ab_semigroup_add
  2816     pordered_cancel_comm_semiring       ~> ordered_cancel_comm_semiring
  2817     pordered_cancel_semiring            ~> ordered_cancel_semiring
  2818     pordered_comm_monoid_add            ~> ordered_comm_monoid_add
  2819     pordered_comm_ring                  ~> ordered_comm_ring
  2820     pordered_comm_semiring              ~> ordered_comm_semiring
  2821     pordered_ring                       ~> ordered_ring
  2822     pordered_ring_abs                   ~> ordered_ring_abs
  2823     pordered_semiring                   ~> ordered_semiring
  2824 
  2825     ordered_ab_group_add                ~> linordered_ab_group_add
  2826     ordered_ab_semigroup_add            ~> linordered_ab_semigroup_add
  2827     ordered_cancel_ab_semigroup_add     ~> linordered_cancel_ab_semigroup_add
  2828     ordered_comm_semiring_strict        ~> linordered_comm_semiring_strict
  2829     ordered_field                       ~> linordered_field
  2830     ordered_field_no_lb                 ~> linordered_field_no_lb
  2831     ordered_field_no_ub                 ~> linordered_field_no_ub
  2832     ordered_field_dense_linear_order    ~> dense_linordered_field
  2833     ordered_idom                        ~> linordered_idom
  2834     ordered_ring                        ~> linordered_ring
  2835     ordered_ring_le_cancel_factor       ~> linordered_ring_le_cancel_factor
  2836     ordered_ring_less_cancel_factor     ~> linordered_ring_less_cancel_factor
  2837     ordered_ring_strict                 ~> linordered_ring_strict
  2838     ordered_semidom                     ~> linordered_semidom
  2839     ordered_semiring                    ~> linordered_semiring
  2840     ordered_semiring_1                  ~> linordered_semiring_1
  2841     ordered_semiring_1_strict           ~> linordered_semiring_1_strict
  2842     ordered_semiring_strict             ~> linordered_semiring_strict
  2843 
  2844   The following slightly odd type classes have been moved to a
  2845   separate theory Library/Lattice_Algebras:
  2846 
  2847     lordered_ab_group_add               ~> lattice_ab_group_add
  2848     lordered_ab_group_add_abs           ~> lattice_ab_group_add_abs
  2849     lordered_ab_group_add_meet          ~> semilattice_inf_ab_group_add
  2850     lordered_ab_group_add_join          ~> semilattice_sup_ab_group_add
  2851     lordered_ring                       ~> lattice_ring
  2852 
  2853 INCOMPATIBILITY.
  2854 
  2855 * Refined field classes:
  2856   - classes division_ring_inverse_zero, field_inverse_zero,
  2857     linordered_field_inverse_zero include rule inverse 0 = 0 --
  2858     subsumes former division_by_zero class;
  2859   - numerous lemmas have been ported from field to division_ring.
  2860 INCOMPATIBILITY.
  2861 
  2862 * Refined algebra theorem collections:
  2863   - dropped theorem group group_simps, use algebra_simps instead;
  2864   - dropped theorem group ring_simps, use field_simps instead;
  2865   - proper theorem collection field_simps subsumes former theorem
  2866     groups field_eq_simps and field_simps;
  2867   - dropped lemma eq_minus_self_iff which is a duplicate for
  2868     equal_neg_zero.
  2869 INCOMPATIBILITY.
  2870 
  2871 * Theory Finite_Set and List: some lemmas have been generalized from
  2872 sets to lattices:
  2873 
  2874   fun_left_comm_idem_inter      ~> fun_left_comm_idem_inf
  2875   fun_left_comm_idem_union      ~> fun_left_comm_idem_sup
  2876   inter_Inter_fold_inter        ~> inf_Inf_fold_inf
  2877   union_Union_fold_union        ~> sup_Sup_fold_sup
  2878   Inter_fold_inter              ~> Inf_fold_inf
  2879   Union_fold_union              ~> Sup_fold_sup
  2880   inter_INTER_fold_inter        ~> inf_INFI_fold_inf
  2881   union_UNION_fold_union        ~> sup_SUPR_fold_sup
  2882   INTER_fold_inter              ~> INFI_fold_inf
  2883   UNION_fold_union              ~> SUPR_fold_sup
  2884 
  2885 * Theory "Complete_Lattice": lemmas top_def and bot_def have been
  2886 replaced by the more convenient lemmas Inf_empty and Sup_empty.
  2887 Dropped lemmas Inf_insert_simp and Sup_insert_simp, which are subsumed
  2888 by Inf_insert and Sup_insert.  Lemmas Inf_UNIV and Sup_UNIV replace
  2889 former Inf_Univ and Sup_Univ.  Lemmas inf_top_right and sup_bot_right
  2890 subsume inf_top and sup_bot respectively.  INCOMPATIBILITY.
  2891 
  2892 * Reorganized theory Multiset: swapped notation of pointwise and
  2893 multiset order:
  2894 
  2895   - pointwise ordering is instance of class order with standard syntax
  2896     <= and <;
  2897   - multiset ordering has syntax <=# and <#; partial order properties
  2898     are provided by means of interpretation with prefix
  2899     multiset_order;
  2900   - less duplication, less historical organization of sections,
  2901     conversion from associations lists to multisets, rudimentary code
  2902     generation;
  2903   - use insert_DiffM2 [symmetric] instead of elem_imp_eq_diff_union,
  2904     if needed.
  2905 
  2906 Renamed:
  2907 
  2908   multiset_eq_conv_count_eq  ~>  multiset_ext_iff
  2909   multi_count_ext  ~>  multiset_ext
  2910   diff_union_inverse2  ~>  diff_union_cancelR
  2911 
  2912 INCOMPATIBILITY.
  2913 
  2914 * Theory Permutation: replaced local "remove" by List.remove1.
  2915 
  2916 * Code generation: ML and OCaml code is decorated with signatures.
  2917 
  2918 * Theory List: added transpose.
  2919 
  2920 * Library/Nat_Bijection.thy is a collection of bijective functions
  2921 between nat and other types, which supersedes the older libraries
  2922 Library/Nat_Int_Bij.thy and HOLCF/NatIso.thy.  INCOMPATIBILITY.
  2923 
  2924   Constants:
  2925   Nat_Int_Bij.nat2_to_nat         ~> prod_encode
  2926   Nat_Int_Bij.nat_to_nat2         ~> prod_decode
  2927   Nat_Int_Bij.int_to_nat_bij      ~> int_encode
  2928   Nat_Int_Bij.nat_to_int_bij      ~> int_decode
  2929   Countable.pair_encode           ~> prod_encode
  2930   NatIso.prod2nat                 ~> prod_encode
  2931   NatIso.nat2prod                 ~> prod_decode
  2932   NatIso.sum2nat                  ~> sum_encode
  2933   NatIso.nat2sum                  ~> sum_decode
  2934   NatIso.list2nat                 ~> list_encode
  2935   NatIso.nat2list                 ~> list_decode
  2936   NatIso.set2nat                  ~> set_encode
  2937   NatIso.nat2set                  ~> set_decode
  2938 
  2939   Lemmas:
  2940   Nat_Int_Bij.bij_nat_to_int_bij  ~> bij_int_decode
  2941   Nat_Int_Bij.nat2_to_nat_inj     ~> inj_prod_encode
  2942   Nat_Int_Bij.nat2_to_nat_surj    ~> surj_prod_encode
  2943   Nat_Int_Bij.nat_to_nat2_inj     ~> inj_prod_decode
  2944   Nat_Int_Bij.nat_to_nat2_surj    ~> surj_prod_decode
  2945   Nat_Int_Bij.i2n_n2i_id          ~> int_encode_inverse
  2946   Nat_Int_Bij.n2i_i2n_id          ~> int_decode_inverse
  2947   Nat_Int_Bij.surj_nat_to_int_bij ~> surj_int_encode
  2948   Nat_Int_Bij.surj_int_to_nat_bij ~> surj_int_decode
  2949   Nat_Int_Bij.inj_nat_to_int_bij  ~> inj_int_encode
  2950   Nat_Int_Bij.inj_int_to_nat_bij  ~> inj_int_decode
  2951   Nat_Int_Bij.bij_nat_to_int_bij  ~> bij_int_encode
  2952   Nat_Int_Bij.bij_int_to_nat_bij  ~> bij_int_decode
  2953 
  2954 * Sledgehammer:
  2955   - Renamed ATP commands:
  2956     atp_info     ~> sledgehammer running_atps
  2957     atp_kill     ~> sledgehammer kill_atps
  2958     atp_messages ~> sledgehammer messages
  2959     atp_minimize ~> sledgehammer minimize
  2960     print_atps   ~> sledgehammer available_atps
  2961     INCOMPATIBILITY.
  2962   - Added user's manual ("isabelle doc sledgehammer").
  2963   - Added option syntax and "sledgehammer_params" to customize
  2964     Sledgehammer's behavior.  See the manual for details.
  2965   - Modified the Isar proof reconstruction code so that it produces
  2966     direct proofs rather than proofs by contradiction.  (This feature
  2967     is still experimental.)
  2968   - Made Isar proof reconstruction work for SPASS, remote ATPs, and in
  2969     full-typed mode.
  2970   - Added support for TPTP syntax for SPASS via the "spass_tptp" ATP.
  2971 
  2972 * Nitpick:
  2973   - Added and implemented "binary_ints" and "bits" options.
  2974   - Added "std" option and implemented support for nonstandard models.
  2975   - Added and implemented "finitize" option to improve the precision
  2976     of infinite datatypes based on a monotonicity analysis.
  2977   - Added support for quotient types.
  2978   - Added support for "specification" and "ax_specification"
  2979     constructs.
  2980   - Added support for local definitions (for "function" and
  2981     "termination" proofs).
  2982   - Added support for term postprocessors.
  2983   - Optimized "Multiset.multiset" and "FinFun.finfun".
  2984   - Improved efficiency of "destroy_constrs" optimization.
  2985   - Fixed soundness bugs related to "destroy_constrs" optimization and
  2986     record getters.
  2987   - Fixed soundness bug related to higher-order constructors.
  2988   - Fixed soundness bug when "full_descrs" is enabled.
  2989   - Improved precision of set constructs.
  2990   - Added "atoms" option.
  2991   - Added cache to speed up repeated Kodkod invocations on the same
  2992     problems.
  2993   - Renamed "MiniSatJNI", "zChaffJNI", "BerkMinAlloy", and
  2994     "SAT4JLight" to "MiniSat_JNI", "zChaff_JNI", "BerkMin_Alloy", and
  2995     "SAT4J_Light".  INCOMPATIBILITY.
  2996   - Removed "skolemize", "uncurry", "sym_break", "flatten_prop",
  2997     "sharing_depth", and "show_skolems" options.  INCOMPATIBILITY.
  2998   - Removed "nitpick_intro" attribute.  INCOMPATIBILITY.
  2999 
  3000 * Method "induct" now takes instantiations of the form t, where t is not
  3001   a variable, as a shorthand for "x == t", where x is a fresh variable.
  3002   If this is not intended, t has to be enclosed in parentheses.
  3003   By default, the equalities generated by definitional instantiations
  3004   are pre-simplified, which may cause parameters of inductive cases
  3005   to disappear, or may even delete some of the inductive cases.
  3006   Use "induct (no_simp)" instead of "induct" to restore the old
  3007   behaviour. The (no_simp) option is also understood by the "cases"
  3008   and "nominal_induct" methods, which now perform pre-simplification, too.
  3009   INCOMPATIBILITY.
  3010 
  3011 
  3012 *** HOLCF ***
  3013 
  3014 * Variable names in lemmas generated by the domain package have
  3015 changed; the naming scheme is now consistent with the HOL datatype
  3016 package.  Some proof scripts may be affected, INCOMPATIBILITY.
  3017 
  3018 * The domain package no longer defines the function "foo_copy" for
  3019 recursive domain "foo".  The reach lemma is now stated directly in
  3020 terms of "foo_take".  Lemmas and proofs that mention "foo_copy" must
  3021 be reformulated in terms of "foo_take", INCOMPATIBILITY.
  3022 
  3023 * Most definedness lemmas generated by the domain package (previously
  3024 of the form "x ~= UU ==> foo$x ~= UU") now have an if-and-only-if form
  3025 like "foo$x = UU <-> x = UU", which works better as a simp rule.
  3026 Proofs that used definedness lemmas as intro rules may break,
  3027 potential INCOMPATIBILITY.
  3028 
  3029 * Induction and casedist rules generated by the domain package now
  3030 declare proper case_names (one called "bottom", and one named for each
  3031 constructor).  INCOMPATIBILITY.
  3032 
  3033 * For mutually-recursive domains, separate "reach" and "take_lemma"
  3034 rules are generated for each domain, INCOMPATIBILITY.
  3035 
  3036   foo_bar.reach       ~> foo.reach  bar.reach
  3037   foo_bar.take_lemmas ~> foo.take_lemma  bar.take_lemma
  3038 
  3039 * Some lemmas generated by the domain package have been renamed for
  3040 consistency with the datatype package, INCOMPATIBILITY.
  3041 
  3042   foo.ind        ~> foo.induct
  3043   foo.finite_ind ~> foo.finite_induct
  3044   foo.coind      ~> foo.coinduct
  3045   foo.casedist   ~> foo.exhaust
  3046   foo.exhaust    ~> foo.nchotomy
  3047 
  3048 * For consistency with other definition packages, the fixrec package
  3049 now generates qualified theorem names, INCOMPATIBILITY.
  3050 
  3051   foo_simps  ~> foo.simps
  3052   foo_unfold ~> foo.unfold
  3053   foo_induct ~> foo.induct
  3054 
  3055 * The "fixrec_simp" attribute has been removed.  The "fixrec_simp"
  3056 method and internal fixrec proofs now use the default simpset instead.
  3057 INCOMPATIBILITY.
  3058 
  3059 * The "contlub" predicate has been removed.  Proof scripts should use
  3060 lemma contI2 in place of monocontlub2cont, INCOMPATIBILITY.
  3061 
  3062 * The "admw" predicate has been removed, INCOMPATIBILITY.
  3063 
  3064 * The constants cpair, cfst, and csnd have been removed in favor of
  3065 Pair, fst, and snd from Isabelle/HOL, INCOMPATIBILITY.
  3066 
  3067 
  3068 *** ML ***
  3069 
  3070 * Antiquotations for basic formal entities:
  3071 
  3072     @{class NAME}         -- type class
  3073     @{class_syntax NAME}  -- syntax representation of the above
  3074 
  3075     @{type_name NAME}     -- logical type
  3076     @{type_abbrev NAME}   -- type abbreviation
  3077     @{nonterminal NAME}   -- type of concrete syntactic category
  3078     @{type_syntax NAME}   -- syntax representation of any of the above
  3079 
  3080     @{const_name NAME}    -- logical constant (INCOMPATIBILITY)
  3081     @{const_abbrev NAME}  -- abbreviated constant
  3082     @{const_syntax NAME}  -- syntax representation of any of the above
  3083 
  3084 * Antiquotation @{syntax_const NAME} ensures that NAME refers to a raw
  3085 syntax constant (cf. 'syntax' command).
  3086 
  3087 * Antiquotation @{make_string} inlines a function to print arbitrary
  3088 values similar to the ML toplevel.  The result is compiler dependent
  3089 and may fall back on "?" in certain situations.
  3090 
  3091 * Diagnostic commands 'ML_val' and 'ML_command' may refer to
  3092 antiquotations @{Isar.state} and @{Isar.goal}.  This replaces impure
  3093 Isar.state() and Isar.goal(), which belong to the old TTY loop and do
  3094 not work with the asynchronous Isar document model.
  3095 
  3096 * Configuration options now admit dynamic default values, depending on
  3097 the context or even global references.
  3098 
  3099 * SHA1.digest digests strings according to SHA-1 (see RFC 3174).  It
  3100 uses an efficient external library if available (for Poly/ML).
  3101 
  3102 * Renamed some important ML structures, while keeping the old names
  3103 for some time as aliases within the structure Legacy:
  3104 
  3105   OuterKeyword  ~>  Keyword
  3106   OuterLex      ~>  Token
  3107   OuterParse    ~>  Parse
  3108   OuterSyntax   ~>  Outer_Syntax
  3109   PrintMode     ~>  Print_Mode
  3110   SpecParse     ~>  Parse_Spec
  3111   ThyInfo       ~>  Thy_Info
  3112   ThyLoad       ~>  Thy_Load
  3113   ThyOutput     ~>  Thy_Output
  3114   TypeInfer     ~>  Type_Infer
  3115 
  3116 Note that "open Legacy" simplifies porting of sources, but forgetting
  3117 to remove it again will complicate porting again in the future.
  3118 
  3119 * Most operations that refer to a global context are named
  3120 accordingly, e.g. Simplifier.global_context or
  3121 ProofContext.init_global.  There are some situations where a global
  3122 context actually works, but under normal circumstances one needs to
  3123 pass the proper local context through the code!
  3124 
  3125 * Discontinued old TheoryDataFun with its copy/init operation -- data
  3126 needs to be pure.  Functor Theory_Data_PP retains the traditional
  3127 Pretty.pp argument to merge, which is absent in the standard
  3128 Theory_Data version.
  3129 
  3130 * Sorts.certify_sort and derived "cert" operations for types and terms
  3131 no longer minimize sorts.  Thus certification at the boundary of the
  3132 inference kernel becomes invariant under addition of class relations,
  3133 which is an important monotonicity principle.  Sorts are now minimized
  3134 in the syntax layer only, at the boundary between the end-user and the
  3135 system.  Subtle INCOMPATIBILITY, may have to use Sign.minimize_sort
  3136 explicitly in rare situations.
  3137 
  3138 * Renamed old-style Drule.standard to Drule.export_without_context, to
  3139 emphasize that this is in no way a standard operation.
  3140 INCOMPATIBILITY.
  3141 
  3142 * Subgoal.FOCUS (and variants): resulting goal state is normalized as
  3143 usual for resolution.  Rare INCOMPATIBILITY.
  3144 
  3145 * Renamed varify/unvarify operations to varify_global/unvarify_global
  3146 to emphasize that these only work in a global situation (which is
  3147 quite rare).
  3148 
  3149 * Curried take and drop in library.ML; negative length is interpreted
  3150 as infinity (as in chop).  Subtle INCOMPATIBILITY.
  3151 
  3152 * Proof terms: type substitutions on proof constants now use canonical
  3153 order of type variables.  INCOMPATIBILITY for tools working with proof
  3154 terms.
  3155 
  3156 * Raw axioms/defs may no longer carry sort constraints, and raw defs
  3157 may no longer carry premises.  User-level specifications are
  3158 transformed accordingly by Thm.add_axiom/add_def.
  3159 
  3160 
  3161 *** System ***
  3162 
  3163 * Discontinued special HOL_USEDIR_OPTIONS for the main HOL image;
  3164 ISABELLE_USEDIR_OPTIONS applies uniformly to all sessions.  Note that
  3165 proof terms are enabled unconditionally in the new HOL-Proofs image.
  3166 
  3167 * Discontinued old ISABELLE and ISATOOL environment settings (legacy
  3168 feature since Isabelle2009).  Use ISABELLE_PROCESS and ISABELLE_TOOL,
  3169 respectively.
  3170 
  3171 * Old lib/scripts/polyml-platform is superseded by the
  3172 ISABELLE_PLATFORM setting variable, which defaults to the 32 bit
  3173 variant, even on a 64 bit machine.  The following example setting
  3174 prefers 64 bit if available:
  3175 
  3176   ML_PLATFORM="${ISABELLE_PLATFORM64:-$ISABELLE_PLATFORM}"
  3177 
  3178 * The preliminary Isabelle/jEdit application demonstrates the emerging
  3179 Isabelle/Scala layer for advanced prover interaction and integration.
  3180 See src/Tools/jEdit or "isabelle jedit" provided by the properly built
  3181 component.
  3182 
  3183 * "IsabelleText" is a Unicode font derived from Bitstream Vera Mono
  3184 and Bluesky TeX fonts.  It provides the usual Isabelle symbols,
  3185 similar to the default assignment of the document preparation system
  3186 (cf. isabellesym.sty).  The Isabelle/Scala class Isabelle_System
  3187 provides some operations for direct access to the font without asking
  3188 the user for manual installation.
  3189 
  3190 
  3191 
  3192 New in Isabelle2009-1 (December 2009)
  3193 -------------------------------------
  3194 
  3195 *** General ***
  3196 
  3197 * Discontinued old form of "escaped symbols" such as \\<forall>.  Only
  3198 one backslash should be used, even in ML sources.
  3199 
  3200 
  3201 *** Pure ***
  3202 
  3203 * Locale interpretation propagates mixins along the locale hierarchy.
  3204 The currently only available mixins are the equations used to map
  3205 local definitions to terms of the target domain of an interpretation.
  3206 
  3207 * Reactivated diagnostic command 'print_interps'.  Use "print_interps
  3208 loc" to print all interpretations of locale "loc" in the theory.
  3209 Interpretations in proofs are not shown.
  3210 
  3211 * Thoroughly revised locales tutorial.  New section on conditional
  3212 interpretation.
  3213 
  3214 * On instantiation of classes, remaining undefined class parameters
  3215 are formally declared.  INCOMPATIBILITY.
  3216 
  3217 
  3218 *** Document preparation ***
  3219 
  3220 * New generalized style concept for printing terms: @{foo (style) ...}
  3221 instead of @{foo_style style ...}  (old form is still retained for
  3222 backward compatibility).  Styles can be also applied for
  3223 antiquotations prop, term_type and typeof.
  3224 
  3225 
  3226 *** HOL ***
  3227 
  3228 * New proof method "smt" for a combination of first-order logic with
  3229 equality, linear and nonlinear (natural/integer/real) arithmetic, and
  3230 fixed-size bitvectors; there is also basic support for higher-order
  3231 features (esp. lambda abstractions).  It is an incomplete decision
  3232 procedure based on external SMT solvers using the oracle mechanism;
  3233 for the SMT solver Z3, this method is proof-producing.  Certificates
  3234 are provided to avoid calling the external solvers solely for
  3235 re-checking proofs.  Due to a remote SMT service there is no need for
  3236 installing SMT solvers locally.  See src/HOL/SMT.
  3237 
  3238 * New commands to load and prove verification conditions generated by
  3239 the Boogie program verifier or derived systems (e.g. the Verifying C
  3240 Compiler (VCC) or Spec#).  See src/HOL/Boogie.
  3241 
  3242 * New counterexample generator tool 'nitpick' based on the Kodkod
  3243 relational model finder.  See src/HOL/Tools/Nitpick and
  3244 src/HOL/Nitpick_Examples.
  3245 
  3246 * New commands 'code_pred' and 'values' to invoke the predicate
  3247 compiler and to enumerate values of inductive predicates.
  3248 
  3249 * A tabled implementation of the reflexive transitive closure.
  3250 
  3251 * New implementation of quickcheck uses generic code generator;
  3252 default generators are provided for all suitable HOL types, records
  3253 and datatypes.  Old quickcheck can be re-activated importing theory
  3254 Library/SML_Quickcheck.
  3255 
  3256 * New testing tool Mirabelle for automated proof tools.  Applies
  3257 several tools and tactics like sledgehammer, metis, or quickcheck, to
  3258 every proof step in a theory.  To be used in batch mode via the
  3259 "mirabelle" utility.
  3260 
  3261 * New proof method "sos" (sum of squares) for nonlinear real
  3262 arithmetic (originally due to John Harison). It requires theory
  3263 Library/Sum_Of_Squares.  It is not a complete decision procedure but
  3264 works well in practice on quantifier-free real arithmetic with +, -,
  3265 *, ^, =, <= and <, i.e. boolean combinations of equalities and
  3266 inequalities between polynomials.  It makes use of external
  3267 semidefinite programming solvers.  Method "sos" generates a
  3268 certificate that can be pasted into the proof thus avoiding the need
  3269 to call an external tool every time the proof is checked.  See
  3270 src/HOL/Library/Sum_Of_Squares.
  3271 
  3272 * New method "linarith" invokes existing linear arithmetic decision
  3273 procedure only.
  3274 
  3275 * New command 'atp_minimal' reduces result produced by Sledgehammer.
  3276 
  3277 * New Sledgehammer option "Full Types" in Proof General settings menu.
  3278 Causes full type information to be output to the ATPs.  This slows
  3279 ATPs down considerably but eliminates a source of unsound "proofs"
  3280 that fail later.
  3281 
  3282 * New method "metisFT": A version of metis that uses full type
  3283 information in order to avoid failures of proof reconstruction.
  3284 
  3285 * New evaluator "approximate" approximates an real valued term using
  3286 the same method as the approximation method.
  3287 
  3288 * Method "approximate" now supports arithmetic expressions as
  3289 boundaries of intervals and implements interval splitting and Taylor
  3290 series expansion.
  3291 
  3292 * ML antiquotation @{code_datatype} inserts definition of a datatype
  3293 generated by the code generator; e.g. see src/HOL/Predicate.thy.
  3294 
  3295 * New theory SupInf of the supremum and infimum operators for sets of
  3296 reals.
  3297 
  3298 * New theory Probability, which contains a development of measure
  3299 theory, eventually leading to Lebesgue integration and probability.
  3300 
  3301 * Extended Multivariate Analysis to include derivation and Brouwer's
  3302 fixpoint theorem.
  3303 
  3304 * Reorganization of number theory, INCOMPATIBILITY:
  3305   - new number theory development for nat and int, in theories Divides
  3306     and GCD as well as in new session Number_Theory
  3307   - some constants and facts now suffixed with _nat and _int
  3308     accordingly
  3309   - former session NumberTheory now named Old_Number_Theory, including
  3310     theories Legacy_GCD and Primes (prefer Number_Theory if possible)
  3311   - moved theory Pocklington from src/HOL/Library to
  3312     src/HOL/Old_Number_Theory
  3313 
  3314 * Theory GCD includes functions Gcd/GCD and Lcm/LCM for the gcd and
  3315 lcm of finite and infinite sets. It is shown that they form a complete
  3316 lattice.
  3317 
  3318 * Class semiring_div requires superclass no_zero_divisors and proof of
  3319 div_mult_mult1; theorems div_mult_mult1, div_mult_mult2,
  3320 div_mult_mult1_if, div_mult_mult1 and div_mult_mult2 have been
  3321 generalized to class semiring_div, subsuming former theorems
  3322 zdiv_zmult_zmult1, zdiv_zmult_zmult1_if, zdiv_zmult_zmult1 and
  3323 zdiv_zmult_zmult2.  div_mult_mult1 is now [simp] by default.
  3324 INCOMPATIBILITY.
  3325 
  3326 * Refinements to lattice classes and sets:
  3327   - less default intro/elim rules in locale variant, more default
  3328     intro/elim rules in class variant: more uniformity
  3329   - lemma ge_sup_conv renamed to le_sup_iff, in accordance with
  3330     le_inf_iff
  3331   - dropped lemma alias inf_ACI for inf_aci (same for sup_ACI and
  3332     sup_aci)
  3333   - renamed ACI to inf_sup_aci
  3334   - new class "boolean_algebra"
  3335   - class "complete_lattice" moved to separate theory
  3336     "Complete_Lattice"; corresponding constants (and abbreviations)
  3337     renamed and with authentic syntax:
  3338     Set.Inf ~>    Complete_Lattice.Inf
  3339     Set.Sup ~>    Complete_Lattice.Sup
  3340     Set.INFI ~>   Complete_Lattice.INFI
  3341     Set.SUPR ~>   Complete_Lattice.SUPR
  3342     Set.Inter ~>  Complete_Lattice.Inter
  3343     Set.Union ~>  Complete_Lattice.Union
  3344     Set.INTER ~>  Complete_Lattice.INTER
  3345     Set.UNION ~>  Complete_Lattice.UNION
  3346   - authentic syntax for
  3347     Set.Pow
  3348     Set.image
  3349   - mere abbreviations:
  3350     Set.empty               (for bot)
  3351     Set.UNIV                (for top)
  3352     Set.inter               (for inf, formerly Set.Int)
  3353     Set.union               (for sup, formerly Set.Un)
  3354     Complete_Lattice.Inter  (for Inf)
  3355     Complete_Lattice.Union  (for Sup)
  3356     Complete_Lattice.INTER  (for INFI)
  3357     Complete_Lattice.UNION  (for SUPR)
  3358   - object-logic definitions as far as appropriate
  3359 
  3360 INCOMPATIBILITY.  Care is required when theorems Int_subset_iff or
  3361 Un_subset_iff are explicitly deleted as default simp rules; then also
  3362 their lattice counterparts le_inf_iff and le_sup_iff have to be
  3363 deleted to achieve the desired effect.
  3364 
  3365 * Rules inf_absorb1, inf_absorb2, sup_absorb1, sup_absorb2 are no simp
  3366 rules by default any longer; the same applies to min_max.inf_absorb1
  3367 etc.  INCOMPATIBILITY.
  3368 
  3369 * Rules sup_Int_eq and sup_Un_eq are no longer declared as
  3370 pred_set_conv by default.  INCOMPATIBILITY.
  3371 
  3372 * Power operations on relations and functions are now one dedicated
  3373 constant "compow" with infix syntax "^^".  Power operation on
  3374 multiplicative monoids retains syntax "^" and is now defined generic
  3375 in class power.  INCOMPATIBILITY.
  3376 
  3377 * Relation composition "R O S" now has a more standard argument order:
  3378 "R O S = {(x, z). EX y. (x, y) : R & (y, z) : S}".  INCOMPATIBILITY,
  3379 rewrite propositions with "S O R" --> "R O S". Proofs may occasionally
  3380 break, since the O_assoc rule was not rewritten like this.  Fix using
  3381 O_assoc[symmetric].  The same applies to the curried version "R OO S".
  3382 
  3383 * Function "Inv" is renamed to "inv_into" and function "inv" is now an
  3384 abbreviation for "inv_into UNIV".  Lemmas are renamed accordingly.
  3385 INCOMPATIBILITY.
  3386 
  3387 * Most rules produced by inductive and datatype package have mandatory
  3388 prefixes.  INCOMPATIBILITY.
  3389 
  3390 * Changed "DERIV_intros" to a dynamic fact, which can be augmented by
  3391 the attribute of the same name.  Each of the theorems in the list
  3392 DERIV_intros assumes composition with an additional function and
  3393 matches a variable to the derivative, which has to be solved by the
  3394 Simplifier.  Hence (auto intro!: DERIV_intros) computes the derivative
  3395 of most elementary terms.  Former Maclauren.DERIV_tac and
  3396 Maclauren.deriv_tac should be replaced by (auto intro!: DERIV_intros).
  3397 INCOMPATIBILITY.
  3398 
  3399 * Code generator attributes follow the usual underscore convention:
  3400     code_unfold     replaces    code unfold
  3401     code_post       replaces    code post
  3402     etc.
  3403   INCOMPATIBILITY.
  3404 
  3405 * Renamed methods:
  3406     sizechange -> size_change
  3407     induct_scheme -> induction_schema
  3408   INCOMPATIBILITY.
  3409 
  3410 * Discontinued abbreviation "arbitrary" of constant "undefined".
  3411 INCOMPATIBILITY, use "undefined" directly.
  3412 
  3413 * Renamed theorems:
  3414     Suc_eq_add_numeral_1 -> Suc_eq_plus1
  3415     Suc_eq_add_numeral_1_left -> Suc_eq_plus1_left
  3416     Suc_plus1 -> Suc_eq_plus1
  3417     *anti_sym -> *antisym*
  3418     vector_less_eq_def -> vector_le_def
  3419   INCOMPATIBILITY.
  3420 
  3421 * Added theorem List.map_map as [simp].  Removed List.map_compose.
  3422 INCOMPATIBILITY.
  3423 
  3424 * Removed predicate "M hassize n" (<--> card M = n & finite M).
  3425 INCOMPATIBILITY.
  3426 
  3427 
  3428 *** HOLCF ***
  3429 
  3430 * Theory Representable defines a class "rep" of domains that are
  3431 representable (via an ep-pair) in the universal domain type "udom".
  3432 Instances are provided for all type constructors defined in HOLCF.
  3433 
  3434 * The 'new_domain' command is a purely definitional version of the
  3435 domain package, for representable domains.  Syntax is identical to the
  3436 old domain package.  The 'new_domain' package also supports indirect
  3437 recursion using previously-defined type constructors.  See
  3438 src/HOLCF/ex/New_Domain.thy for examples.
  3439 
  3440 * Method "fixrec_simp" unfolds one step of a fixrec-defined constant
  3441 on the left-hand side of an equation, and then performs
  3442 simplification.  Rewriting is done using rules declared with the
  3443 "fixrec_simp" attribute.  The "fixrec_simp" method is intended as a
  3444 replacement for "fixpat"; see src/HOLCF/ex/Fixrec_ex.thy for examples.
  3445 
  3446 * The pattern-match compiler in 'fixrec' can now handle constructors
  3447 with HOL function types.  Pattern-match combinators for the Pair
  3448 constructor are pre-configured.
  3449 
  3450 * The 'fixrec' package now produces better fixed-point induction rules
  3451 for mutually-recursive definitions:  Induction rules have conclusions
  3452 of the form "P foo bar" instead of "P <foo, bar>".
  3453 
  3454 * The constant "sq_le" (with infix syntax "<<" or "\<sqsubseteq>") has
  3455 been renamed to "below".  The name "below" now replaces "less" in many
  3456 theorem names.  (Legacy theorem names using "less" are still supported
  3457 as well.)
  3458 
  3459 * The 'fixrec' package now supports "bottom patterns".  Bottom
  3460 patterns can be used to generate strictness rules, or to make
  3461 functions more strict (much like the bang-patterns supported by the
  3462 Glasgow Haskell Compiler).  See src/HOLCF/ex/Fixrec_ex.thy for
  3463 examples.
  3464 
  3465 
  3466 *** ML ***
  3467 
  3468 * Support for Poly/ML 5.3.0, with improved reporting of compiler
  3469 errors and run-time exceptions, including detailed source positions.
  3470 
  3471 * Structure Name_Space (formerly NameSpace) now manages uniquely
  3472 identified entries, with some additional information such as source
  3473 position, logical grouping etc.
  3474 
  3475 * Theory and context data is now introduced by the simplified and
  3476 modernized functors Theory_Data, Proof_Data, Generic_Data.  Data needs
  3477 to be pure, but the old TheoryDataFun for mutable data (with explicit
  3478 copy operation) is still available for some time.
  3479 
  3480 * Structure Synchronized (cf. src/Pure/Concurrent/synchronized.ML)
  3481 provides a high-level programming interface to synchronized state
  3482 variables with atomic update.  This works via pure function
  3483 application within a critical section -- its runtime should be as
  3484 short as possible; beware of deadlocks if critical code is nested,
  3485 either directly or indirectly via other synchronized variables!
  3486 
  3487 * Structure Unsynchronized (cf. src/Pure/ML-Systems/unsynchronized.ML)
  3488 wraps raw ML references, explicitly indicating their non-thread-safe
  3489 behaviour.  The Isar toplevel keeps this structure open, to
  3490 accommodate Proof General as well as quick and dirty interactive
  3491 experiments with references.
  3492 
  3493 * PARALLEL_CHOICE and PARALLEL_GOALS provide basic support for
  3494 parallel tactical reasoning.
  3495 
  3496 * Tacticals Subgoal.FOCUS, Subgoal.FOCUS_PREMS, Subgoal.FOCUS_PARAMS
  3497 are similar to SUBPROOF, but are slightly more flexible: only the
  3498 specified parts of the subgoal are imported into the context, and the
  3499 body tactic may introduce new subgoals and schematic variables.
  3500 
  3501 * Old tactical METAHYPS, which does not observe the proof context, has
  3502 been renamed to Old_Goals.METAHYPS and awaits deletion.  Use SUBPROOF
  3503 or Subgoal.FOCUS etc.
  3504 
  3505 * Renamed functor TableFun to Table, and GraphFun to Graph.  (Since
  3506 functors have their own ML name space there is no point to mark them
  3507 separately.)  Minor INCOMPATIBILITY.
  3508 
  3509 * Renamed NamedThmsFun to Named_Thms.  INCOMPATIBILITY.
  3510 
  3511 * Renamed several structures FooBar to Foo_Bar.  Occasional,
  3512 INCOMPATIBILITY.
  3513 
  3514 * Operations of structure Skip_Proof no longer require quick_and_dirty
  3515 mode, which avoids critical setmp.
  3516 
  3517 * Eliminated old Attrib.add_attributes, Method.add_methods and related
  3518 combinators for "args".  INCOMPATIBILITY, need to use simplified
  3519 Attrib/Method.setup introduced in Isabelle2009.
  3520 
  3521 * Proper context for simpset_of, claset_of, clasimpset_of.  May fall
  3522 back on global_simpset_of, global_claset_of, global_clasimpset_of as
  3523 last resort.  INCOMPATIBILITY.
  3524 
  3525 * Display.pretty_thm now requires a proper context (cf. former
  3526 ProofContext.pretty_thm).  May fall back on Display.pretty_thm_global
  3527 or even Display.pretty_thm_without_context as last resort.
  3528 INCOMPATIBILITY.
  3529 
  3530 * Discontinued Display.pretty_ctyp/cterm etc.  INCOMPATIBILITY, use
  3531 Syntax.pretty_typ/term directly, preferably with proper context
  3532 instead of global theory.
  3533 
  3534 
  3535 *** System ***
  3536 
  3537 * Further fine tuning of parallel proof checking, scales up to 8 cores
  3538 (max. speedup factor 5.0).  See also Goal.parallel_proofs in ML and
  3539 usedir option -q.
  3540 
  3541 * Support for additional "Isabelle components" via etc/components, see
  3542 also the system manual.
  3543 
  3544 * The isabelle makeall tool now operates on all components with
  3545 IsaMakefile, not just hardwired "logics".
  3546 
  3547 * Removed "compress" option from isabelle-process and isabelle usedir;
  3548 this is always enabled.
  3549 
  3550 * Discontinued support for Poly/ML 4.x versions.
  3551 
  3552 * Isabelle tool "wwwfind" provides web interface for 'find_theorems'
  3553 on a given logic image.  This requires the lighttpd webserver and is
  3554 currently supported on Linux only.
  3555 
  3556 
  3557 
  3558 New in Isabelle2009 (April 2009)
  3559 --------------------------------
  3560 
  3561 *** General ***
  3562 
  3563 * Simplified main Isabelle executables, with less surprises on
  3564 case-insensitive file-systems (such as Mac OS).
  3565 
  3566   - The main Isabelle tool wrapper is now called "isabelle" instead of
  3567     "isatool."
  3568 
  3569   - The former "isabelle" alias for "isabelle-process" has been
  3570     removed (should rarely occur to regular users).
  3571 
  3572   - The former "isabelle-interface" and its alias "Isabelle" have been
  3573     removed (interfaces are now regular Isabelle tools).
  3574 
  3575 Within scripts and make files, the Isabelle environment variables
  3576 ISABELLE_TOOL and ISABELLE_PROCESS replace old ISATOOL and ISABELLE,
  3577 respectively.  (The latter are still available as legacy feature.)
  3578 
  3579 The old isabelle-interface wrapper could react in confusing ways if
  3580 the interface was uninstalled or changed otherwise.  Individual
  3581 interface tool configuration is now more explicit, see also the
  3582 Isabelle system manual.  In particular, Proof General is now available
  3583 via "isabelle emacs".
  3584 
  3585 INCOMPATIBILITY, need to adapt derivative scripts.  Users may need to
  3586 purge installed copies of Isabelle executables and re-run "isabelle
  3587 install -p ...", or use symlinks.
  3588 
  3589 * The default for ISABELLE_HOME_USER is now ~/.isabelle instead of the
  3590 old ~/isabelle, which was slightly non-standard and apt to cause
  3591 surprises on case-insensitive file-systems (such as Mac OS).
  3592 
  3593 INCOMPATIBILITY, need to move existing ~/isabelle/etc,
  3594 ~/isabelle/heaps, ~/isabelle/browser_info to the new place.  Special
  3595 care is required when using older releases of Isabelle.  Note that
  3596 ISABELLE_HOME_USER can be changed in Isabelle/etc/settings of any
  3597 Isabelle distribution, in order to use the new ~/.isabelle uniformly.
  3598 
  3599 * Proofs of fully specified statements are run in parallel on
  3600 multi-core systems.  A speedup factor of 2.5 to 3.2 can be expected on
  3601 a regular 4-core machine, if the initial heap space is made reasonably
  3602 large (cf. Poly/ML option -H).  (Requires Poly/ML 5.2.1 or later.)
  3603 
  3604 * The main reference manuals ("isar-ref", "implementation", and
  3605 "system") have been updated and extended.  Formally checked references
  3606 as hyperlinks are now available uniformly.
  3607 
  3608 
  3609 *** Pure ***
  3610 
  3611 * Complete re-implementation of locales.  INCOMPATIBILITY in several
  3612 respects.  The most important changes are listed below.  See the
  3613 Tutorial on Locales ("locales" manual) for details.
  3614 
  3615 - In locale expressions, instantiation replaces renaming.  Parameters
  3616 must be declared in a for clause.  To aid compatibility with previous
  3617 parameter inheritance, in locale declarations, parameters that are not
  3618 'touched' (instantiation position "_" or omitted) are implicitly added
  3619 with their syntax at the beginning of the for clause.
  3620 
  3621 - Syntax from abbreviations and definitions in locales is available in
  3622 locale expressions and context elements.  The latter is particularly
  3623 useful in locale declarations.
  3624 
  3625 - More flexible mechanisms to qualify names generated by locale
  3626 expressions.  Qualifiers (prefixes) may be specified in locale
  3627 expressions, and can be marked as mandatory (syntax: "name!:") or
  3628 optional (syntax "name?:").  The default depends for plain "name:"
  3629 depends on the situation where a locale expression is used: in
  3630 commands 'locale' and 'sublocale' prefixes are optional, in
  3631 'interpretation' and 'interpret' prefixes are mandatory.  The old
  3632 implicit qualifiers derived from the parameter names of a locale are
  3633 no longer generated.
  3634 
  3635 - Command "sublocale l < e" replaces "interpretation l < e".  The
  3636 instantiation clause in "interpretation" and "interpret" (square
  3637 brackets) is no longer available.  Use locale expressions.
  3638 
  3639 - When converting proof scripts, mandatory qualifiers in
  3640 'interpretation' and 'interpret' should be retained by default, even
  3641 if this is an INCOMPATIBILITY compared to former behavior.  In the
  3642 worst case, use the "name?:" form for non-mandatory ones.  Qualifiers
  3643 in locale expressions range over a single locale instance only.
  3644 
  3645 - Dropped locale element "includes".  This is a major INCOMPATIBILITY.
  3646 In existing theorem specifications replace the includes element by the
  3647 respective context elements of the included locale, omitting those
  3648 that are already present in the theorem specification.  Multiple
  3649 assume elements of a locale should be replaced by a single one
  3650 involving the locale predicate.  In the proof body, declarations (most
  3651 notably theorems) may be regained by interpreting the respective
  3652 locales in the proof context as required (command "interpret").
  3653 
  3654 If using "includes" in replacement of a target solely because the
  3655 parameter types in the theorem are not as general as in the target,
  3656 consider declaring a new locale with additional type constraints on
  3657 the parameters (context element "constrains").
  3658 
  3659 - Discontinued "locale (open)".  INCOMPATIBILITY.
  3660 
  3661 - Locale interpretation commands no longer attempt to simplify goal.
  3662 INCOMPATIBILITY: in rare situations the generated goal differs.  Use
  3663 methods intro_locales and unfold_locales to clarify.
  3664 
  3665 - Locale interpretation commands no longer accept interpretation
  3666 attributes.  INCOMPATIBILITY.
  3667 
  3668 * Class declaration: so-called "base sort" must not be given in import
  3669 list any longer, but is inferred from the specification.  Particularly
  3670 in HOL, write
  3671 
  3672     class foo = ...
  3673 
  3674 instead of
  3675 
  3676     class foo = type + ...
  3677 
  3678 * Class target: global versions of theorems stemming do not carry a
  3679 parameter prefix any longer.  INCOMPATIBILITY.
  3680 
  3681 * Class 'instance' command no longer accepts attached definitions.
  3682 INCOMPATIBILITY, use proper 'instantiation' target instead.
  3683 
  3684 * Recovered hiding of consts, which was accidentally broken in
  3685 Isabelle2007.  Potential INCOMPATIBILITY, ``hide const c'' really
  3686 makes c inaccessible; consider using ``hide (open) const c'' instead.
  3687 
  3688 * Slightly more coherent Pure syntax, with updated documentation in
  3689 isar-ref manual.  Removed locales meta_term_syntax and
  3690 meta_conjunction_syntax: TERM and &&& (formerly &&) are now permanent,
  3691 INCOMPATIBILITY in rare situations.  Note that &&& should not be used
  3692 directly in regular applications.
  3693 
  3694 * There is a new syntactic category "float_const" for signed decimal
  3695 fractions (e.g. 123.45 or -123.45).
  3696 
  3697 * Removed exotic 'token_translation' command.  INCOMPATIBILITY, use ML
  3698 interface with 'setup' command instead.
  3699 
  3700 * Command 'local_setup' is similar to 'setup', but operates on a local
  3701 theory context.
  3702 
  3703 * The 'axiomatization' command now only works within a global theory
  3704 context.  INCOMPATIBILITY.
  3705 
  3706 * Goal-directed proof now enforces strict proof irrelevance wrt. sort
  3707 hypotheses.  Sorts required in the course of reasoning need to be
  3708 covered by the constraints in the initial statement, completed by the
  3709 type instance information of the background theory.  Non-trivial sort
  3710 hypotheses, which rarely occur in practice, may be specified via
  3711 vacuous propositions of the form SORT_CONSTRAINT('a::c).  For example:
  3712 
  3713   lemma assumes "SORT_CONSTRAINT('a::empty)" shows False ...
  3714 
  3715 The result contains an implicit sort hypotheses as before --
  3716 SORT_CONSTRAINT premises are eliminated as part of the canonical rule
  3717 normalization.
  3718 
  3719 * Generalized Isar history, with support for linear undo, direct state
  3720 addressing etc.
  3721 
  3722 * Changed defaults for unify configuration options:
  3723 
  3724   unify_trace_bound = 50 (formerly 25)
  3725   unify_search_bound = 60 (formerly 30)
  3726 
  3727 * Different bookkeeping for code equations (INCOMPATIBILITY):
  3728 
  3729   a) On theory merge, the last set of code equations for a particular
  3730      constant is taken (in accordance with the policy applied by other
  3731      parts of the code generator framework).
  3732 
  3733   b) Code equations stemming from explicit declarations (e.g. code
  3734      attribute) gain priority over default code equations stemming
  3735      from definition, primrec, fun etc.
  3736 
  3737 * Keyword 'code_exception' now named 'code_abort'.  INCOMPATIBILITY.
  3738 
  3739 * Unified theorem tables for both code generators.  Thus [code
  3740 func] has disappeared and only [code] remains.  INCOMPATIBILITY.
  3741 
  3742 * Command 'find_consts' searches for constants based on type and name
  3743 patterns, e.g.
  3744 
  3745     find_consts "_ => bool"
  3746 
  3747 By default, matching is against subtypes, but it may be restricted to
  3748 the whole type.  Searching by name is possible.  Multiple queries are
  3749 conjunctive and queries may be negated by prefixing them with a
  3750 hyphen:
  3751 
  3752     find_consts strict: "_ => bool" name: "Int" -"int => int"
  3753 
  3754 * New 'find_theorems' criterion "solves" matches theorems that
  3755 directly solve the current goal (modulo higher-order unification).
  3756 
  3757 * Auto solve feature for main theorem statements: whenever a new goal
  3758 is stated, "find_theorems solves" is called; any theorems that could
  3759 solve the lemma directly are listed as part of the goal state.
  3760 Cf. associated options in Proof General Isabelle settings menu,
  3761 enabled by default, with reasonable timeout for pathological cases of
  3762 higher-order unification.
  3763 
  3764 
  3765 *** Document preparation ***
  3766 
  3767 * Antiquotation @{lemma} now imitates a regular terminal proof,
  3768 demanding keyword 'by' and supporting the full method expression
  3769 syntax just like the Isar command 'by'.
  3770 
  3771 
  3772 *** HOL ***
  3773 
  3774 * Integrated main parts of former image HOL-Complex with HOL.  Entry
  3775 points Main and Complex_Main remain as before.
  3776 
  3777 * Logic image HOL-Plain provides a minimal HOL with the most important
  3778 tools available (inductive, datatype, primrec, ...).  This facilitates
  3779 experimentation and tool development.  Note that user applications
  3780 (and library theories) should never refer to anything below theory
  3781 Main, as before.
  3782 
  3783 * Logic image HOL-Main stops at theory Main, and thus facilitates
  3784 experimentation due to shorter build times.
  3785 
  3786 * Logic image HOL-NSA contains theories of nonstandard analysis which
  3787 were previously part of former HOL-Complex.  Entry point Hyperreal
  3788 remains valid, but theories formerly using Complex_Main should now use
  3789 new entry point Hypercomplex.
  3790 
  3791 * Generic ATP manager for Sledgehammer, based on ML threads instead of
  3792 Posix processes.  Avoids potentially expensive forking of the ML
  3793 process.  New thread-based implementation also works on non-Unix
  3794 platforms (Cygwin).  Provers are no longer hardwired, but defined
  3795 within the theory via plain ML wrapper functions.  Basic Sledgehammer
  3796 commands are covered in the isar-ref manual.
  3797 
  3798 * Wrapper scripts for remote SystemOnTPTP service allows to use
  3799 sledgehammer without local ATP installation (Vampire etc.). Other
  3800 provers may be included via suitable ML wrappers, see also
  3801 src/HOL/ATP_Linkup.thy.
  3802 
  3803 * ATP selection (E/Vampire/Spass) is now via Proof General's settings
  3804 menu.
  3805 
  3806 * The metis method no longer fails because the theorem is too trivial
  3807 (contains the empty clause).
  3808 
  3809 * The metis method now fails in the usual manner, rather than raising
  3810 an exception, if it determines that it cannot prove the theorem.
  3811 
  3812 * Method "coherent" implements a prover for coherent logic (see also
  3813 src/Tools/coherent.ML).
  3814 
  3815 * Constants "undefined" and "default" replace "arbitrary".  Usually
  3816 "undefined" is the right choice to replace "arbitrary", though
  3817 logically there is no difference.  INCOMPATIBILITY.
  3818 
  3819 * Command "value" now integrates different evaluation mechanisms.  The
  3820 result of the first successful evaluation mechanism is printed.  In
  3821 square brackets a particular named evaluation mechanisms may be
  3822 specified (currently, [SML], [code] or [nbe]).  See further
  3823 src/HOL/ex/Eval_Examples.thy.
  3824 
  3825 * Normalization by evaluation now allows non-leftlinear equations.
  3826 Declare with attribute [code nbe].
  3827 
  3828 * Methods "case_tac" and "induct_tac" now refer to the very same rules
  3829 as the structured Isar versions "cases" and "induct", cf. the
  3830 corresponding "cases" and "induct" attributes.  Mutual induction rules
  3831 are now presented as a list of individual projections
  3832 (e.g. foo_bar.inducts for types foo and bar); the old format with
  3833 explicit HOL conjunction is no longer supported.  INCOMPATIBILITY, in
  3834 rare situations a different rule is selected --- notably nested tuple
  3835 elimination instead of former prod.exhaust: use explicit (case_tac t
  3836 rule: prod.exhaust) here.
  3837 
  3838 * Attributes "cases", "induct", "coinduct" support "del" option.
  3839 
  3840 * Removed fact "case_split_thm", which duplicates "case_split".
  3841 
  3842 * The option datatype has been moved to a new theory Option.  Renamed
  3843 option_map to Option.map, and o2s to Option.set, INCOMPATIBILITY.
  3844 
  3845 * New predicate "strict_mono" classifies strict functions on partial
  3846 orders.  With strict functions on linear orders, reasoning about
  3847 (in)equalities is facilitated by theorems "strict_mono_eq",
  3848 "strict_mono_less_eq" and "strict_mono_less".
  3849 
  3850 * Some set operations are now proper qualified constants with
  3851 authentic syntax.  INCOMPATIBILITY:
  3852 
  3853     op Int ~>   Set.Int
  3854     op Un ~>    Set.Un
  3855     INTER ~>    Set.INTER
  3856     UNION ~>    Set.UNION
  3857     Inter ~>    Set.Inter
  3858     Union ~>    Set.Union
  3859     {} ~>       Set.empty
  3860     UNIV ~>     Set.UNIV
  3861 
  3862 * Class complete_lattice with operations Inf, Sup, INFI, SUPR now in
  3863 theory Set.
  3864 
  3865 * Auxiliary class "itself" has disappeared -- classes without any
  3866 parameter are treated as expected by the 'class' command.
  3867 
  3868 * Leibnitz's Series for Pi and the arcus tangens and logarithm series.
  3869 
  3870 * Common decision procedures (Cooper, MIR, Ferrack, Approximation,
  3871 Dense_Linear_Order) are now in directory HOL/Decision_Procs.
  3872 
  3873 * Theory src/HOL/Decision_Procs/Approximation provides the new proof
  3874 method "approximation".  It proves formulas on real values by using
  3875 interval arithmetic.  In the formulas are also the transcendental
  3876 functions sin, cos, tan, atan, ln, exp and the constant pi are
  3877 allowed. For examples see
  3878 src/HOL/Descision_Procs/ex/Approximation_Ex.thy.
  3879 
  3880 * Theory "Reflection" now resides in HOL/Library.
  3881 
  3882 * Entry point to Word library now simply named "Word".
  3883 INCOMPATIBILITY.
  3884 
  3885 * Made source layout more coherent with logical distribution
  3886 structure:
  3887 
  3888     src/HOL/Library/RType.thy ~> src/HOL/Typerep.thy
  3889     src/HOL/Library/Code_Message.thy ~> src/HOL/
  3890     src/HOL/Library/GCD.thy ~> src/HOL/
  3891     src/HOL/Library/Order_Relation.thy ~> src/HOL/
  3892     src/HOL/Library/Parity.thy ~> src/HOL/
  3893     src/HOL/Library/Univ_Poly.thy ~> src/HOL/
  3894     src/HOL/Real/ContNotDenum.thy ~> src/HOL/Library/
  3895     src/HOL/Real/Lubs.thy ~> src/HOL/
  3896     src/HOL/Real/PReal.thy ~> src/HOL/
  3897     src/HOL/Real/Rational.thy ~> src/HOL/
  3898     src/HOL/Real/RComplete.thy ~> src/HOL/
  3899     src/HOL/Real/RealDef.thy ~> src/HOL/
  3900     src/HOL/Real/RealPow.thy ~> src/HOL/
  3901     src/HOL/Real/Real.thy ~> src/HOL/
  3902     src/HOL/Complex/Complex_Main.thy ~> src/HOL/
  3903     src/HOL/Complex/Complex.thy ~> src/HOL/
  3904     src/HOL/Complex/FrechetDeriv.thy ~> src/HOL/Library/
  3905     src/HOL/Complex/Fundamental_Theorem_Algebra.thy ~> src/HOL/Library/
  3906     src/HOL/Hyperreal/Deriv.thy ~> src/HOL/
  3907     src/HOL/Hyperreal/Fact.thy ~> src/HOL/
  3908     src/HOL/Hyperreal/Integration.thy ~> src/HOL/
  3909     src/HOL/Hyperreal/Lim.thy ~> src/HOL/
  3910     src/HOL/Hyperreal/Ln.thy ~> src/HOL/
  3911     src/HOL/Hyperreal/Log.thy ~> src/HOL/
  3912     src/HOL/Hyperreal/MacLaurin.thy ~> src/HOL/
  3913     src/HOL/Hyperreal/NthRoot.thy ~> src/HOL/
  3914     src/HOL/Hyperreal/Series.thy ~> src/HOL/
  3915     src/HOL/Hyperreal/SEQ.thy ~> src/HOL/
  3916     src/HOL/Hyperreal/Taylor.thy ~> src/HOL/
  3917     src/HOL/Hyperreal/Transcendental.thy ~> src/HOL/
  3918     src/HOL/Real/Float ~> src/HOL/Library/
  3919     src/HOL/Real/HahnBanach ~> src/HOL/HahnBanach
  3920     src/HOL/Real/RealVector.thy ~> src/HOL/
  3921 
  3922     src/HOL/arith_data.ML ~> src/HOL/Tools
  3923     src/HOL/hologic.ML ~> src/HOL/Tools
  3924     src/HOL/simpdata.ML ~> src/HOL/Tools
  3925     src/HOL/int_arith1.ML ~> src/HOL/Tools/int_arith.ML
  3926     src/HOL/int_factor_simprocs.ML ~> src/HOL/Tools
  3927     src/HOL/nat_simprocs.ML ~> src/HOL/Tools
  3928     src/HOL/Real/float_arith.ML ~> src/HOL/Tools
  3929     src/HOL/Real/float_syntax.ML ~> src/HOL/Tools
  3930     src/HOL/Real/rat_arith.ML ~> src/HOL/Tools
  3931     src/HOL/Real/real_arith.ML ~> src/HOL/Tools
  3932 
  3933     src/HOL/Library/Array.thy ~> src/HOL/Imperative_HOL
  3934     src/HOL/Library/Heap_Monad.thy ~> src/HOL/Imperative_HOL
  3935     src/HOL/Library/Heap.thy ~> src/HOL/Imperative_HOL
  3936     src/HOL/Library/Imperative_HOL.thy ~> src/HOL/Imperative_HOL
  3937     src/HOL/Library/Ref.thy ~> src/HOL/Imperative_HOL
  3938     src/HOL/Library/Relational.thy ~> src/HOL/Imperative_HOL
  3939 
  3940 * If methods "eval" and "evaluation" encounter a structured proof
  3941 state with !!/==>, only the conclusion is evaluated to True (if
  3942 possible), avoiding strange error messages.
  3943 
  3944 * Method "sizechange" automates termination proofs using (a
  3945 modification of) the size-change principle.  Requires SAT solver.  See
  3946 src/HOL/ex/Termination.thy for examples.
  3947 
  3948 * Simplifier: simproc for let expressions now unfolds if bound
  3949 variable occurs at most once in let expression body.  INCOMPATIBILITY.
  3950 
  3951 * Method "arith": Linear arithmetic now ignores all inequalities when
  3952 fast_arith_neq_limit is exceeded, instead of giving up entirely.
  3953 
  3954 * New attribute "arith" for facts that should always be used
  3955 automatically by arithmetic. It is intended to be used locally in
  3956 proofs, e.g.
  3957 
  3958   assumes [arith]: "x > 0"
  3959 
  3960 Global usage is discouraged because of possible performance impact.
  3961 
  3962 * New classes "top" and "bot" with corresponding operations "top" and
  3963 "bot" in theory Orderings; instantiation of class "complete_lattice"
  3964 requires instantiation of classes "top" and "bot".  INCOMPATIBILITY.
  3965 
  3966 * Changed definition lemma "less_fun_def" in order to provide an
  3967 instance for preorders on functions; use lemma "less_le" instead.
  3968 INCOMPATIBILITY.
  3969 
  3970 * Theory Orderings: class "wellorder" moved here, with explicit
  3971 induction rule "less_induct" as assumption.  For instantiation of
  3972 "wellorder" by means of predicate "wf", use rule wf_wellorderI.
  3973 INCOMPATIBILITY.
  3974 
  3975 * Theory Orderings: added class "preorder" as superclass of "order".
  3976 INCOMPATIBILITY: Instantiation proofs for order, linorder
  3977 etc. slightly changed.  Some theorems named order_class.* now named
  3978 preorder_class.*.
  3979 
  3980 * Theory Relation: renamed "refl" to "refl_on", "reflexive" to "refl,
  3981 "diag" to "Id_on".
  3982 
  3983 * Theory Finite_Set: added a new fold combinator of type
  3984 
  3985   ('a => 'b => 'b) => 'b => 'a set => 'b
  3986 
  3987 Occasionally this is more convenient than the old fold combinator
  3988 which is now defined in terms of the new one and renamed to
  3989 fold_image.
  3990 
  3991 * Theories Ring_and_Field and OrderedGroup: The lemmas "group_simps"
  3992 and "ring_simps" have been replaced by "algebra_simps" (which can be
  3993 extended with further lemmas!).  At the moment both still exist but
  3994 the former will disappear at some point.
  3995 
  3996 * Theory Power: Lemma power_Suc is now declared as a simp rule in
  3997 class recpower.  Type-specific simp rules for various recpower types
  3998 have been removed.  INCOMPATIBILITY, rename old lemmas as follows:
  3999 
  4000 rat_power_0    -> power_0
  4001 rat_power_Suc  -> power_Suc
  4002 realpow_0      -> power_0
  4003 realpow_Suc    -> power_Suc
  4004 complexpow_0   -> power_0
  4005 complexpow_Suc -> power_Suc
  4006 power_poly_0   -> power_0
  4007 power_poly_Suc -> power_Suc
  4008 
  4009 * Theories Ring_and_Field and Divides: Definition of "op dvd" has been
  4010 moved to separate class dvd in Ring_and_Field; a couple of lemmas on
  4011 dvd has been generalized to class comm_semiring_1.  Likewise a bunch
  4012 of lemmas from Divides has been generalized from nat to class
  4013 semiring_div.  INCOMPATIBILITY.  This involves the following theorem
  4014 renames resulting from duplicate elimination:
  4015 
  4016     dvd_def_mod ~>          dvd_eq_mod_eq_0
  4017     zero_dvd_iff ~>         dvd_0_left_iff
  4018     dvd_0 ~>                dvd_0_right
  4019     DIVISION_BY_ZERO_DIV ~> div_by_0
  4020     DIVISION_BY_ZERO_MOD ~> mod_by_0
  4021     mult_div ~>             div_mult_self2_is_id
  4022     mult_mod ~>             mod_mult_self2_is_0
  4023 
  4024 * Theory IntDiv: removed many lemmas that are instances of class-based
  4025 generalizations (from Divides and Ring_and_Field).  INCOMPATIBILITY,
  4026 rename old lemmas as follows:
  4027 
  4028 dvd_diff               -> nat_dvd_diff
  4029 dvd_zminus_iff         -> dvd_minus_iff
  4030 mod_add1_eq            -> mod_add_eq
  4031 mod_mult1_eq           -> mod_mult_right_eq
  4032 mod_mult1_eq'          -> mod_mult_left_eq
  4033 mod_mult_distrib_mod   -> mod_mult_eq
  4034 nat_mod_add_left_eq    -> mod_add_left_eq
  4035 nat_mod_add_right_eq   -> mod_add_right_eq
  4036 nat_mod_div_trivial    -> mod_div_trivial
  4037 nat_mod_mod_trivial    -> mod_mod_trivial
  4038 zdiv_zadd_self1        -> div_add_self1
  4039 zdiv_zadd_self2        -> div_add_self2
  4040 zdiv_zmult_self1       -> div_mult_self2_is_id
  4041 zdiv_zmult_self2       -> div_mult_self1_is_id
  4042 zdvd_triv_left         -> dvd_triv_left
  4043 zdvd_triv_right        -> dvd_triv_right
  4044 zdvd_zmult_cancel_disj -> dvd_mult_cancel_left
  4045 zmod_eq0_zdvd_iff      -> dvd_eq_mod_eq_0[symmetric]
  4046 zmod_zadd_left_eq      -> mod_add_left_eq
  4047 zmod_zadd_right_eq     -> mod_add_right_eq
  4048 zmod_zadd_self1        -> mod_add_self1
  4049 zmod_zadd_self2        -> mod_add_self2
  4050 zmod_zadd1_eq          -> mod_add_eq
  4051 zmod_zdiff1_eq         -> mod_diff_eq
  4052 zmod_zdvd_zmod         -> mod_mod_cancel
  4053 zmod_zmod_cancel       -> mod_mod_cancel
  4054 zmod_zmult_self1       -> mod_mult_self2_is_0
  4055 zmod_zmult_self2       -> mod_mult_self1_is_0
  4056 zmod_1                 -> mod_by_1
  4057 zdiv_1                 -> div_by_1
  4058 zdvd_abs1              -> abs_dvd_iff
  4059 zdvd_abs2              -> dvd_abs_iff
  4060 zdvd_refl              -> dvd_refl
  4061 zdvd_trans             -> dvd_trans
  4062 zdvd_zadd              -> dvd_add
  4063 zdvd_zdiff             -> dvd_diff
  4064 zdvd_zminus_iff        -> dvd_minus_iff
  4065 zdvd_zminus2_iff       -> minus_dvd_iff
  4066 zdvd_zmultD            -> dvd_mult_right
  4067 zdvd_zmultD2           -> dvd_mult_left
  4068 zdvd_zmult_mono        -> mult_dvd_mono
  4069 zdvd_0_right           -> dvd_0_right
  4070 zdvd_0_left            -> dvd_0_left_iff
  4071 zdvd_1_left            -> one_dvd
  4072 zminus_dvd_iff         -> minus_dvd_iff
  4073 
  4074 * Theory Rational: 'Fract k 0' now equals '0'.  INCOMPATIBILITY.
  4075 
  4076 * The real numbers offer decimal input syntax: 12.34 is translated
  4077 into 1234/10^2. This translation is not reversed upon output.
  4078 
  4079 * Theory Library/Polynomial defines an abstract type 'a poly of
  4080 univariate polynomials with coefficients of type 'a.  In addition to
  4081 the standard ring operations, it also supports div and mod.  Code
  4082 generation is also supported, using list-style constructors.
  4083 
  4084 * Theory Library/Inner_Product defines a class of real_inner for real
  4085 inner product spaces, with an overloaded operation inner :: 'a => 'a
  4086 => real.  Class real_inner is a subclass of real_normed_vector from
  4087 theory RealVector.
  4088 
  4089 * Theory Library/Product_Vector provides instances for the product
  4090 type 'a * 'b of several classes from RealVector and Inner_Product.
  4091 Definitions of addition, subtraction, scalar multiplication, norms,
  4092 and inner products are included.
  4093 
  4094 * Theory Library/Bit defines the field "bit" of integers modulo 2.  In
  4095 addition to the field operations, numerals and case syntax are also
  4096 supported.
  4097 
  4098 * Theory Library/Diagonalize provides constructive version of Cantor's
  4099 first diagonalization argument.
  4100 
  4101 * Theory Library/GCD: Curried operations gcd, lcm (for nat) and zgcd,
  4102 zlcm (for int); carried together from various gcd/lcm developements in
  4103 the HOL Distribution.  Constants zgcd and zlcm replace former igcd and
  4104 ilcm; corresponding theorems renamed accordingly.  INCOMPATIBILITY,
  4105 may recover tupled syntax as follows:
  4106 
  4107     hide (open) const gcd
  4108     abbreviation gcd where
  4109       "gcd == (%(a, b). GCD.gcd a b)"
  4110     notation (output)
  4111       GCD.gcd ("gcd '(_, _')")
  4112 
  4113 The same works for lcm, zgcd, zlcm.
  4114 
  4115 * Theory Library/Nat_Infinity: added addition, numeral syntax and more
  4116 instantiations for algebraic structures.  Removed some duplicate
  4117 theorems.  Changes in simp rules.  INCOMPATIBILITY.
  4118 
  4119 * ML antiquotation @{code} takes a constant as argument and generates
  4120 corresponding code in background and inserts name of the corresponding
  4121 resulting ML value/function/datatype constructor binding in place.
  4122 All occurrences of @{code} with a single ML block are generated
  4123 simultaneously.  Provides a generic and safe interface for
  4124 instrumentalizing code generation.  See
  4125 src/HOL/Decision_Procs/Ferrack.thy for a more ambitious application.
  4126 In future you ought to refrain from ad-hoc compiling generated SML
  4127 code on the ML toplevel.  Note that (for technical reasons) @{code}
  4128 cannot refer to constants for which user-defined serializations are
  4129 set.  Refer to the corresponding ML counterpart directly in that
  4130 cases.
  4131 
  4132 * Command 'rep_datatype': instead of theorem names the command now
  4133 takes a list of terms denoting the constructors of the type to be
  4134 represented as datatype.  The characteristic theorems have to be
  4135 proven.  INCOMPATIBILITY.  Also observe that the following theorems
  4136 have disappeared in favour of existing ones:
  4137 
  4138     unit_induct                 ~> unit.induct
  4139     prod_induct                 ~> prod.induct
  4140     sum_induct                  ~> sum.induct
  4141     Suc_Suc_eq                  ~> nat.inject
  4142     Suc_not_Zero Zero_not_Suc   ~> nat.distinct
  4143 
  4144 
  4145 *** HOL-Algebra ***
  4146 
  4147 * New locales for orders and lattices where the equivalence relation
  4148 is not restricted to equality.  INCOMPATIBILITY: all order and lattice
  4149 locales use a record structure with field eq for the equivalence.
  4150 
  4151 * New theory of factorial domains.
  4152 
  4153 * Units_l_inv and Units_r_inv are now simp rules by default.
  4154 INCOMPATIBILITY.  Simplifier proof that require deletion of l_inv
  4155 and/or r_inv will now also require deletion of these lemmas.
  4156 
  4157 * Renamed the following theorems, INCOMPATIBILITY:
  4158 
  4159 UpperD ~> Upper_memD
  4160 LowerD ~> Lower_memD
  4161 least_carrier ~> least_closed
  4162 greatest_carrier ~> greatest_closed
  4163 greatest_Lower_above ~> greatest_Lower_below
  4164 one_zero ~> carrier_one_zero
  4165 one_not_zero ~> carrier_one_not_zero  (collision with assumption)
  4166 
  4167 
  4168 *** HOL-Nominal ***
  4169 
  4170 * Nominal datatypes can now contain type-variables.
  4171 
  4172 * Commands 'nominal_inductive' and 'equivariance' work with local
  4173 theory targets.
  4174 
  4175 * Nominal primrec can now works with local theory targets and its
  4176 specification syntax now conforms to the general format as seen in
  4177 'inductive' etc.
  4178 
  4179 * Method "perm_simp" honours the standard simplifier attributes
  4180 (no_asm), (no_asm_use) etc.
  4181 
  4182 * The new predicate #* is defined like freshness, except that on the
  4183 left hand side can be a set or list of atoms.
  4184 
  4185 * Experimental command 'nominal_inductive2' derives strong induction
  4186 principles for inductive definitions.  In contrast to
  4187 'nominal_inductive', which can only deal with a fixed number of
  4188 binders, it can deal with arbitrary expressions standing for sets of
  4189 atoms to be avoided.  The only inductive definition we have at the
  4190 moment that needs this generalisation is the typing rule for Lets in
  4191 the algorithm W:
  4192 
  4193  Gamma |- t1 : T1   (x,close Gamma T1)::Gamma |- t2 : T2   x#Gamma
  4194  -----------------------------------------------------------------
  4195          Gamma |- Let x be t1 in t2 : T2
  4196 
  4197 In this rule one wants to avoid all the binders that are introduced by
  4198 "close Gamma T1".  We are looking for other examples where this
  4199 feature might be useful.  Please let us know.
  4200 
  4201 
  4202 *** HOLCF ***
  4203 
  4204 * Reimplemented the simplification procedure for proving continuity
  4205 subgoals.  The new simproc is extensible; users can declare additional
  4206 continuity introduction rules with the attribute [cont2cont].
  4207 
  4208 * The continuity simproc now uses a different introduction rule for
  4209 solving continuity subgoals on terms with lambda abstractions.  In
  4210 some rare cases the new simproc may fail to solve subgoals that the
  4211 old one could solve, and "simp add: cont2cont_LAM" may be necessary.
  4212 Potential INCOMPATIBILITY.
  4213 
  4214 * Command 'fixrec': specification syntax now conforms to the general
  4215 format as seen in 'inductive' etc.  See src/HOLCF/ex/Fixrec_ex.thy for
  4216 examples.  INCOMPATIBILITY.
  4217 
  4218 
  4219 *** ZF ***
  4220 
  4221 * Proof of Zorn's Lemma for partial orders.
  4222 
  4223 
  4224 *** ML ***
  4225 
  4226 * Multithreading for Poly/ML 5.1/5.2 is no longer supported, only for
  4227 Poly/ML 5.2.1 or later.  Important note: the TimeLimit facility
  4228 depends on multithreading, so timouts will not work before Poly/ML
  4229 5.2.1!
  4230 
  4231 * High-level support for concurrent ML programming, see
  4232 src/Pure/Cuncurrent.  The data-oriented model of "future values" is
  4233 particularly convenient to organize independent functional
  4234 computations.  The concept of "synchronized variables" provides a
  4235 higher-order interface for components with shared state, avoiding the
  4236 delicate details of mutexes and condition variables.  (Requires
  4237 Poly/ML 5.2.1 or later.)
  4238 
  4239 * ML bindings produced via Isar commands are stored within the Isar
  4240 context (theory or proof).  Consequently, commands like 'use' and 'ML'
  4241 become thread-safe and work with undo as expected (concerning
  4242 top-level bindings, not side-effects on global references).
  4243 INCOMPATIBILITY, need to provide proper Isar context when invoking the
  4244 compiler at runtime; really global bindings need to be given outside a
  4245 theory.  (Requires Poly/ML 5.2 or later.)
  4246 
  4247 * Command 'ML_prf' is analogous to 'ML' but works within a proof
  4248 context.  Top-level ML bindings are stored within the proof context in
  4249 a purely sequential fashion, disregarding the nested proof structure.
  4250 ML bindings introduced by 'ML_prf' are discarded at the end of the
  4251 proof.  (Requires Poly/ML 5.2 or later.)
  4252 
  4253 * Simplified ML attribute and method setup, cf. functions Attrib.setup
  4254 and Method.setup, as well as Isar commands 'attribute_setup' and
  4255 'method_setup'.  INCOMPATIBILITY for 'method_setup', need to simplify
  4256 existing code accordingly, or use plain 'setup' together with old
  4257 Method.add_method.
  4258 
  4259 * Simplified ML oracle interface Thm.add_oracle promotes 'a -> cterm
  4260 to 'a -> thm, while results are always tagged with an authentic oracle
  4261 name.  The Isar command 'oracle' is now polymorphic, no argument type
  4262 is specified.  INCOMPATIBILITY, need to simplify existing oracle code
  4263 accordingly.  Note that extra performance may be gained by producing
  4264 the cterm carefully, avoiding slow Thm.cterm_of.
  4265 
  4266 * Simplified interface for defining document antiquotations via
  4267 ThyOutput.antiquotation, ThyOutput.output, and optionally
  4268 ThyOutput.maybe_pretty_source.  INCOMPATIBILITY, need to simplify user
  4269 antiquotations accordingly, see src/Pure/Thy/thy_output.ML for common
  4270 examples.
  4271 
  4272 * More systematic treatment of long names, abstract name bindings, and
  4273 name space operations.  Basic operations on qualified names have been
  4274 move from structure NameSpace to Long_Name, e.g. Long_Name.base_name,
  4275 Long_Name.append.  Old type bstring has been mostly replaced by
  4276 abstract type binding (see structure Binding), which supports precise
  4277 qualification by packages and local theory targets, as well as proper
  4278 tracking of source positions.  INCOMPATIBILITY, need to wrap old
  4279 bstring values into Binding.name, or better pass through abstract
  4280 bindings everywhere.  See further src/Pure/General/long_name.ML,
  4281 src/Pure/General/binding.ML and src/Pure/General/name_space.ML
  4282 
  4283 * Result facts (from PureThy.note_thms, ProofContext.note_thms,
  4284 LocalTheory.note etc.) now refer to the *full* internal name, not the
  4285 bstring as before.  INCOMPATIBILITY, not detected by ML type-checking!
  4286 
  4287 * Disposed old type and term read functions (Sign.read_def_typ,
  4288 Sign.read_typ, Sign.read_def_terms, Sign.read_term,
  4289 Thm.read_def_cterms, Thm.read_cterm etc.).  INCOMPATIBILITY, should
  4290 use regular Syntax.read_typ, Syntax.read_term, Syntax.read_typ_global,
  4291 Syntax.read_term_global etc.; see also OldGoals.read_term as last
  4292 resort for legacy applications.
  4293 
  4294 * Disposed old declarations, tactics, tactic combinators that refer to
  4295 the simpset or claset of an implicit theory (such as Addsimps,
  4296 Simp_tac, SIMPSET).  INCOMPATIBILITY, should use @{simpset} etc. in
  4297 embedded ML text, or local_simpset_of with a proper context passed as
  4298 explicit runtime argument.
  4299 
  4300 * Rules and tactics that read instantiations (read_instantiate,
  4301 res_inst_tac, thin_tac, subgoal_tac etc.) now demand a proper proof
  4302 context, which is required for parsing and type-checking.  Moreover,
  4303 the variables are specified as plain indexnames, not string encodings
  4304 thereof.  INCOMPATIBILITY.
  4305 
  4306 * Generic Toplevel.add_hook interface allows to analyze the result of
  4307 transactions.  E.g. see src/Pure/ProofGeneral/proof_general_pgip.ML
  4308 for theorem dependency output of transactions resulting in a new
  4309 theory state.
  4310 
  4311 * ML antiquotations: block-structured compilation context indicated by
  4312 \<lbrace> ... \<rbrace>; additional antiquotation forms:
  4313 
  4314   @{binding name}                         - basic name binding
  4315   @{let ?pat = term}                      - term abbreviation (HO matching)
  4316   @{note name = fact}                     - fact abbreviation
  4317   @{thm fact}                             - singleton fact (with attributes)
  4318   @{thms fact}                            - general fact (with attributes)
  4319   @{lemma prop by method}                 - singleton goal
  4320   @{lemma prop by meth1 meth2}            - singleton goal
  4321   @{lemma prop1 ... propN by method}      - general goal
  4322   @{lemma prop1 ... propN by meth1 meth2} - general goal
  4323   @{lemma (open) ...}                     - open derivation
  4324 
  4325 
  4326 *** System ***
  4327 
  4328 * The Isabelle "emacs" tool provides a specific interface to invoke
  4329 Proof General / Emacs, with more explicit failure if that is not
  4330 installed (the old isabelle-interface script silently falls back on
  4331 isabelle-process).  The PROOFGENERAL_HOME setting determines the
  4332 installation location of the Proof General distribution.
  4333 
  4334 * Isabelle/lib/classes/Pure.jar provides basic support to integrate
  4335 the Isabelle process into a JVM/Scala application.  See
  4336 Isabelle/lib/jedit/plugin for a minimal example.  (The obsolete Java
  4337 process wrapper has been discontinued.)
  4338 
  4339 * Added homegrown Isabelle font with unicode layout, see lib/fonts.
  4340 
  4341 * Various status messages (with exact source position information) are
  4342 emitted, if proper markup print mode is enabled.  This allows
  4343 user-interface components to provide detailed feedback on internal
  4344 prover operations.
  4345 
  4346 
  4347 
  4348 New in Isabelle2008 (June 2008)
  4349 -------------------------------
  4350 
  4351 *** General ***
  4352 
  4353 * The Isabelle/Isar Reference Manual (isar-ref) has been reorganized
  4354 and updated, with formally checked references as hyperlinks.
  4355 
  4356 * Theory loader: use_thy (and similar operations) no longer set the
  4357 implicit ML context, which was occasionally hard to predict and in
  4358 conflict with concurrency.  INCOMPATIBILITY, use ML within Isar which
  4359 provides a proper context already.
  4360 
  4361 * Theory loader: old-style ML proof scripts being *attached* to a thy
  4362 file are no longer supported.  INCOMPATIBILITY, regular 'uses' and
  4363 'use' within a theory file will do the job.
  4364 
  4365 * Name space merge now observes canonical order, i.e. the second space
  4366 is inserted into the first one, while existing entries in the first
  4367 space take precedence.  INCOMPATIBILITY in rare situations, may try to
  4368 swap theory imports.
  4369 
  4370 * Syntax: symbol \<chi> is now considered a letter.  Potential
  4371 INCOMPATIBILITY in identifier syntax etc.
  4372 
  4373 * Outer syntax: string tokens no longer admit escaped white space,
  4374 which was an accidental (undocumented) feature.  INCOMPATIBILITY, use
  4375 white space without escapes.
  4376 
  4377 * Outer syntax: string tokens may contain arbitrary character codes
  4378 specified via 3 decimal digits (as in SML).  E.g. "foo\095bar" for
  4379 "foo_bar".
  4380 
  4381 
  4382 *** Pure ***
  4383 
  4384 * Context-dependent token translations.  Default setup reverts locally
  4385 fixed variables, and adds hilite markup for undeclared frees.
  4386 
  4387 * Unused theorems can be found using the new command 'unused_thms'.
  4388 There are three ways of invoking it:
  4389 
  4390 (1) unused_thms
  4391      Only finds unused theorems in the current theory.
  4392 
  4393 (2) unused_thms thy_1 ... thy_n -
  4394      Finds unused theorems in the current theory and all of its ancestors,
  4395      excluding the theories thy_1 ... thy_n and all of their ancestors.
  4396 
  4397 (3) unused_thms thy_1 ... thy_n - thy'_1 ... thy'_m
  4398      Finds unused theorems in the theories thy'_1 ... thy'_m and all of
  4399      their ancestors, excluding the theories thy_1 ... thy_n and all of
  4400      their ancestors.
  4401 
  4402 In order to increase the readability of the list produced by
  4403 unused_thms, theorems that have been created by a particular instance
  4404 of a theory command such as 'inductive' or 'function' are considered
  4405 to belong to the same "group", meaning that if at least one theorem in
  4406 this group is used, the other theorems in the same group are no longer
  4407 reported as unused.  Moreover, if all theorems in the group are
  4408 unused, only one theorem in the group is displayed.
  4409 
  4410 Note that proof objects have to be switched on in order for
  4411 unused_thms to work properly (i.e. !proofs must be >= 1, which is
  4412 usually the case when using Proof General with the default settings).
  4413 
  4414 * Authentic naming of facts disallows ad-hoc overwriting of previous
  4415 theorems within the same name space.  INCOMPATIBILITY, need to remove
  4416 duplicate fact bindings, or even accidental fact duplications.  Note
  4417 that tools may maintain dynamically scoped facts systematically, using
  4418 PureThy.add_thms_dynamic.
  4419 
  4420 * Command 'hide' now allows to hide from "fact" name space as well.
  4421 
  4422 * Eliminated destructive theorem database, simpset, claset, and
  4423 clasimpset.  Potential INCOMPATIBILITY, really need to observe linear
  4424 update of theories within ML code.
  4425 
  4426 * Eliminated theory ProtoPure and CPure, leaving just one Pure theory.
  4427 INCOMPATIBILITY, object-logics depending on former Pure require
  4428 additional setup PureThy.old_appl_syntax_setup; object-logics
  4429 depending on former CPure need to refer to Pure.
  4430 
  4431 * Commands 'use' and 'ML' are now purely functional, operating on
  4432 theory/local_theory.  Removed former 'ML_setup' (on theory), use 'ML'
  4433 instead.  Added 'ML_val' as mere diagnostic replacement for 'ML'.
  4434 INCOMPATIBILITY.
  4435 
  4436 * Command 'setup': discontinued implicit version with ML reference.
  4437 
  4438 * Instantiation target allows for simultaneous specification of class
  4439 instance operations together with an instantiation proof.
  4440 Type-checking phase allows to refer to class operations uniformly.
  4441 See src/HOL/Complex/Complex.thy for an Isar example and
  4442 src/HOL/Library/Eval.thy for an ML example.
  4443 
  4444 * Indexing of literal facts: be more serious about including only
  4445 facts from the visible specification/proof context, but not the
  4446 background context (locale etc.).  Affects `prop` notation and method
  4447 "fact".  INCOMPATIBILITY: need to name facts explicitly in rare
  4448 situations.
  4449 
  4450 * Method "cases", "induct", "coinduct": removed obsolete/undocumented
  4451 "(open)" option, which used to expose internal bound variables to the
  4452 proof text.
  4453 
  4454 * Isar statements: removed obsolete case "rule_context".
  4455 INCOMPATIBILITY, better use explicit fixes/assumes.
  4456 
  4457 * Locale proofs: default proof step now includes 'unfold_locales';
  4458 hence 'proof' without argument may be used to unfold locale
  4459 predicates.
  4460 
  4461 
  4462 *** Document preparation ***
  4463 
  4464 * Simplified pdfsetup.sty: color/hyperref is used unconditionally for
  4465 both pdf and dvi (hyperlinks usually work in xdvi as well); removed
  4466 obsolete thumbpdf setup (contemporary PDF viewers do this on the
  4467 spot); renamed link color from "darkblue" to "linkcolor" (default
  4468 value unchanged, can be redefined via \definecolor); no longer sets
  4469 "a4paper" option (unnecessary or even intrusive).
  4470 
  4471 * Antiquotation @{lemma A method} proves proposition A by the given
  4472 method (either a method name or a method name plus (optional) method
  4473 arguments in parentheses) and prints A just like @{prop A}.
  4474 
  4475 
  4476 *** HOL ***
  4477 
  4478 * New primrec package.  Specification syntax conforms in style to
  4479 definition/function/....  No separate induction rule is provided.  The
  4480 "primrec" command distinguishes old-style and new-style specifications
  4481 by syntax.  The former primrec package is now named OldPrimrecPackage.
  4482 When adjusting theories, beware: constants stemming from new-style
  4483 primrec specifications have authentic syntax.
  4484 
  4485 * Metis prover is now an order of magnitude faster, and also works
  4486 with multithreading.
  4487 
  4488 * Metis: the maximum number of clauses that can be produced from a
  4489 theorem is now given by the attribute max_clauses.  Theorems that
  4490 exceed this number are ignored, with a warning printed.
  4491 
  4492 * Sledgehammer no longer produces structured proofs by default. To
  4493 enable, declare [[sledgehammer_full = true]].  Attributes
  4494 reconstruction_modulus, reconstruction_sorts renamed
  4495 sledgehammer_modulus, sledgehammer_sorts.  INCOMPATIBILITY.
  4496 
  4497 * Method "induct_scheme" derives user-specified induction rules
  4498 from well-founded induction and completeness of patterns. This factors
  4499 out some operations that are done internally by the function package
  4500 and makes them available separately.  See
  4501 src/HOL/ex/Induction_Scheme.thy for examples.
  4502 
  4503 * More flexible generation of measure functions for termination
  4504 proofs: Measure functions can be declared by proving a rule of the
  4505 form "is_measure f" and giving it the [measure_function] attribute.
  4506 The "is_measure" predicate is logically meaningless (always true), and
  4507 just guides the heuristic.  To find suitable measure functions, the
  4508 termination prover sets up the goal "is_measure ?f" of the appropriate
  4509 type and generates all solutions by prolog-style backwards proof using
  4510 the declared rules.
  4511 
  4512 This setup also deals with rules like 
  4513 
  4514   "is_measure f ==> is_measure (list_size f)"
  4515 
  4516 which accommodates nested datatypes that recurse through lists.
  4517 Similar rules are predeclared for products and option types.
  4518 
  4519 * Turned the type of sets "'a set" into an abbreviation for "'a => bool"
  4520 
  4521   INCOMPATIBILITIES:
  4522 
  4523   - Definitions of overloaded constants on sets have to be replaced by
  4524     definitions on => and bool.
  4525 
  4526   - Some definitions of overloaded operators on sets can now be proved
  4527     using the definitions of the operators on => and bool.  Therefore,
  4528     the following theorems have been renamed:
  4529 
  4530       subset_def   -> subset_eq
  4531       psubset_def  -> psubset_eq
  4532       set_diff_def -> set_diff_eq
  4533       Compl_def    -> Compl_eq
  4534       Sup_set_def  -> Sup_set_eq
  4535       Inf_set_def  -> Inf_set_eq
  4536       sup_set_def  -> sup_set_eq
  4537       inf_set_def  -> inf_set_eq
  4538 
  4539   - Due to the incompleteness of the HO unification algorithm, some
  4540     rules such as subst may require manual instantiation, if some of
  4541     the unknowns in the rule is a set.
  4542 
  4543   - Higher order unification and forward proofs:
  4544     The proof pattern
  4545 
  4546       have "P (S::'a set)" <...>
  4547       then have "EX S. P S" ..
  4548 
  4549     no longer works (due to the incompleteness of the HO unification
  4550     algorithm) and must be replaced by the pattern
  4551 
  4552       have "EX S. P S"
  4553       proof
  4554         show "P S" <...>
  4555       qed
  4556 
  4557   - Calculational reasoning with subst (or similar rules):
  4558     The proof pattern
  4559 
  4560       have "P (S::'a set)" <...>
  4561       also have "S = T" <...>
  4562       finally have "P T" .
  4563 
  4564     no longer works (for similar reasons as the previous example) and
  4565     must be replaced by something like
  4566 
  4567       have "P (S::'a set)" <...>
  4568       moreover have "S = T" <...>
  4569       ultimately have "P T" by simp
  4570 
  4571   - Tactics or packages written in ML code:
  4572     Code performing pattern matching on types via
  4573 
  4574       Type ("set", [T]) => ...
  4575 
  4576     must be rewritten. Moreover, functions like strip_type or
  4577     binder_types no longer return the right value when applied to a
  4578     type of the form
  4579 
  4580       T1 => ... => Tn => U => bool
  4581 
  4582     rather than
  4583 
  4584       T1 => ... => Tn => U set
  4585 
  4586 * Merged theories Wellfounded_Recursion, Accessible_Part and
  4587 Wellfounded_Relations to theory Wellfounded.
  4588 
  4589 * Explicit class "eq" for executable equality.  INCOMPATIBILITY.
  4590 
  4591 * Class finite no longer treats UNIV as class parameter.  Use class
  4592 enum from theory Library/Enum instead to achieve a similar effect.
  4593 INCOMPATIBILITY.
  4594 
  4595 * Theory List: rule list_induct2 now has explicitly named cases "Nil"
  4596 and "Cons".  INCOMPATIBILITY.
  4597 
  4598 * HOL (and FOL): renamed variables in rules imp_elim and swap.
  4599 Potential INCOMPATIBILITY.
  4600 
  4601 * Theory Product_Type: duplicated lemmas split_Pair_apply and
  4602 injective_fst_snd removed, use split_eta and prod_eqI instead.
  4603 Renamed upd_fst to apfst and upd_snd to apsnd.  INCOMPATIBILITY.
  4604 
  4605 * Theory Nat: removed redundant lemmas that merely duplicate lemmas of
  4606 the same name in theory Orderings:
  4607 
  4608   less_trans
  4609   less_linear
  4610   le_imp_less_or_eq
  4611   le_less_trans
  4612   less_le_trans
  4613   less_not_sym
  4614   less_asym
  4615 
  4616 Renamed less_imp_le to less_imp_le_nat, and less_irrefl to
  4617 less_irrefl_nat.  Potential INCOMPATIBILITY due to more general types
  4618 and different variable names.
  4619 
  4620 * Library/Option_ord.thy: Canonical order on option type.
  4621 
  4622 * Library/RBT.thy: Red-black trees, an efficient implementation of
  4623 finite maps.
  4624 
  4625 * Library/Countable.thy: Type class for countable types.
  4626 
  4627 * Theory Int: The representation of numerals has changed.  The infix
  4628 operator BIT and the bit datatype with constructors B0 and B1 have
  4629 disappeared.  INCOMPATIBILITY, use "Int.Bit0 x" and "Int.Bit1 y" in
  4630 place of "x BIT bit.B0" and "y BIT bit.B1", respectively.  Theorems
  4631 involving BIT, B0, or B1 have been renamed with "Bit0" or "Bit1"
  4632 accordingly.
  4633 
  4634 * Theory Nat: definition of <= and < on natural numbers no longer
  4635 depend on well-founded relations.  INCOMPATIBILITY.  Definitions
  4636 le_def and less_def have disappeared.  Consider lemmas not_less
  4637 [symmetric, where ?'a = nat] and less_eq [symmetric] instead.
  4638 
  4639 * Theory Finite_Set: locales ACf, ACe, ACIf, ACIfSL and ACIfSLlin
  4640 (whose purpose mainly is for various fold_set functionals) have been
  4641 abandoned in favor of the existing algebraic classes
  4642 ab_semigroup_mult, comm_monoid_mult, ab_semigroup_idem_mult,
  4643 lower_semilattice (resp. upper_semilattice) and linorder.
  4644 INCOMPATIBILITY.
  4645 
  4646 * Theory Transitive_Closure: induct and cases rules now declare proper
  4647 case_names ("base" and "step").  INCOMPATIBILITY.
  4648 
  4649 * Theorem Inductive.lfp_ordinal_induct generalized to complete
  4650 lattices.  The form set-specific version is available as
  4651 Inductive.lfp_ordinal_induct_set.
  4652 
  4653 * Renamed theorems "power.simps" to "power_int.simps".
  4654 INCOMPATIBILITY.
  4655 
  4656 * Class semiring_div provides basic abstract properties of semirings
  4657 with division and modulo operations.  Subsumes former class dvd_mod.
  4658 
  4659 * Merged theories IntDef, Numeral and IntArith into unified theory
  4660 Int.  INCOMPATIBILITY.
  4661 
  4662 * Theory Library/Code_Index: type "index" now represents natural
  4663 numbers rather than integers.  INCOMPATIBILITY.
  4664 
  4665 * New class "uminus" with operation "uminus" (split of from class
  4666 "minus" which now only has operation "minus", binary).
  4667 INCOMPATIBILITY.
  4668 
  4669 * Constants "card", "internal_split", "option_map" now with authentic
  4670 syntax.  INCOMPATIBILITY.
  4671 
  4672 * Definitions subset_def, psubset_def, set_diff_def, Compl_def,
  4673 le_bool_def, less_bool_def, le_fun_def, less_fun_def, inf_bool_def,
  4674 sup_bool_def, Inf_bool_def, Sup_bool_def, inf_fun_def, sup_fun_def,
  4675 Inf_fun_def, Sup_fun_def, inf_set_def, sup_set_def, Inf_set_def,
  4676 Sup_set_def, le_def, less_def, option_map_def now with object
  4677 equality.  INCOMPATIBILITY.
  4678 
  4679 * Records. Removed K_record, and replaced it by pure lambda term
  4680 %x. c. The simplifier setup is now more robust against eta expansion.
  4681 INCOMPATIBILITY: in cases explicitly referring to K_record.
  4682 
  4683 * Library/Multiset: {#a, b, c#} abbreviates {#a#} + {#b#} + {#c#}.
  4684 
  4685 * Library/ListVector: new theory of arithmetic vector operations.
  4686 
  4687 * Library/Order_Relation: new theory of various orderings as sets of
  4688 pairs.  Defines preorders, partial orders, linear orders and
  4689 well-orders on sets and on types.
  4690 
  4691 
  4692 *** ZF ***
  4693 
  4694 * Renamed some theories to allow to loading both ZF and HOL in the
  4695 same session:
  4696 
  4697   Datatype  -> Datatype_ZF
  4698   Inductive -> Inductive_ZF
  4699   Int       -> Int_ZF
  4700   IntDiv    -> IntDiv_ZF
  4701   Nat       -> Nat_ZF
  4702   List      -> List_ZF
  4703   Main      -> Main_ZF
  4704 
  4705 INCOMPATIBILITY: ZF theories that import individual theories below
  4706 Main might need to be adapted.  Regular theory Main is still
  4707 available, as trivial extension of Main_ZF.
  4708 
  4709 
  4710 *** ML ***
  4711 
  4712 * ML within Isar: antiquotation @{const name} or @{const
  4713 name(typargs)} produces statically-checked Const term.
  4714 
  4715 * Functor NamedThmsFun: data is available to the user as dynamic fact
  4716 (of the same name).  Removed obsolete print command.
  4717 
  4718 * Removed obsolete "use_legacy_bindings" function.
  4719 
  4720 * The ``print mode'' is now a thread-local value derived from a global
  4721 template (the former print_mode reference), thus access becomes
  4722 non-critical.  The global print_mode reference is for session
  4723 management only; user-code should use print_mode_value,
  4724 print_mode_active, PrintMode.setmp etc.  INCOMPATIBILITY.
  4725 
  4726 * Functions system/system_out provide a robust way to invoke external
  4727 shell commands, with propagation of interrupts (requires Poly/ML
  4728 5.2.1).  Do not use OS.Process.system etc. from the basis library!
  4729 
  4730 
  4731 *** System ***
  4732 
  4733 * Default settings: PROOFGENERAL_OPTIONS no longer impose xemacs ---
  4734 in accordance with Proof General 3.7, which prefers GNU emacs.
  4735 
  4736 * isatool tty runs Isabelle process with plain tty interaction;
  4737 optional line editor may be specified via ISABELLE_LINE_EDITOR
  4738 setting, the default settings attempt to locate "ledit" and "rlwrap".
  4739 
  4740 * isatool browser now works with Cygwin as well, using general
  4741 "javapath" function defined in Isabelle process environment.
  4742 
  4743 * YXML notation provides a simple and efficient alternative to
  4744 standard XML transfer syntax.  See src/Pure/General/yxml.ML and
  4745 isatool yxml as described in the Isabelle system manual.
  4746 
  4747 * JVM class isabelle.IsabelleProcess (located in Isabelle/lib/classes)
  4748 provides general wrapper for managing an Isabelle process in a robust
  4749 fashion, with ``cooked'' output from stdin/stderr.
  4750 
  4751 * Rudimentary Isabelle plugin for jEdit (see Isabelle/lib/jedit),
  4752 based on Isabelle/JVM process wrapper (see Isabelle/lib/classes).
  4753 
  4754 * Removed obsolete THIS_IS_ISABELLE_BUILD feature.  NB: the documented
  4755 way of changing the user's settings is via
  4756 ISABELLE_HOME_USER/etc/settings, which is a fully featured bash
  4757 script.
  4758 
  4759 * Multithreading.max_threads := 0 refers to the number of actual CPU
  4760 cores of the underlying machine, which is a good starting point for
  4761 optimal performance tuning.  The corresponding usedir option -M allows
  4762 "max" as an alias for "0".  WARNING: does not work on certain versions
  4763 of Mac OS (with Poly/ML 5.1).
  4764 
  4765 * isabelle-process: non-ML sessions are run with "nice", to reduce the
  4766 adverse effect of Isabelle flooding interactive front-ends (notably
  4767 ProofGeneral / XEmacs).
  4768 
  4769 
  4770 
  4771 New in Isabelle2007 (November 2007)
  4772 -----------------------------------
  4773 
  4774 *** General ***
  4775 
  4776 * More uniform information about legacy features, notably a
  4777 warning/error of "Legacy feature: ...", depending on the state of the
  4778 tolerate_legacy_features flag (default true). FUTURE INCOMPATIBILITY:
  4779 legacy features will disappear eventually.
  4780 
  4781 * Theory syntax: the header format ``theory A = B + C:'' has been
  4782 discontinued in favour of ``theory A imports B C begin''.  Use isatool
  4783 fixheaders to convert existing theory files.  INCOMPATIBILITY.
  4784 
  4785 * Theory syntax: the old non-Isar theory file format has been
  4786 discontinued altogether.  Note that ML proof scripts may still be used
  4787 with Isar theories; migration is usually quite simple with the ML
  4788 function use_legacy_bindings.  INCOMPATIBILITY.
  4789 
  4790 * Theory syntax: some popular names (e.g. 'class', 'declaration',
  4791 'fun', 'help', 'if') are now keywords.  INCOMPATIBILITY, use double
  4792 quotes.
  4793 
  4794 * Theory loader: be more serious about observing the static theory
  4795 header specifications (including optional directories), but not the
  4796 accidental file locations of previously successful loads.  The strict
  4797 update policy of former update_thy is now already performed by
  4798 use_thy, so the former has been removed; use_thys updates several
  4799 theories simultaneously, just as 'imports' within a theory header
  4800 specification, but without merging the results.  Potential
  4801 INCOMPATIBILITY: may need to refine theory headers and commands
  4802 ROOT.ML which depend on load order.
  4803 
  4804 * Theory loader: optional support for content-based file
  4805 identification, instead of the traditional scheme of full physical
  4806 path plus date stamp; configured by the ISABELLE_FILE_IDENT setting
  4807 (cf. the system manual).  The new scheme allows to work with
  4808 non-finished theories in persistent session images, such that source
  4809 files may be moved later on without requiring reloads.
  4810 
  4811 * Theory loader: old-style ML proof scripts being *attached* to a thy
  4812 file (with the same base name as the theory) are considered a legacy
  4813 feature, which will disappear eventually. Even now, the theory loader
  4814 no longer maintains dependencies on such files.
  4815 
  4816 * Syntax: the scope for resolving ambiguities via type-inference is
  4817 now limited to individual terms, instead of whole simultaneous
  4818 specifications as before. This greatly reduces the complexity of the
  4819 syntax module and improves flexibility by separating parsing and
  4820 type-checking. INCOMPATIBILITY: additional type-constraints (explicit
  4821 'fixes' etc.) are required in rare situations.
  4822 
  4823 * Syntax: constants introduced by new-style packages ('definition',
  4824 'abbreviation' etc.) are passed through the syntax module in
  4825 ``authentic mode''. This means that associated mixfix annotations
  4826 really stick to such constants, independently of potential name space
  4827 ambiguities introduced later on. INCOMPATIBILITY: constants in parse
  4828 trees are represented slightly differently, may need to adapt syntax
  4829 translations accordingly. Use CONST marker in 'translations' and
  4830 @{const_syntax} antiquotation in 'parse_translation' etc.
  4831 
  4832 * Legacy goal package: reduced interface to the bare minimum required
  4833 to keep existing proof scripts running.  Most other user-level
  4834 functions are now part of the OldGoals structure, which is *not* open
  4835 by default (consider isatool expandshort before open OldGoals).
  4836 Removed top_sg, prin, printyp, pprint_term/typ altogether, because
  4837 these tend to cause confusion about the actual goal (!) context being
  4838 used here, which is not necessarily the same as the_context().
  4839 
  4840 * Command 'find_theorems': supports "*" wild-card in "name:"
  4841 criterion; "with_dups" option.  Certain ProofGeneral versions might
  4842 support a specific search form (see ProofGeneral/CHANGES).
  4843 
  4844 * The ``prems limit'' option (cf. ProofContext.prems_limit) is now -1
  4845 by default, which means that "prems" (and also "fixed variables") are
  4846 suppressed from proof state output.  Note that the ProofGeneral
  4847 settings mechanism allows to change and save options persistently, but
  4848 older versions of Isabelle will fail to start up if a negative prems
  4849 limit is imposed.
  4850 
  4851 * Local theory targets may be specified by non-nested blocks of
  4852 ``context/locale/class ... begin'' followed by ``end''.  The body may
  4853 contain definitions, theorems etc., including any derived mechanism
  4854 that has been implemented on top of these primitives.  This concept
  4855 generalizes the existing ``theorem (in ...)'' towards more versatility
  4856 and scalability.
  4857 
  4858 * Proof General interface: proper undo of final 'end' command;
  4859 discontinued Isabelle/classic mode (ML proof scripts).
  4860 
  4861 
  4862 *** Document preparation ***
  4863 
  4864 * Added antiquotation @{theory name} which prints the given name,
  4865 after checking that it refers to a valid ancestor theory in the
  4866 current context.
  4867 
  4868 * Added antiquotations @{ML_type text} and @{ML_struct text} which
  4869 check the given source text as ML type/structure, printing verbatim.
  4870 
  4871 * Added antiquotation @{abbrev "c args"} which prints the abbreviation
  4872 "c args == rhs" given in the current context.  (Any number of
  4873 arguments may be given on the LHS.)
  4874 
  4875 
  4876 *** Pure ***
  4877 
  4878 * The 'class' package offers a combination of axclass and locale to
  4879 achieve Haskell-like type classes in Isabelle.  Definitions and
  4880 theorems within a class context produce both relative results (with
  4881 implicit parameters according to the locale context), and polymorphic
  4882 constants with qualified polymorphism (according to the class
  4883 context).  Within the body context of a 'class' target, a separate
  4884 syntax layer ("user space type system") takes care of converting
  4885 between global polymorphic consts and internal locale representation.
  4886 See src/HOL/ex/Classpackage.thy for examples (as well as main HOL).
  4887 "isatool doc classes" provides a tutorial.
  4888 
  4889 * Generic code generator framework allows to generate executable
  4890 code for ML and Haskell (including Isabelle classes).  A short usage
  4891 sketch:
  4892 
  4893     internal compilation:
  4894         export_code <list of constants (term syntax)> in SML
  4895     writing SML code to a file:
  4896         export_code <list of constants (term syntax)> in SML <filename>
  4897     writing OCaml code to a file:
  4898         export_code <list of constants (term syntax)> in OCaml <filename>
  4899     writing Haskell code to a bunch of files:
  4900         export_code <list of constants (term syntax)> in Haskell <filename>
  4901 
  4902     evaluating closed propositions to True/False using code generation:
  4903         method ``eval''
  4904 
  4905 Reasonable default setup of framework in HOL.
  4906 
  4907 Theorem attributs for selecting and transforming function equations theorems:
  4908 
  4909     [code fun]:        select a theorem as function equation for a specific constant
  4910     [code fun del]:    deselect a theorem as function equation for a specific constant
  4911     [code inline]:     select an equation theorem for unfolding (inlining) in place
  4912     [code inline del]: deselect an equation theorem for unfolding (inlining) in place
  4913 
  4914 User-defined serializations (target in {SML, OCaml, Haskell}):
  4915 
  4916     code_const <and-list of constants (term syntax)>
  4917       {(target) <and-list of const target syntax>}+
  4918 
  4919     code_type <and-list of type constructors>
  4920       {(target) <and-list of type target syntax>}+
  4921 
  4922     code_instance <and-list of instances>
  4923       {(target)}+
  4924         where instance ::= <type constructor> :: <class>
  4925 
  4926     code_class <and_list of classes>
  4927       {(target) <and-list of class target syntax>}+
  4928         where class target syntax ::= <class name> {where {<classop> == <target syntax>}+}?
  4929 
  4930 code_instance and code_class only are effective to target Haskell.
  4931 
  4932 For example usage see src/HOL/ex/Codegenerator.thy and
  4933 src/HOL/ex/Codegenerator_Pretty.thy.  A separate tutorial on code
  4934 generation from Isabelle/HOL theories is available via "isatool doc
  4935 codegen".
  4936 
  4937 * Code generator: consts in 'consts_code' Isar commands are now
  4938 referred to by usual term syntax (including optional type
  4939 annotations).
  4940 
  4941 * Command 'no_translations' removes translation rules from theory
  4942 syntax.
  4943 
  4944 * Overloaded definitions are now actually checked for acyclic
  4945 dependencies.  The overloading scheme is slightly more general than
  4946 that of Haskell98, although Isabelle does not demand an exact
  4947 correspondence to type class and instance declarations.
  4948 INCOMPATIBILITY, use ``defs (unchecked overloaded)'' to admit more
  4949 exotic versions of overloading -- at the discretion of the user!
  4950 
  4951 Polymorphic constants are represented via type arguments, i.e. the
  4952 instantiation that matches an instance against the most general
  4953 declaration given in the signature.  For example, with the declaration
  4954 c :: 'a => 'a => 'a, an instance c :: nat => nat => nat is represented
  4955 as c(nat).  Overloading is essentially simultaneous structural
  4956 recursion over such type arguments.  Incomplete specification patterns
  4957 impose global constraints on all occurrences, e.g. c('a * 'a) on the
  4958 LHS means that more general c('a * 'b) will be disallowed on any RHS.
  4959 Command 'print_theory' outputs the normalized system of recursive
  4960 equations, see section "definitions".
  4961 
  4962 * Configuration options are maintained within the theory or proof
  4963 context (with name and type bool/int/string), providing a very simple
  4964 interface to a poor-man's version of general context data.  Tools may
  4965 declare options in ML (e.g. using Attrib.config_int) and then refer to
  4966 these values using Config.get etc.  Users may change options via an
  4967 associated attribute of the same name.  This form of context
  4968 declaration works particularly well with commands 'declare' or
  4969 'using', for example ``declare [[foo = 42]]''.  Thus it has become
  4970 very easy to avoid global references, which would not observe Isar
  4971 toplevel undo/redo and fail to work with multithreading.
  4972 
  4973 Various global ML references of Pure and HOL have been turned into
  4974 configuration options:
  4975 
  4976   Unify.search_bound		unify_search_bound
  4977   Unify.trace_bound		unify_trace_bound
  4978   Unify.trace_simp		unify_trace_simp
  4979   Unify.trace_types		unify_trace_types
  4980   Simplifier.simp_depth_limit	simp_depth_limit
  4981   Blast.depth_limit		blast_depth_limit
  4982   DatatypeProp.dtK		datatype_distinctness_limit
  4983   fast_arith_neq_limit  	fast_arith_neq_limit
  4984   fast_arith_split_limit	fast_arith_split_limit
  4985 
  4986 * Named collections of theorems may be easily installed as context
  4987 data using the functor NamedThmsFun (see also
  4988 src/Pure/Tools/named_thms.ML).  The user may add or delete facts via
  4989 attributes; there is also a toplevel print command.  This facility is
  4990 just a common case of general context data, which is the preferred way
  4991 for anything more complex than just a list of facts in canonical
  4992 order.
  4993 
  4994 * Isar: command 'declaration' augments a local theory by generic
  4995 declaration functions written in ML.  This enables arbitrary content
  4996 being added to the context, depending on a morphism that tells the
  4997 difference of the original declaration context wrt. the application
  4998 context encountered later on.
  4999 
  5000 * Isar: proper interfaces for simplification procedures.  Command
  5001 'simproc_setup' declares named simprocs (with match patterns, and body
  5002 text in ML).  Attribute "simproc" adds/deletes simprocs in the current
  5003 context.  ML antiquotation @{simproc name} retrieves named simprocs.
  5004 
  5005 * Isar: an extra pair of brackets around attribute declarations
  5006 abbreviates a theorem reference involving an internal dummy fact,
  5007 which will be ignored later --- only the effect of the attribute on
  5008 the background context will persist.  This form of in-place
  5009 declarations is particularly useful with commands like 'declare' and
  5010 'using', for example ``have A using [[simproc a]] by simp''.
  5011 
  5012 * Isar: method "assumption" (and implicit closing of subproofs) now
  5013 takes simple non-atomic goal assumptions into account: after applying
  5014 an assumption as a rule the resulting subgoals are solved by atomic
  5015 assumption steps.  This is particularly useful to finish 'obtain'
  5016 goals, such as "!!x. (!!x. P x ==> thesis) ==> P x ==> thesis",
  5017 without referring to the original premise "!!x. P x ==> thesis" in the
  5018 Isar proof context.  POTENTIAL INCOMPATIBILITY: method "assumption" is
  5019 more permissive.
  5020 
  5021 * Isar: implicit use of prems from the Isar proof context is
  5022 considered a legacy feature.  Common applications like ``have A .''
  5023 may be replaced by ``have A by fact'' or ``note `A`''.  In general,
  5024 referencing facts explicitly here improves readability and
  5025 maintainability of proof texts.
  5026 
  5027 * Isar: improper proof element 'guess' is like 'obtain', but derives
  5028 the obtained context from the course of reasoning!  For example:
  5029 
  5030   assume "EX x y. A x & B y"   -- "any previous fact"
  5031   then guess x and y by clarify
  5032 
  5033 This technique is potentially adventurous, depending on the facts and
  5034 proof tools being involved here.
  5035 
  5036 * Isar: known facts from the proof context may be specified as literal
  5037 propositions, using ASCII back-quote syntax.  This works wherever
  5038 named facts used to be allowed so far, in proof commands, proof
  5039 methods, attributes etc.  Literal facts are retrieved from the context
  5040 according to unification of type and term parameters.  For example,
  5041 provided that "A" and "A ==> B" and "!!x. P x ==> Q x" are known
  5042 theorems in the current context, then these are valid literal facts:
  5043 `A` and `A ==> B` and `!!x. P x ==> Q x" as well as `P a ==> Q a` etc.
  5044 
  5045 There is also a proof method "fact" which does the same composition
  5046 for explicit goal states, e.g. the following proof texts coincide with
  5047 certain special cases of literal facts:
  5048 
  5049   have "A" by fact                 ==  note `A`
  5050   have "A ==> B" by fact           ==  note `A ==> B`
  5051   have "!!x. P x ==> Q x" by fact  ==  note `!!x. P x ==> Q x`
  5052   have "P a ==> Q a" by fact       ==  note `P a ==> Q a`
  5053 
  5054 * Isar: ":" (colon) is no longer a symbolic identifier character in
  5055 outer syntax.  Thus symbolic identifiers may be used without
  5056 additional white space in declarations like this: ``assume *: A''.
  5057 
  5058 * Isar: 'print_facts' prints all local facts of the current context,
  5059 both named and unnamed ones.
  5060 
  5061 * Isar: 'def' now admits simultaneous definitions, e.g.:
  5062 
  5063   def x == "t" and y == "u"
  5064 
  5065 * Isar: added command 'unfolding', which is structurally similar to
  5066 'using', but affects both the goal state and facts by unfolding given
  5067 rewrite rules.  Thus many occurrences of the 'unfold' method or
  5068 'unfolded' attribute may be replaced by first-class proof text.
  5069 
  5070 * Isar: methods 'unfold' / 'fold', attributes 'unfolded' / 'folded',
  5071 and command 'unfolding' now all support object-level equalities
  5072 (potentially conditional).  The underlying notion of rewrite rule is
  5073 analogous to the 'rule_format' attribute, but *not* that of the
  5074 Simplifier (which is usually more generous).
  5075 
  5076 * Isar: the new attribute [rotated n] (default n = 1) rotates the
  5077 premises of a theorem by n. Useful in conjunction with drule.
  5078 
  5079 * Isar: the goal restriction operator [N] (default N = 1) evaluates a
  5080 method expression within a sandbox consisting of the first N
  5081 sub-goals, which need to exist.  For example, ``simp_all [3]''
  5082 simplifies the first three sub-goals, while (rule foo, simp_all)[]
  5083 simplifies all new goals that emerge from applying rule foo to the
  5084 originally first one.
  5085 
  5086 * Isar: schematic goals are no longer restricted to higher-order
  5087 patterns; e.g. ``lemma "?P(?x)" by (rule TrueI)'' now works as
  5088 expected.
  5089 
  5090 * Isar: the conclusion of a long theorem statement is now either
  5091 'shows' (a simultaneous conjunction, as before), or 'obtains'
  5092 (essentially a disjunction of cases with local parameters and
  5093 assumptions).  The latter allows to express general elimination rules
  5094 adequately; in this notation common elimination rules look like this:
  5095 
  5096   lemma exE:    -- "EX x. P x ==> (!!x. P x ==> thesis) ==> thesis"
  5097     assumes "EX x. P x"
  5098     obtains x where "P x"
  5099 
  5100   lemma conjE:  -- "A & B ==> (A ==> B ==> thesis) ==> thesis"
  5101     assumes "A & B"
  5102     obtains A and B
  5103 
  5104   lemma disjE:  -- "A | B ==> (A ==> thesis) ==> (B ==> thesis) ==> thesis"
  5105     assumes "A | B"
  5106     obtains
  5107       A
  5108     | B
  5109 
  5110 The subsequent classical rules even refer to the formal "thesis"
  5111 explicitly:
  5112 
  5113   lemma classical:     -- "(~ thesis ==> thesis) ==> thesis"
  5114     obtains "~ thesis"
  5115 
  5116   lemma Peirce's_Law:  -- "((thesis ==> something) ==> thesis) ==> thesis"
  5117     obtains "thesis ==> something"
  5118 
  5119 The actual proof of an 'obtains' statement is analogous to that of the
  5120 Isar proof element 'obtain', only that there may be several cases.
  5121 Optional case names may be specified in parentheses; these will be
  5122 available both in the present proof and as annotations in the
  5123 resulting rule, for later use with the 'cases' method (cf. attribute
  5124 case_names).
  5125 
  5126 * Isar: the assumptions of a long theorem statement are available as
  5127 "assms" fact in the proof context.  This is more appropriate than the
  5128 (historical) "prems", which refers to all assumptions of the current
  5129 context, including those from the target locale, proof body etc.
  5130 
  5131 * Isar: 'print_statement' prints theorems from the current theory or
  5132 proof context in long statement form, according to the syntax of a
  5133 top-level lemma.
  5134 
  5135 * Isar: 'obtain' takes an optional case name for the local context
  5136 introduction rule (default "that").
  5137 
  5138 * Isar: removed obsolete 'concl is' patterns.  INCOMPATIBILITY, use
  5139 explicit (is "_ ==> ?foo") in the rare cases where this still happens
  5140 to occur.
  5141 
  5142 * Pure: syntax "CONST name" produces a fully internalized constant
  5143 according to the current context.  This is particularly useful for
  5144 syntax translations that should refer to internal constant
  5145 representations independently of name spaces.
  5146 
  5147 * Pure: syntax constant for foo (binder "FOO ") is called "foo_binder"
  5148 instead of "FOO ". This allows multiple binder declarations to coexist
  5149 in the same context.  INCOMPATIBILITY.
  5150 
  5151 * Isar/locales: 'notation' provides a robust interface to the 'syntax'
  5152 primitive that also works in a locale context (both for constants and
  5153 fixed variables). Type declaration and internal syntactic representation
  5154 of given constants retrieved from the context. Likewise, the
  5155 'no_notation' command allows to remove given syntax annotations from the
  5156 current context.
  5157 
  5158 * Isar/locales: new derived specification elements 'axiomatization',
  5159 'definition', 'abbreviation', which support type-inference, admit
  5160 object-level specifications (equality, equivalence).  See also the
  5161 isar-ref manual.  Examples:
  5162 
  5163   axiomatization
  5164     eq  (infix "===" 50) where
  5165     eq_refl: "x === x" and eq_subst: "x === y ==> P x ==> P y"
  5166 
  5167   definition "f x y = x + y + 1"
  5168   definition g where "g x = f x x"
  5169 
  5170   abbreviation
  5171     neq  (infix "=!=" 50) where
  5172     "x =!= y == ~ (x === y)"
  5173 
  5174 These specifications may be also used in a locale context.  Then the
  5175 constants being introduced depend on certain fixed parameters, and the
  5176 constant name is qualified by the locale base name.  An internal
  5177 abbreviation takes care for convenient input and output, making the
  5178 parameters implicit and using the original short name.  See also
  5179 src/HOL/ex/Abstract_NAT.thy for an example of deriving polymorphic
  5180 entities from a monomorphic theory.
  5181 
  5182 Presently, abbreviations are only available 'in' a target locale, but
  5183 not inherited by general import expressions.  Also note that
  5184 'abbreviation' may be used as a type-safe replacement for 'syntax' +
  5185 'translations' in common applications.  The "no_abbrevs" print mode
  5186 prevents folding of abbreviations in term output.
  5187 
  5188 Concrete syntax is attached to specified constants in internal form,
  5189 independently of name spaces.  The parse tree representation is
  5190 slightly different -- use 'notation' instead of raw 'syntax', and
  5191 'translations' with explicit "CONST" markup to accommodate this.
  5192 
  5193 * Pure/Isar: unified syntax for new-style specification mechanisms
  5194 (e.g.  'definition', 'abbreviation', or 'inductive' in HOL) admits
  5195 full type inference and dummy patterns ("_").  For example:
  5196 
  5197   definition "K x _ = x"
  5198 
  5199   inductive conj for A B
  5200   where "A ==> B ==> conj A B"
  5201 
  5202 * Pure: command 'print_abbrevs' prints all constant abbreviations of
  5203 the current context.  Print mode "no_abbrevs" prevents inversion of
  5204 abbreviations on output.
  5205 
  5206 * Isar/locales: improved parameter handling: use of locales "var" and
  5207 "struct" no longer necessary; - parameter renamings are no longer
  5208 required to be injective.  For example, this allows to define
  5209 endomorphisms as locale endom = homom mult mult h.
  5210 
  5211 * Isar/locales: changed the way locales with predicates are defined.
  5212 Instead of accumulating the specification, the imported expression is
  5213 now an interpretation.  INCOMPATIBILITY: different normal form of
  5214 locale expressions.  In particular, in interpretations of locales with
  5215 predicates, goals repesenting already interpreted fragments are not
  5216 removed automatically.  Use methods `intro_locales' and
  5217 `unfold_locales'; see below.
  5218 
  5219 * Isar/locales: new methods `intro_locales' and `unfold_locales'
  5220 provide backward reasoning on locales predicates.  The methods are
  5221 aware of interpretations and discharge corresponding goals.
  5222 `intro_locales' is less aggressive then `unfold_locales' and does not
  5223 unfold predicates to assumptions.
  5224 
  5225 * Isar/locales: the order in which locale fragments are accumulated
  5226 has changed.  This enables to override declarations from fragments due
  5227 to interpretations -- for example, unwanted simp rules.
  5228 
  5229 * Isar/locales: interpretation in theories and proof contexts has been
  5230 extended.  One may now specify (and prove) equations, which are
  5231 unfolded in interpreted theorems.  This is useful for replacing
  5232 defined concepts (constants depending on locale parameters) by
  5233 concepts already existing in the target context.  Example:
  5234 
  5235   interpretation partial_order ["op <= :: [int, int] => bool"]
  5236     where "partial_order.less (op <=) (x::int) y = (x < y)"
  5237 
  5238 Typically, the constant `partial_order.less' is created by a
  5239 definition specification element in the context of locale
  5240 partial_order.
  5241 
  5242 * Method "induct": improved internal context management to support
  5243 local fixes and defines on-the-fly. Thus explicit meta-level
  5244 connectives !!  and ==> are rarely required anymore in inductive goals
  5245 (using object-logic connectives for this purpose has been long
  5246 obsolete anyway). Common proof patterns are explained in
  5247 src/HOL/Induct/Common_Patterns.thy, see also
  5248 src/HOL/Isar_examples/Puzzle.thy and src/HOL/Lambda for realistic
  5249 examples.
  5250 
  5251 * Method "induct": improved handling of simultaneous goals. Instead of
  5252 introducing object-level conjunction, the statement is now split into
  5253 several conclusions, while the corresponding symbolic cases are nested
  5254 accordingly. INCOMPATIBILITY, proofs need to be structured explicitly,
  5255 see src/HOL/Induct/Common_Patterns.thy, for example.
  5256 
  5257 * Method "induct": mutual induction rules are now specified as a list
  5258 of rule sharing the same induction cases. HOL packages usually provide
  5259 foo_bar.inducts for mutually defined items foo and bar (e.g. inductive
  5260 predicates/sets or datatypes). INCOMPATIBILITY, users need to specify
  5261 mutual induction rules differently, i.e. like this:
  5262 
  5263   (induct rule: foo_bar.inducts)
  5264   (induct set: foo bar)
  5265   (induct pred: foo bar)
  5266   (induct type: foo bar)
  5267 
  5268 The ML function ProjectRule.projections turns old-style rules into the
  5269 new format.
  5270 
  5271 * Method "coinduct": dual of induction, see
  5272 src/HOL/Library/Coinductive_List.thy for various examples.
  5273 
  5274 * Method "cases", "induct", "coinduct": the ``(open)'' option is
  5275 considered a legacy feature.
  5276 
  5277 * Attribute "symmetric" produces result with standardized schematic
  5278 variables (index 0).  Potential INCOMPATIBILITY.
  5279 
  5280 * Simplifier: by default the simplifier trace only shows top level
  5281 rewrites now. That is, trace_simp_depth_limit is set to 1 by
  5282 default. Thus there is less danger of being flooded by the trace. The
  5283 trace indicates where parts have been suppressed.
  5284   
  5285 * Provers/classical: removed obsolete classical version of elim_format
  5286 attribute; classical elim/dest rules are now treated uniformly when
  5287 manipulating the claset.
  5288 
  5289 * Provers/classical: stricter checks to ensure that supplied intro,
  5290 dest and elim rules are well-formed; dest and elim rules must have at
  5291 least one premise.
  5292 
  5293 * Provers/classical: attributes dest/elim/intro take an optional
  5294 weight argument for the rule (just as the Pure versions).  Weights are
  5295 ignored by automated tools, but determine the search order of single
  5296 rule steps.
  5297 
  5298 * Syntax: input syntax now supports dummy variable binding "%_. b",
  5299 where the body does not mention the bound variable.  Note that dummy
  5300 patterns implicitly depend on their context of bounds, which makes
  5301 "{_. _}" match any set comprehension as expected.  Potential
  5302 INCOMPATIBILITY -- parse translations need to cope with syntactic
  5303 constant "_idtdummy" in the binding position.
  5304 
  5305 * Syntax: removed obsolete syntactic constant "_K" and its associated
  5306 parse translation.  INCOMPATIBILITY -- use dummy abstraction instead,
  5307 for example "A -> B" => "Pi A (%_. B)".
  5308 
  5309 * Pure: 'class_deps' command visualizes the subclass relation, using
  5310 the graph browser tool.
  5311 
  5312 * Pure: 'print_theory' now suppresses certain internal declarations by
  5313 default; use '!' option for full details.
  5314 
  5315 
  5316 *** HOL ***
  5317 
  5318 * Method "metis" proves goals by applying the Metis general-purpose
  5319 resolution prover (see also http://gilith.com/software/metis/).
  5320 Examples are in the directory MetisExamples.  WARNING: the
  5321 Isabelle/HOL-Metis integration does not yet work properly with
  5322 multi-threading.
  5323   
  5324 * Command 'sledgehammer' invokes external automatic theorem provers as
  5325 background processes.  It generates calls to the "metis" method if
  5326 successful. These can be pasted into the proof.  Users do not have to
  5327 wait for the automatic provers to return.  WARNING: does not really
  5328 work with multi-threading.
  5329 
  5330 * New "auto_quickcheck" feature tests outermost goal statements for
  5331 potential counter-examples.  Controlled by ML references
  5332 auto_quickcheck (default true) and auto_quickcheck_time_limit (default
  5333 5000 milliseconds).  Fails silently if statements is outside of
  5334 executable fragment, or any other codgenerator problem occurs.
  5335 
  5336 * New constant "undefined" with axiom "undefined x = undefined".
  5337 
  5338 * Added class "HOL.eq", allowing for code generation with polymorphic
  5339 equality.
  5340 
  5341 * Some renaming of class constants due to canonical name prefixing in
  5342 the new 'class' package:
  5343 
  5344     HOL.abs ~> HOL.abs_class.abs
  5345     HOL.divide ~> HOL.divide_class.divide
  5346     0 ~> HOL.zero_class.zero
  5347     1 ~> HOL.one_class.one
  5348     op + ~> HOL.plus_class.plus
  5349     op - ~> HOL.minus_class.minus
  5350     uminus ~> HOL.minus_class.uminus
  5351     op * ~> HOL.times_class.times
  5352     op < ~> HOL.ord_class.less
  5353     op <= > HOL.ord_class.less_eq
  5354     Nat.power ~> Power.power_class.power
  5355     Nat.size ~> Nat.size_class.size
  5356     Numeral.number_of ~> Numeral.number_class.number_of
  5357     FixedPoint.Inf ~> Lattices.complete_lattice_class.Inf
  5358     FixedPoint.Sup ~> Lattices.complete_lattice_class.Sup
  5359     Orderings.min ~> Orderings.ord_class.min
  5360     Orderings.max ~> Orderings.ord_class.max
  5361     Divides.op div ~> Divides.div_class.div
  5362     Divides.op mod ~> Divides.div_class.mod
  5363     Divides.op dvd ~> Divides.div_class.dvd
  5364 
  5365 INCOMPATIBILITY.  Adaptions may be required in the following cases:
  5366 
  5367 a) User-defined constants using any of the names "plus", "minus",
  5368 "times", "less" or "less_eq". The standard syntax translations for
  5369 "+", "-" and "*" may go wrong.  INCOMPATIBILITY: use more specific
  5370 names.
  5371 
  5372 b) Variables named "plus", "minus", "times", "less", "less_eq"
  5373 INCOMPATIBILITY: use more specific names.
  5374 
  5375 c) Permutative equations (e.g. "a + b = b + a")
  5376 Since the change of names also changes the order of terms, permutative
  5377 rewrite rules may get applied in a different order. Experience shows
  5378 that this is rarely the case (only two adaptions in the whole Isabelle
  5379 distribution).  INCOMPATIBILITY: rewrite proofs
  5380 
  5381 d) ML code directly refering to constant names
  5382 This in general only affects hand-written proof tactics, simprocs and
  5383 so on.  INCOMPATIBILITY: grep your sourcecode and replace names.
  5384 Consider using @{const_name} antiquotation.
  5385 
  5386 * New class "default" with associated constant "default".
  5387 
  5388 * Function "sgn" is now overloaded and available on int, real, complex
  5389 (and other numeric types), using class "sgn".  Two possible defs of
  5390 sgn are given as equational assumptions in the classes sgn_if and
  5391 sgn_div_norm; ordered_idom now also inherits from sgn_if.
  5392 INCOMPATIBILITY.
  5393 
  5394 * Locale "partial_order" now unified with class "order" (cf. theory
  5395 Orderings), added parameter "less".  INCOMPATIBILITY.
  5396 
  5397 * Renamings in classes "order" and "linorder": facts "refl", "trans" and
  5398 "cases" to "order_refl", "order_trans" and "linorder_cases", to avoid
  5399 clashes with HOL "refl" and "trans".  INCOMPATIBILITY.
  5400 
  5401 * Classes "order" and "linorder": potential INCOMPATIBILITY due to
  5402 changed order of proof goals in instance proofs.
  5403 
  5404 * The transitivity reasoner for partial and linear orders is set up
  5405 for classes "order" and "linorder".  Instances of the reasoner are available
  5406 in all contexts importing or interpreting the corresponding locales.
  5407 Method "order" invokes the reasoner separately; the reasoner
  5408 is also integrated with the Simplifier as a solver.  Diagnostic
  5409 command 'print_orders' shows the available instances of the reasoner
  5410 in the current context.
  5411 
  5412 * Localized monotonicity predicate in theory "Orderings"; integrated
  5413 lemmas max_of_mono and min_of_mono with this predicate.
  5414 INCOMPATIBILITY.
  5415 
  5416 * Formulation of theorem "dense" changed slightly due to integration
  5417 with new class dense_linear_order.
  5418 
  5419 * Uniform lattice theory development in HOL.
  5420 
  5421     constants "meet" and "join" now named "inf" and "sup"
  5422     constant "Meet" now named "Inf"
  5423 
  5424     classes "meet_semilorder" and "join_semilorder" now named
  5425       "lower_semilattice" and "upper_semilattice"
  5426     class "lorder" now named "lattice"
  5427     class "comp_lat" now named "complete_lattice"
  5428 
  5429     Instantiation of lattice classes allows explicit definitions
  5430     for "inf" and "sup" operations (or "Inf" and "Sup" for complete lattices).
  5431 
  5432   INCOMPATIBILITY.  Theorem renames:
  5433 
  5434     meet_left_le            ~> inf_le1
  5435     meet_right_le           ~> inf_le2
  5436     join_left_le            ~> sup_ge1
  5437     join_right_le           ~> sup_ge2
  5438     meet_join_le            ~> inf_sup_ord
  5439     le_meetI                ~> le_infI
  5440     join_leI                ~> le_supI
  5441     le_meet                 ~> le_inf_iff
  5442     le_join                 ~> ge_sup_conv
  5443     meet_idempotent         ~> inf_idem
  5444     join_idempotent         ~> sup_idem
  5445     meet_comm               ~> inf_commute
  5446     join_comm               ~> sup_commute
  5447     meet_leI1               ~> le_infI1
  5448     meet_leI2               ~> le_infI2
  5449     le_joinI1               ~> le_supI1
  5450     le_joinI2               ~> le_supI2
  5451     meet_assoc              ~> inf_assoc
  5452     join_assoc              ~> sup_assoc
  5453     meet_left_comm          ~> inf_left_commute
  5454     meet_left_idempotent    ~> inf_left_idem
  5455     join_left_comm          ~> sup_left_commute
  5456     join_left_idempotent    ~> sup_left_idem
  5457     meet_aci                ~> inf_aci
  5458     join_aci                ~> sup_aci
  5459     le_def_meet             ~> le_iff_inf
  5460     le_def_join             ~> le_iff_sup
  5461     join_absorp2            ~> sup_absorb2
  5462     join_absorp1            ~> sup_absorb1
  5463     meet_absorp1            ~> inf_absorb1
  5464     meet_absorp2            ~> inf_absorb2
  5465     meet_join_absorp        ~> inf_sup_absorb
  5466     join_meet_absorp        ~> sup_inf_absorb
  5467     distrib_join_le         ~> distrib_sup_le
  5468     distrib_meet_le         ~> distrib_inf_le
  5469 
  5470     add_meet_distrib_left   ~> add_inf_distrib_left
  5471     add_join_distrib_left   ~> add_sup_distrib_left
  5472     is_join_neg_meet        ~> is_join_neg_inf
  5473     is_meet_neg_join        ~> is_meet_neg_sup
  5474     add_meet_distrib_right  ~> add_inf_distrib_right
  5475     add_join_distrib_right  ~> add_sup_distrib_right
  5476     add_meet_join_distribs  ~> add_sup_inf_distribs
  5477     join_eq_neg_meet        ~> sup_eq_neg_inf
  5478     meet_eq_neg_join        ~> inf_eq_neg_sup
  5479     add_eq_meet_join        ~> add_eq_inf_sup
  5480     meet_0_imp_0            ~> inf_0_imp_0
  5481     join_0_imp_0            ~> sup_0_imp_0
  5482     meet_0_eq_0             ~> inf_0_eq_0
  5483     join_0_eq_0             ~> sup_0_eq_0
  5484     neg_meet_eq_join        ~> neg_inf_eq_sup
  5485     neg_join_eq_meet        ~> neg_sup_eq_inf
  5486     join_eq_if              ~> sup_eq_if
  5487 
  5488     mono_meet               ~> mono_inf
  5489     mono_join               ~> mono_sup
  5490     meet_bool_eq            ~> inf_bool_eq
  5491     join_bool_eq            ~> sup_bool_eq
  5492     meet_fun_eq             ~> inf_fun_eq
  5493     join_fun_eq             ~> sup_fun_eq
  5494     meet_set_eq             ~> inf_set_eq
  5495     join_set_eq             ~> sup_set_eq
  5496     meet1_iff               ~> inf1_iff
  5497     meet2_iff               ~> inf2_iff
  5498     meet1I                  ~> inf1I
  5499     meet2I                  ~> inf2I
  5500     meet1D1                 ~> inf1D1
  5501     meet2D1                 ~> inf2D1
  5502     meet1D2                 ~> inf1D2
  5503     meet2D2                 ~> inf2D2
  5504     meet1E                  ~> inf1E
  5505     meet2E                  ~> inf2E
  5506     join1_iff               ~> sup1_iff
  5507     join2_iff               ~> sup2_iff
  5508     join1I1                 ~> sup1I1
  5509     join2I1                 ~> sup2I1
  5510     join1I1                 ~> sup1I1
  5511     join2I2                 ~> sup1I2
  5512     join1CI                 ~> sup1CI
  5513     join2CI                 ~> sup2CI
  5514     join1E                  ~> sup1E
  5515     join2E                  ~> sup2E
  5516 
  5517     is_meet_Meet            ~> is_meet_Inf
  5518     Meet_bool_def           ~> Inf_bool_def
  5519     Meet_fun_def            ~> Inf_fun_def
  5520     Meet_greatest           ~> Inf_greatest
  5521     Meet_lower              ~> Inf_lower
  5522     Meet_set_def            ~> Inf_set_def
  5523 
  5524     Sup_def                 ~> Sup_Inf
  5525     Sup_bool_eq             ~> Sup_bool_def
  5526     Sup_fun_eq              ~> Sup_fun_def
  5527     Sup_set_eq              ~> Sup_set_def
  5528 
  5529     listsp_meetI            ~> listsp_infI
  5530     listsp_meet_eq          ~> listsp_inf_eq
  5531 
  5532     meet_min                ~> inf_min
  5533     join_max                ~> sup_max
  5534 
  5535 * Added syntactic class "size"; overloaded constant "size" now has
  5536 type "'a::size ==> bool"
  5537 
  5538 * Internal reorganisation of `size' of datatypes: size theorems
  5539 "foo.size" are no longer subsumed by "foo.simps" (but are still
  5540 simplification rules by default!); theorems "prod.size" now named
  5541 "*.size".
  5542 
  5543 * Class "div" now inherits from class "times" rather than "type".
  5544 INCOMPATIBILITY.
  5545 
  5546 * HOL/Finite_Set: "name-space" locales Lattice, Distrib_lattice,
  5547 Linorder etc.  have disappeared; operations defined in terms of
  5548 fold_set now are named Inf_fin, Sup_fin.  INCOMPATIBILITY.
  5549 
  5550 * HOL/Nat: neq0_conv no longer declared as iff.  INCOMPATIBILITY.
  5551 
  5552 * HOL-Word: New extensive library and type for generic, fixed size
  5553 machine words, with arithmetic, bit-wise, shifting and rotating
  5554 operations, reflection into int, nat, and bool lists, automation for
  5555 linear arithmetic (by automatic reflection into nat or int), including
  5556 lemmas on overflow and monotonicity.  Instantiated to all appropriate
  5557 arithmetic type classes, supporting automatic simplification of
  5558 numerals on all operations.
  5559 
  5560 * Library/Boolean_Algebra: locales for abstract boolean algebras.
  5561 
  5562 * Library/Numeral_Type: numbers as types, e.g. TYPE(32).
  5563 
  5564 * Code generator library theories:
  5565   - Code_Integer represents HOL integers by big integer literals in target
  5566     languages.
  5567   - Code_Char represents HOL characters by character literals in target
  5568     languages.
  5569   - Code_Char_chr like Code_Char, but also offers treatment of character
  5570     codes; includes Code_Integer.
  5571   - Executable_Set allows to generate code for finite sets using lists.
  5572   - Executable_Rat implements rational numbers as triples (sign, enumerator,
  5573     denominator).
  5574   - Executable_Real implements a subset of real numbers, namly those
  5575     representable by rational numbers.
  5576   - Efficient_Nat implements natural numbers by integers, which in general will
  5577     result in higher efficency; pattern matching with 0/Suc is eliminated;
  5578     includes Code_Integer.
  5579   - Code_Index provides an additional datatype index which is mapped to
  5580     target-language built-in integers.
  5581   - Code_Message provides an additional datatype message_string which is isomorphic to
  5582     strings; messages are mapped to target-language strings.
  5583 
  5584 * New package for inductive predicates
  5585 
  5586   An n-ary predicate p with m parameters z_1, ..., z_m can now be defined via
  5587 
  5588     inductive
  5589       p :: "U_1 => ... => U_m => T_1 => ... => T_n => bool"
  5590       for z_1 :: U_1 and ... and z_n :: U_m
  5591     where
  5592       rule_1: "... ==> p z_1 ... z_m t_1_1 ... t_1_n"
  5593     | ...
  5594 
  5595   with full support for type-inference, rather than
  5596 
  5597     consts s :: "U_1 => ... => U_m => (T_1 * ... * T_n) set"
  5598 
  5599     abbreviation p :: "U_1 => ... => U_m => T_1 => ... => T_n => bool"
  5600     where "p z_1 ... z_m x_1 ... x_n == (x_1, ..., x_n) : s z_1 ... z_m"
  5601 
  5602     inductive "s z_1 ... z_m"
  5603     intros
  5604       rule_1: "... ==> (t_1_1, ..., t_1_n) : s z_1 ... z_m"
  5605       ...
  5606 
  5607   For backward compatibility, there is a wrapper allowing inductive
  5608   sets to be defined with the new package via
  5609 
  5610     inductive_set
  5611       s :: "U_1 => ... => U_m => (T_1 * ... * T_n) set"
  5612       for z_1 :: U_1 and ... and z_n :: U_m
  5613     where
  5614       rule_1: "... ==> (t_1_1, ..., t_1_n) : s z_1 ... z_m"
  5615     | ...
  5616 
  5617   or
  5618 
  5619     inductive_set
  5620       s :: "U_1 => ... => U_m => (T_1 * ... * T_n) set"
  5621       and p :: "U_1 => ... => U_m => T_1 => ... => T_n => bool"
  5622       for z_1 :: U_1 and ... and z_n :: U_m
  5623     where
  5624       "p z_1 ... z_m x_1 ... x_n == (x_1, ..., x_n) : s z_1 ... z_m"
  5625     | rule_1: "... ==> p z_1 ... z_m t_1_1 ... t_1_n"
  5626     | ...
  5627 
  5628   if the additional syntax "p ..." is required.
  5629 
  5630   Numerous examples can be found in the subdirectories src/HOL/Auth,
  5631   src/HOL/Bali, src/HOL/Induct, and src/HOL/MicroJava.
  5632 
  5633   INCOMPATIBILITIES:
  5634 
  5635   - Since declaration and definition of inductive sets or predicates
  5636     is no longer separated, abbreviations involving the newly
  5637     introduced sets or predicates must be specified together with the
  5638     introduction rules after the 'where' keyword (see above), rather
  5639     than before the actual inductive definition.
  5640 
  5641   - The variables in induction and elimination rules are now
  5642     quantified in the order of their occurrence in the introduction
  5643     rules, rather than in alphabetical order. Since this may break
  5644     some proofs, these proofs either have to be repaired, e.g. by
  5645     reordering the variables a_i_1 ... a_i_{k_i} in Isar 'case'
  5646     statements of the form
  5647 
  5648       case (rule_i a_i_1 ... a_i_{k_i})
  5649 
  5650     or the old order of quantification has to be restored by explicitly adding
  5651     meta-level quantifiers in the introduction rules, i.e.
  5652 
  5653       | rule_i: "!!a_i_1 ... a_i_{k_i}. ... ==> p z_1 ... z_m t_i_1 ... t_i_n"
  5654 
  5655   - The format of the elimination rules is now
  5656 
  5657       p z_1 ... z_m x_1 ... x_n ==>
  5658         (!!a_1_1 ... a_1_{k_1}. x_1 = t_1_1 ==> ... ==> x_n = t_1_n ==> ... ==> P)
  5659         ==> ... ==> P
  5660 
  5661     for predicates and
  5662 
  5663       (x_1, ..., x_n) : s z_1 ... z_m ==>
  5664         (!!a_1_1 ... a_1_{k_1}. x_1 = t_1_1 ==> ... ==> x_n = t_1_n ==> ... ==> P)
  5665         ==> ... ==> P
  5666 
  5667     for sets rather than
  5668 
  5669       x : s z_1 ... z_m ==>
  5670         (!!a_1_1 ... a_1_{k_1}. x = (t_1_1, ..., t_1_n) ==> ... ==> P)
  5671         ==> ... ==> P
  5672 
  5673     This may require terms in goals to be expanded to n-tuples
  5674     (e.g. using case_tac or simplification with the split_paired_all
  5675     rule) before the above elimination rule is applicable.
  5676 
  5677   - The elimination or case analysis rules for (mutually) inductive
  5678     sets or predicates are now called "p_1.cases" ... "p_k.cases". The
  5679     list of rules "p_1_..._p_k.elims" is no longer available.
  5680 
  5681 * New package "function"/"fun" for general recursive functions,
  5682 supporting mutual and nested recursion, definitions in local contexts,
  5683 more general pattern matching and partiality. See HOL/ex/Fundefs.thy
  5684 for small examples, and the separate tutorial on the function
  5685 package. The old recdef "package" is still available as before, but
  5686 users are encouraged to use the new package.
  5687 
  5688 * Method "lexicographic_order" automatically synthesizes termination
  5689 relations as lexicographic combinations of size measures. 
  5690 
  5691 * Case-expressions allow arbitrary constructor-patterns (including
  5692 "_") and take their order into account, like in functional
  5693 programming.  Internally, this is translated into nested
  5694 case-expressions; missing cases are added and mapped to the predefined
  5695 constant "undefined". In complicated cases printing may no longer show
  5696 the original input but the internal form. Lambda-abstractions allow
  5697 the same form of pattern matching: "% pat1 => e1 | ..." is an
  5698 abbreviation for "%x. case x of pat1 => e1 | ..." where x is a new
  5699 variable.
  5700 
  5701 * IntDef: The constant "int :: nat => int" has been removed; now "int"
  5702 is an abbreviation for "of_nat :: nat => int". The simplification
  5703 rules for "of_nat" have been changed to work like "int" did
  5704 previously.  Potential INCOMPATIBILITY:
  5705   - "of_nat (Suc m)" simplifies to "1 + of_nat m" instead of "of_nat m + 1"
  5706   - of_nat_diff and of_nat_mult are no longer default simp rules
  5707 
  5708 * Method "algebra" solves polynomial equations over (semi)rings using
  5709 Groebner bases. The (semi)ring structure is defined by locales and the
  5710 tool setup depends on that generic context. Installing the method for
  5711 a specific type involves instantiating the locale and possibly adding
  5712 declarations for computation on the coefficients.  The method is
  5713 already instantiated for natural numbers and for the axiomatic class
  5714 of idoms with numerals.  See also the paper by Chaieb and Wenzel at
  5715 CALCULEMUS 2007 for the general principles underlying this
  5716 architecture of context-aware proof-tools.
  5717 
  5718 * Method "ferrack" implements quantifier elimination over
  5719 special-purpose dense linear orders using locales (analogous to
  5720 "algebra"). The method is already installed for class
  5721 {ordered_field,recpower,number_ring} which subsumes real, hyperreal,
  5722 rat, etc.
  5723 
  5724 * Former constant "List.op @" now named "List.append".  Use ML
  5725 antiquotations @{const_name List.append} or @{term " ... @ ... "} to
  5726 circumvent possible incompatibilities when working on ML level.
  5727 
  5728 * primrec: missing cases mapped to "undefined" instead of "arbitrary".
  5729 
  5730 * New function listsum :: 'a list => 'a for arbitrary monoids.
  5731 Special syntax: "SUM x <- xs. f x" (and latex variants)
  5732 
  5733 * New syntax for Haskell-like list comprehension (input only), eg.
  5734 [(x,y). x <- xs, y <- ys, x ~= y], see also src/HOL/List.thy.
  5735 
  5736 * The special syntax for function "filter" has changed from [x :
  5737 xs. P] to [x <- xs. P] to avoid an ambiguity caused by list
  5738 comprehension syntax, and for uniformity.  INCOMPATIBILITY.
  5739 
  5740 * [a..b] is now defined for arbitrary linear orders.  It used to be
  5741 defined on nat only, as an abbreviation for [a..<Suc b]
  5742 INCOMPATIBILITY.
  5743 
  5744 * Renamed lemma "set_take_whileD"  to "set_takeWhileD".
  5745 
  5746 * New functions "sorted" and "sort" in src/HOL/List.thy.
  5747 
  5748 * New lemma collection field_simps (an extension of ring_simps) for
  5749 manipulating (in)equations involving division. Multiplies with all
  5750 denominators that can be proved to be non-zero (in equations) or
  5751 positive/negative (in inequations).
  5752 
  5753 * Lemma collections ring_eq_simps, group_eq_simps and ring_distrib
  5754 have been improved and renamed to ring_simps, group_simps and
  5755 ring_distribs.  Removed lemmas field_xyz in theory Ring_and_Field
  5756 because they were subsumed by lemmas xyz.  INCOMPATIBILITY.
  5757 
  5758 * Theory Library/Commutative_Ring: switched from recdef to function
  5759 package; constants add, mul, pow now curried.  Infix syntax for
  5760 algebraic operations.
  5761 
  5762 * Dropped redundant lemma def_imp_eq in favor of meta_eq_to_obj_eq.
  5763 INCOMPATIBILITY.
  5764 
  5765 * Dropped redundant lemma if_def2 in favor of if_bool_eq_conj.
  5766 INCOMPATIBILITY.
  5767 
  5768 * HOL/records: generalised field-update to take a function on the
  5769 field rather than the new value: r(|A := x|) is translated to A_update
  5770 (K x) r The K-combinator that is internally used is called K_record.
  5771 INCOMPATIBILITY: Usage of the plain update functions has to be
  5772 adapted.
  5773  
  5774 * Class "semiring_0" now contains annihilation axioms x * 0 = 0 and 0
  5775 * x = 0, which are required for a semiring.  Richer structures do not
  5776 inherit from semiring_0 anymore, because this property is a theorem
  5777 there, not an axiom.  INCOMPATIBILITY: In instances of semiring_0,
  5778 there is more to prove, but this is mostly trivial.
  5779 
  5780 * Class "recpower" is generalized to arbitrary monoids, not just
  5781 commutative semirings.  INCOMPATIBILITY: may need to incorporate
  5782 commutativity or semiring properties additionally.
  5783 
  5784 * Constant "List.list_all2" in List.thy now uses authentic syntax.
  5785 INCOMPATIBILITY: translations containing list_all2 may go wrong,
  5786 better use 'abbreviation'.
  5787 
  5788 * Renamed constant "List.op mem" to "List.member".  INCOMPATIBILITY.
  5789 
  5790 * Numeral syntax: type 'bin' which was a mere type copy of 'int' has
  5791 been abandoned in favour of plain 'int'.  INCOMPATIBILITY --
  5792 significant changes for setting up numeral syntax for types:
  5793   - New constants Numeral.pred and Numeral.succ instead
  5794       of former Numeral.bin_pred and Numeral.bin_succ.
  5795   - Use integer operations instead of bin_add, bin_mult and so on.
  5796   - Numeral simplification theorems named Numeral.numeral_simps instead of Bin_simps.
  5797   - ML structure Bin_Simprocs now named Int_Numeral_Base_Simprocs.
  5798 
  5799 See src/HOL/Integ/IntArith.thy for an example setup.
  5800 
  5801 * Command 'normal_form' computes the normal form of a term that may
  5802 contain free variables.  For example ``normal_form "rev [a, b, c]"''
  5803 produces ``[b, c, a]'' (without proof).  This command is suitable for
  5804 heavy-duty computations because the functions are compiled to ML
  5805 first.  Correspondingly, a method "normalization" is provided.  See
  5806 further src/HOL/ex/NormalForm.thy and src/Tools/nbe.ML.
  5807 
  5808 * Alternative iff syntax "A <-> B" for equality on bool (with priority
  5809 25 like -->); output depends on the "iff" print_mode, the default is
  5810 "A = B" (with priority 50).
  5811 
  5812 * Relations less (<) and less_eq (<=) are also available on type bool.
  5813 Modified syntax to disallow nesting without explicit parentheses,
  5814 e.g. "(x < y) < z" or "x < (y < z)", but NOT "x < y < z".  Potential
  5815 INCOMPATIBILITY.
  5816 
  5817 * "LEAST x:A. P" expands to "LEAST x. x:A & P" (input only).
  5818 
  5819 * Relation composition operator "op O" now has precedence 75 and binds
  5820 stronger than union and intersection. INCOMPATIBILITY.
  5821 
  5822 * The old set interval syntax "{m..n(}" (and relatives) has been
  5823 removed.  Use "{m..<n}" (and relatives) instead.
  5824 
  5825 * In the context of the assumption "~(s = t)" the Simplifier rewrites
  5826 "t = s" to False (by simproc "neq").  INCOMPATIBILITY, consider using
  5827 ``declare [[simproc del: neq]]''.
  5828 
  5829 * Simplifier: "m dvd n" where m and n are numbers is evaluated to
  5830 True/False.
  5831 
  5832 * Theorem Cons_eq_map_conv no longer declared as "simp".
  5833 
  5834 * Theorem setsum_mult renamed to setsum_right_distrib.
  5835 
  5836 * Prefer ex1I over ex_ex1I in single-step reasoning, e.g. by the
  5837 ``rule'' method.
  5838 
  5839 * Reimplemented methods "sat" and "satx", with several improvements:
  5840 goals no longer need to be stated as "<prems> ==> False", equivalences
  5841 (i.e. "=" on type bool) are handled, variable names of the form
  5842 "lit_<n>" are no longer reserved, significant speedup.
  5843 
  5844 * Methods "sat" and "satx" can now replay MiniSat proof traces.
  5845 zChaff is still supported as well.
  5846 
  5847 * 'inductive' and 'datatype': provide projections of mutual rules,
  5848 bundled as foo_bar.inducts;
  5849 
  5850 * Library: moved theories Parity, GCD, Binomial, Infinite_Set to
  5851 Library.
  5852 
  5853 * Library: moved theory Accessible_Part to main HOL.
  5854 
  5855 * Library: added theory Coinductive_List of potentially infinite lists
  5856 as greatest fixed-point.
  5857 
  5858 * Library: added theory AssocList which implements (finite) maps as
  5859 association lists.
  5860 
  5861 * Method "evaluation" solves goals (i.e. a boolean expression)
  5862 efficiently by compiling it to ML.  The goal is "proved" (via an
  5863 oracle) if it evaluates to True.
  5864 
  5865 * Linear arithmetic now splits certain operators (e.g. min, max, abs)
  5866 also when invoked by the simplifier.  This results in the Simplifier
  5867 being more powerful on arithmetic goals.  INCOMPATIBILITY.
  5868 Configuration option fast_arith_split_limit=0 recovers the old
  5869 behavior.
  5870 
  5871 * Support for hex (0x20) and binary (0b1001) numerals.
  5872 
  5873 * New method: reify eqs (t), where eqs are equations for an
  5874 interpretation I :: 'a list => 'b => 'c and t::'c is an optional
  5875 parameter, computes a term s::'b and a list xs::'a list and proves the
  5876 theorem I xs s = t. This is also known as reification or quoting. The
  5877 resulting theorem is applied to the subgoal to substitute t with I xs
  5878 s.  If t is omitted, the subgoal itself is reified.
  5879 
  5880 * New method: reflection corr_thm eqs (t). The parameters eqs and (t)
  5881 are as explained above. corr_thm is a theorem for I vs (f t) = I vs t,
  5882 where f is supposed to be a computable function (in the sense of code
  5883 generattion). The method uses reify to compute s and xs as above then
  5884 applies corr_thm and uses normalization by evaluation to "prove" f s =
  5885 r and finally gets the theorem t = r, which is again applied to the
  5886 subgoal. An Example is available in src/HOL/ex/ReflectionEx.thy.
  5887 
  5888 * Reflection: Automatic reification now handels binding, an example is
  5889 available in src/HOL/ex/ReflectionEx.thy
  5890 
  5891 * HOL-Statespace: ``State Spaces: The Locale Way'' introduces a
  5892 command 'statespace' that is similar to 'record', but introduces an
  5893 abstract specification based on the locale infrastructure instead of
  5894 HOL types.  This leads to extra flexibility in composing state spaces,
  5895 in particular multiple inheritance and renaming of components.
  5896 
  5897 
  5898 *** HOL-Complex ***
  5899 
  5900 * Hyperreal: Functions root and sqrt are now defined on negative real
  5901 inputs so that root n (- x) = - root n x and sqrt (- x) = - sqrt x.
  5902 Nonnegativity side conditions have been removed from many lemmas, so
  5903 that more subgoals may now be solved by simplification; potential
  5904 INCOMPATIBILITY.
  5905 
  5906 * Real: new type classes formalize real normed vector spaces and
  5907 algebras, using new overloaded constants scaleR :: real => 'a => 'a
  5908 and norm :: 'a => real.
  5909 
  5910 * Real: constant of_real :: real => 'a::real_algebra_1 injects from
  5911 reals into other types. The overloaded constant Reals :: 'a set is now
  5912 defined as range of_real; potential INCOMPATIBILITY.
  5913 
  5914 * Real: proper support for ML code generation, including 'quickcheck'.
  5915 Reals are implemented as arbitrary precision rationals.
  5916 
  5917 * Hyperreal: Several constants that previously worked only for the
  5918 reals have been generalized, so they now work over arbitrary vector
  5919 spaces. Type annotations may need to be added in some cases; potential
  5920 INCOMPATIBILITY.
  5921 
  5922   Infinitesimal  :: ('a::real_normed_vector) star set
  5923   HFinite        :: ('a::real_normed_vector) star set
  5924   HInfinite      :: ('a::real_normed_vector) star set
  5925   approx         :: ('a::real_normed_vector) star => 'a star => bool
  5926   monad          :: ('a::real_normed_vector) star => 'a star set
  5927   galaxy         :: ('a::real_normed_vector) star => 'a star set
  5928   (NS)LIMSEQ     :: [nat => 'a::real_normed_vector, 'a] => bool
  5929   (NS)convergent :: (nat => 'a::real_normed_vector) => bool
  5930   (NS)Bseq       :: (nat => 'a::real_normed_vector) => bool
  5931   (NS)Cauchy     :: (nat => 'a::real_normed_vector) => bool
  5932   (NS)LIM        :: ['a::real_normed_vector => 'b::real_normed_vector, 'a, 'b] => bool
  5933   is(NS)Cont     :: ['a::real_normed_vector => 'b::real_normed_vector, 'a] => bool
  5934   deriv          :: ['a::real_normed_field => 'a, 'a, 'a] => bool
  5935   sgn            :: 'a::real_normed_vector => 'a
  5936   exp            :: 'a::{recpower,real_normed_field,banach} => 'a
  5937 
  5938 * Complex: Some complex-specific constants are now abbreviations for
  5939 overloaded ones: complex_of_real = of_real, cmod = norm, hcmod =
  5940 hnorm.  Other constants have been entirely removed in favor of the
  5941 polymorphic versions (INCOMPATIBILITY):
  5942 
  5943   approx        <-- capprox
  5944   HFinite       <-- CFinite
  5945   HInfinite     <-- CInfinite
  5946   Infinitesimal <-- CInfinitesimal
  5947   monad         <-- cmonad
  5948   galaxy        <-- cgalaxy
  5949   (NS)LIM       <-- (NS)CLIM, (NS)CRLIM
  5950   is(NS)Cont    <-- is(NS)Contc, is(NS)contCR
  5951   (ns)deriv     <-- (ns)cderiv
  5952 
  5953 
  5954 *** HOL-Algebra ***
  5955 
  5956 * Formalisation of ideals and the quotient construction over rings.
  5957 
  5958 * Order and lattice theory no longer based on records.
  5959 INCOMPATIBILITY.
  5960 
  5961 * Renamed lemmas least_carrier -> least_closed and greatest_carrier ->
  5962 greatest_closed.  INCOMPATIBILITY.
  5963 
  5964 * Method algebra is now set up via an attribute.  For examples see
  5965 Ring.thy.  INCOMPATIBILITY: the method is now weaker on combinations
  5966 of algebraic structures.
  5967 
  5968 * Renamed theory CRing to Ring.
  5969 
  5970 
  5971 *** HOL-Nominal ***
  5972 
  5973 * Substantial, yet incomplete support for nominal datatypes (binding
  5974 structures) based on HOL-Nominal logic.  See src/HOL/Nominal and
  5975 src/HOL/Nominal/Examples.  Prospective users should consult
  5976 http://isabelle.in.tum.de/nominal/
  5977 
  5978 
  5979 *** ML ***
  5980 
  5981 * ML basics: just one true type int, which coincides with IntInf.int
  5982 (even on SML/NJ).
  5983 
  5984 * ML within Isar: antiquotations allow to embed statically-checked
  5985 formal entities in the source, referring to the context available at
  5986 compile-time.  For example:
  5987 
  5988 ML {* @{sort "{zero,one}"} *}
  5989 ML {* @{typ "'a => 'b"} *}
  5990 ML {* @{term "%x. x"} *}
  5991 ML {* @{prop "x == y"} *}
  5992 ML {* @{ctyp "'a => 'b"} *}
  5993 ML {* @{cterm "%x. x"} *}
  5994 ML {* @{cprop "x == y"} *}
  5995 ML {* @{thm asm_rl} *}
  5996 ML {* @{thms asm_rl} *}
  5997 ML {* @{type_name c} *}
  5998 ML {* @{type_syntax c} *}
  5999 ML {* @{const_name c} *}
  6000 ML {* @{const_syntax c} *}
  6001 ML {* @{context} *}
  6002 ML {* @{theory} *}
  6003 ML {* @{theory Pure} *}
  6004 ML {* @{theory_ref} *}
  6005 ML {* @{theory_ref Pure} *}
  6006 ML {* @{simpset} *}
  6007 ML {* @{claset} *}
  6008 ML {* @{clasimpset} *}
  6009 
  6010 The same works for sources being ``used'' within an Isar context.
  6011 
  6012 * ML in Isar: improved error reporting; extra verbosity with
  6013 ML_Context.trace enabled.
  6014 
  6015 * Pure/General/table.ML: the join operations now works via exceptions
  6016 DUP/SAME instead of type option. This is simpler in simple cases, and
  6017 admits slightly more efficient complex applications.
  6018 
  6019 * Pure: 'advanced' translation functions (parse_translation etc.) now
  6020 use Context.generic instead of just theory.
  6021 
  6022 * Pure: datatype Context.generic joins theory/Proof.context and
  6023 provides some facilities for code that works in either kind of
  6024 context, notably GenericDataFun for uniform theory and proof data.
  6025 
  6026 * Pure: simplified internal attribute type, which is now always
  6027 Context.generic * thm -> Context.generic * thm. Global (theory) vs.
  6028 local (Proof.context) attributes have been discontinued, while
  6029 minimizing code duplication. Thm.rule_attribute and
  6030 Thm.declaration_attribute build canonical attributes; see also structure
  6031 Context for further operations on Context.generic, notably
  6032 GenericDataFun. INCOMPATIBILITY, need to adapt attribute type
  6033 declarations and definitions.
  6034 
  6035 * Context data interfaces (Theory/Proof/GenericDataFun): removed
  6036 name/print, uninitialized data defaults to ad-hoc copy of empty value,
  6037 init only required for impure data. INCOMPATIBILITY: empty really need
  6038 to be empty (no dependencies on theory content!)
  6039 
  6040 * Pure/kernel: consts certification ignores sort constraints given in
  6041 signature declarations. (This information is not relevant to the
  6042 logic, but only for type inference.) SIGNIFICANT INTERNAL CHANGE,
  6043 potential INCOMPATIBILITY.
  6044 
  6045 * Pure: axiomatic type classes are now purely definitional, with
  6046 explicit proofs of class axioms and super class relations performed
  6047 internally. See Pure/axclass.ML for the main internal interfaces --
  6048 notably AxClass.define_class supercedes AxClass.add_axclass, and
  6049 AxClass.axiomatize_class/classrel/arity supersede
  6050 Sign.add_classes/classrel/arities.
  6051 
  6052 * Pure/Isar: Args/Attrib parsers operate on Context.generic --
  6053 global/local versions on theory vs. Proof.context have been
  6054 discontinued; Attrib.syntax and Method.syntax have been adapted
  6055 accordingly.  INCOMPATIBILITY, need to adapt parser expressions for
  6056 attributes, methods, etc.
  6057 
  6058 * Pure: several functions of signature "... -> theory -> theory * ..."
  6059 have been reoriented to "... -> theory -> ... * theory" in order to
  6060 allow natural usage in combination with the ||>, ||>>, |-> and
  6061 fold_map combinators.
  6062 
  6063 * Pure: official theorem names (closed derivations) and additional
  6064 comments (tags) are now strictly separate.  Name hints -- which are
  6065 maintained as tags -- may be attached any time without affecting the
  6066 derivation.
  6067 
  6068 * Pure: primitive rule lift_rule now takes goal cterm instead of an
  6069 actual goal state (thm).  Use Thm.lift_rule (Thm.cprem_of st i) to
  6070 achieve the old behaviour.
  6071 
  6072 * Pure: the "Goal" constant is now called "prop", supporting a
  6073 slightly more general idea of ``protecting'' meta-level rule
  6074 statements.
  6075 
  6076 * Pure: Logic.(un)varify only works in a global context, which is now
  6077 enforced instead of silently assumed.  INCOMPATIBILITY, may use
  6078 Logic.legacy_(un)varify as temporary workaround.
  6079 
  6080 * Pure: structure Name provides scalable operations for generating
  6081 internal variable names, notably Name.variants etc.  This replaces
  6082 some popular functions from term.ML:
  6083 
  6084   Term.variant		->  Name.variant
  6085   Term.variantlist	->  Name.variant_list
  6086   Term.invent_names	->  Name.invent_list
  6087 
  6088 Note that low-level renaming rarely occurs in new code -- operations
  6089 from structure Variable are used instead (see below).
  6090 
  6091 * Pure: structure Variable provides fundamental operations for proper
  6092 treatment of fixed/schematic variables in a context.  For example,
  6093 Variable.import introduces fixes for schematics of given facts and
  6094 Variable.export reverses the effect (up to renaming) -- this replaces
  6095 various freeze_thaw operations.
  6096 
  6097 * Pure: structure Goal provides simple interfaces for
  6098 init/conclude/finish and tactical prove operations (replacing former
  6099 Tactic.prove).  Goal.prove is the canonical way to prove results
  6100 within a given context; Goal.prove_global is a degraded version for
  6101 theory level goals, including a global Drule.standard.  Note that
  6102 OldGoals.prove_goalw_cterm has long been obsolete, since it is
  6103 ill-behaved in a local proof context (e.g. with local fixes/assumes or
  6104 in a locale context).
  6105 
  6106 * Pure/Syntax: generic interfaces for parsing (Syntax.parse_term etc.)
  6107 and type checking (Syntax.check_term etc.), with common combinations
  6108 (Syntax.read_term etc.). These supersede former Sign.read_term etc.
  6109 which are considered legacy and await removal.
  6110 
  6111 * Pure/Syntax: generic interfaces for type unchecking
  6112 (Syntax.uncheck_terms etc.) and unparsing (Syntax.unparse_term etc.),
  6113 with common combinations (Syntax.pretty_term, Syntax.string_of_term
  6114 etc.).  Former Sign.pretty_term, Sign.string_of_term etc. are still
  6115 available for convenience, but refer to the very same operations using
  6116 a mere theory instead of a full context.
  6117 
  6118 * Isar: simplified treatment of user-level errors, using exception
  6119 ERROR of string uniformly.  Function error now merely raises ERROR,
  6120 without any side effect on output channels.  The Isar toplevel takes
  6121 care of proper display of ERROR exceptions.  ML code may use plain
  6122 handle/can/try; cat_error may be used to concatenate errors like this:
  6123 
  6124   ... handle ERROR msg => cat_error msg "..."
  6125 
  6126 Toplevel ML code (run directly or through the Isar toplevel) may be
  6127 embedded into the Isar toplevel with exception display/debug like
  6128 this:
  6129 
  6130   Isar.toplevel (fn () => ...)
  6131 
  6132 INCOMPATIBILITY, removed special transform_error facilities, removed
  6133 obsolete variants of user-level exceptions (ERROR_MESSAGE,
  6134 Context.PROOF, ProofContext.CONTEXT, Proof.STATE, ProofHistory.FAIL)
  6135 -- use plain ERROR instead.
  6136 
  6137 * Isar: theory setup now has type (theory -> theory), instead of a
  6138 list.  INCOMPATIBILITY, may use #> to compose setup functions.
  6139 
  6140 * Isar: ML toplevel pretty printer for type Proof.context, subject to
  6141 ProofContext.debug/verbose flags.
  6142 
  6143 * Isar: Toplevel.theory_to_proof admits transactions that modify the
  6144 theory before entering a proof state.  Transactions now always see a
  6145 quasi-functional intermediate checkpoint, both in interactive and
  6146 batch mode.
  6147 
  6148 * Isar: simplified interfaces for outer syntax.  Renamed
  6149 OuterSyntax.add_keywords to OuterSyntax.keywords.  Removed
  6150 OuterSyntax.add_parsers -- this functionality is now included in
  6151 OuterSyntax.command etc.  INCOMPATIBILITY.
  6152 
  6153 * Simplifier: the simpset of a running simplification process now
  6154 contains a proof context (cf. Simplifier.the_context), which is the
  6155 very context that the initial simpset has been retrieved from (by
  6156 simpset_of/local_simpset_of).  Consequently, all plug-in components
  6157 (solver, looper etc.) may depend on arbitrary proof data.
  6158 
  6159 * Simplifier.inherit_context inherits the proof context (plus the
  6160 local bounds) of the current simplification process; any simproc
  6161 etc. that calls the Simplifier recursively should do this!  Removed
  6162 former Simplifier.inherit_bounds, which is already included here --
  6163 INCOMPATIBILITY.  Tools based on low-level rewriting may even have to
  6164 specify an explicit context using Simplifier.context/theory_context.
  6165 
  6166 * Simplifier/Classical Reasoner: more abstract interfaces
  6167 change_simpset/claset for modifying the simpset/claset reference of a
  6168 theory; raw versions simpset/claset_ref etc. have been discontinued --
  6169 INCOMPATIBILITY.
  6170 
  6171 * Provers: more generic wrt. syntax of object-logics, avoid hardwired
  6172 "Trueprop" etc.
  6173 
  6174 
  6175 *** System ***
  6176 
  6177 * settings: the default heap location within ISABELLE_HOME_USER now
  6178 includes ISABELLE_IDENTIFIER.  This simplifies use of multiple
  6179 Isabelle installations.
  6180 
  6181 * isabelle-process: option -S (secure mode) disables some critical
  6182 operations, notably runtime compilation and evaluation of ML source
  6183 code.
  6184 
  6185 * Basic Isabelle mode for jEdit, see Isabelle/lib/jedit/.
  6186 
  6187 * Support for parallel execution, using native multicore support of
  6188 Poly/ML 5.1.  The theory loader exploits parallelism when processing
  6189 independent theories, according to the given theory header
  6190 specifications. The maximum number of worker threads is specified via
  6191 usedir option -M or the "max-threads" setting in Proof General. A
  6192 speedup factor of 1.5--3.5 can be expected on a 4-core machine, and up
  6193 to 6 on a 8-core machine.  User-code needs to observe certain
  6194 guidelines for thread-safe programming, see appendix A in the Isar
  6195 Implementation manual.
  6196 
  6197 
  6198 
  6199 New in Isabelle2005 (October 2005)
  6200 ----------------------------------
  6201 
  6202 *** General ***
  6203 
  6204 * Theory headers: the new header syntax for Isar theories is
  6205 
  6206   theory <name>
  6207   imports <theory1> ... <theoryN>
  6208   uses <file1> ... <fileM>
  6209   begin
  6210 
  6211 where the 'uses' part is optional.  The previous syntax
  6212 
  6213   theory <name> = <theory1> + ... + <theoryN>:
  6214 
  6215 will disappear in the next release.  Use isatool fixheaders to convert
  6216 existing theory files.  Note that there is no change in ancient
  6217 non-Isar theories now, but these will disappear soon.
  6218 
  6219 * Theory loader: parent theories can now also be referred to via
  6220 relative and absolute paths.
  6221 
  6222 * Command 'find_theorems' searches for a list of criteria instead of a
  6223 list of constants. Known criteria are: intro, elim, dest, name:string,
  6224 simp:term, and any term. Criteria can be preceded by '-' to select
  6225 theorems that do not match. Intro, elim, dest select theorems that
  6226 match the current goal, name:s selects theorems whose fully qualified
  6227 name contain s, and simp:term selects all simplification rules whose
  6228 lhs match term.  Any other term is interpreted as pattern and selects
  6229 all theorems matching the pattern. Available in ProofGeneral under
  6230 'ProofGeneral -> Find Theorems' or C-c C-f.  Example:
  6231 
  6232   C-c C-f (100) "(_::nat) + _ + _" intro -name: "HOL."
  6233 
  6234 prints the last 100 theorems matching the pattern "(_::nat) + _ + _",
  6235 matching the current goal as introduction rule and not having "HOL."
  6236 in their name (i.e. not being defined in theory HOL).
  6237 
  6238 * Command 'thms_containing' has been discontinued in favour of
  6239 'find_theorems'; INCOMPATIBILITY.
  6240</