NEWS
author haftmann
Thu Jul 19 22:21:59 2012 +0200 (2012-07-19)
changeset 48371 3a5a5a992519
parent 48317 e5420161d11d
child 48402 327ebf1c42a8
permissions -rw-r--r--
export code relatively to master directory
     1 Isabelle NEWS -- history user-relevant changes
     2 ==============================================
     3 
     4 New in this Isabelle version
     5 ----------------------------
     6 
     7 *** General ***
     8 
     9 * Discontinued obsolete method fastsimp / tactic fast_simp_tac, which
    10 is called fastforce / fast_force_tac already since Isabelle2011-1.
    11 
    12 * Updated and extended "isar-ref" manual, reduced remaining material
    13 in old "ref" manual.
    14 
    15 
    16 *** Pure ***
    17 
    18 * Command "export_code": relative file names are interpreted
    19 relatively to master directory of current theory rather than
    20 the rather arbitrary current working directory.
    21 INCOMPATIBILITY.
    22 
    23 * Discontinued obsolete attribute "COMP".  Potential INCOMPATIBILITY,
    24 use regular rule composition via "OF" / "THEN", or explicit proof
    25 structure instead.  Note that Isabelle/ML provides a variety of
    26 operators like COMP, INCR_COMP, COMP_INCR, which need to be applied
    27 with some care where this is really required.
    28 
    29 
    30 *** HOL ***
    31 
    32 * Simproc "finite_Collect" rewrites set comprehensions into pointfree
    33 expressions.
    34 
    35 * Quickcheck:
    36 
    37   - added an optimisation for equality premises.
    38     It is switched on by default, and can be switched off by setting
    39     the configuration quickcheck_optimise_equality to false.    
    40 
    41 * The SMT solver Z3 has now by default a restricted set of directly
    42 supported features. For the full set of features (div/mod, nonlinear
    43 arithmetic, datatypes/records) with potential proof reconstruction
    44 failures, enable the configuration option "z3_with_extensions".
    45 Minor INCOMPATIBILITY.
    46 
    47 * Sledgehammer:
    48 
    49   - Rationalized type encodings ("type_enc" option).
    50   - Renamed options:
    51       max_relevant ~> max_facts
    52       relevance_thresholds ~> fact_thresholds
    53 
    54 
    55 *** Document preparation ***
    56 
    57 * Default for \<euro> is now based on eurosym package, instead of
    58 slightly exotic babel/greek.
    59 
    60 
    61 *** System ***
    62 
    63 * Discontinued support for Poly/ML 5.2.1, which was the last version
    64 without exception positions and advanced ML compiler/toplevel
    65 configuration.
    66 
    67 
    68 
    69 New in Isabelle2012 (May 2012)
    70 ------------------------------
    71 
    72 *** General ***
    73 
    74 * Prover IDE (PIDE) improvements:
    75 
    76   - more robust Sledgehammer integration (as before the sledgehammer
    77     command-line needs to be typed into the source buffer)
    78   - markup for bound variables
    79   - markup for types of term variables (displayed as tooltips)
    80   - support for user-defined Isar commands within the running session
    81   - improved support for Unicode outside original 16bit range
    82     e.g. glyph for \<A> (thanks to jEdit 4.5.1)
    83 
    84 * Forward declaration of outer syntax keywords within the theory
    85 header -- minor INCOMPATIBILITY for user-defined commands.  Allow new
    86 commands to be used in the same theory where defined.
    87 
    88 * Auxiliary contexts indicate block structure for specifications with
    89 additional parameters and assumptions.  Such unnamed contexts may be
    90 nested within other targets, like 'theory', 'locale', 'class',
    91 'instantiation' etc.  Results from the local context are generalized
    92 accordingly and applied to the enclosing target context.  Example:
    93 
    94   context
    95     fixes x y z :: 'a
    96     assumes xy: "x = y" and yz: "y = z"
    97   begin
    98 
    99   lemma my_trans: "x = z" using xy yz by simp
   100 
   101   end
   102 
   103   thm my_trans
   104 
   105 The most basic application is to factor-out context elements of
   106 several fixes/assumes/shows theorem statements, e.g. see
   107 ~~/src/HOL/Isar_Examples/Group_Context.thy
   108 
   109 Any other local theory specification element works within the "context
   110 ... begin ... end" block as well.
   111 
   112 * Bundled declarations associate attributed fact expressions with a
   113 given name in the context.  These may be later included in other
   114 contexts.  This allows to manage context extensions casually, without
   115 the logical dependencies of locales and locale interpretation.  See
   116 commands 'bundle', 'include', 'including' etc. in the isar-ref manual.
   117 
   118 * Commands 'lemmas' and 'theorems' allow local variables using 'for'
   119 declaration, and results are standardized before being stored.  Thus
   120 old-style "standard" after instantiation or composition of facts
   121 becomes obsolete.  Minor INCOMPATIBILITY, due to potential change of
   122 indices of schematic variables.
   123 
   124 * Rule attributes in local theory declarations (e.g. locale or class)
   125 are now statically evaluated: the resulting theorem is stored instead
   126 of the original expression.  INCOMPATIBILITY in rare situations, where
   127 the historic accident of dynamic re-evaluation in interpretations
   128 etc. was exploited.
   129 
   130 * New tutorial "Programming and Proving in Isabelle/HOL"
   131 ("prog-prove").  It completely supersedes "A Tutorial Introduction to
   132 Structured Isar Proofs" ("isar-overview"), which has been removed.  It
   133 also supersedes "Isabelle/HOL, A Proof Assistant for Higher-Order
   134 Logic" as the recommended beginners tutorial, but does not cover all
   135 of the material of that old tutorial.
   136 
   137 * Updated and extended reference manuals: "isar-ref",
   138 "implementation", "system"; reduced remaining material in old "ref"
   139 manual.
   140 
   141 
   142 *** Pure ***
   143 
   144 * Command 'definition' no longer exports the foundational "raw_def"
   145 into the user context.  Minor INCOMPATIBILITY, may use the regular
   146 "def" result with attribute "abs_def" to imitate the old version.
   147 
   148 * Attribute "abs_def" turns an equation of the form "f x y == t" into
   149 "f == %x y. t", which ensures that "simp" or "unfold" steps always
   150 expand it.  This also works for object-logic equality.  (Formerly
   151 undocumented feature.)
   152 
   153 * Sort constraints are now propagated in simultaneous statements, just
   154 like type constraints.  INCOMPATIBILITY in rare situations, where
   155 distinct sorts used to be assigned accidentally.  For example:
   156 
   157   lemma "P (x::'a::foo)" and "Q (y::'a::bar)"  -- "now illegal"
   158 
   159   lemma "P (x::'a)" and "Q (y::'a::bar)"
   160     -- "now uniform 'a::bar instead of default sort for first occurrence (!)"
   161 
   162 * Rule composition via attribute "OF" (or ML functions OF/MRS) is more
   163 tolerant against multiple unifiers, as long as the final result is
   164 unique.  (As before, rules are composed in canonical right-to-left
   165 order to accommodate newly introduced premises.)
   166 
   167 * Renamed some inner syntax categories:
   168 
   169     num ~> num_token
   170     xnum ~> xnum_token
   171     xstr ~> str_token
   172 
   173 Minor INCOMPATIBILITY.  Note that in practice "num_const" or
   174 "num_position" etc. are mainly used instead (which also include
   175 position information via constraints).
   176 
   177 * Simplified configuration options for syntax ambiguity: see
   178 "syntax_ambiguity_warning" and "syntax_ambiguity_limit" in isar-ref
   179 manual.  Minor INCOMPATIBILITY.
   180 
   181 * Discontinued configuration option "syntax_positions": atomic terms
   182 in parse trees are always annotated by position constraints.
   183 
   184 * Old code generator for SML and its commands 'code_module',
   185 'code_library', 'consts_code', 'types_code' have been discontinued.
   186 Use commands of the generic code generator instead.  INCOMPATIBILITY.
   187 
   188 * Redundant attribute "code_inline" has been discontinued. Use
   189 "code_unfold" instead.  INCOMPATIBILITY.
   190 
   191 * Dropped attribute "code_unfold_post" in favor of the its dual
   192 "code_abbrev", which yields a common pattern in definitions like
   193 
   194   definition [code_abbrev]: "f = t"
   195 
   196 INCOMPATIBILITY.
   197 
   198 * Obsolete 'types' command has been discontinued.  Use 'type_synonym'
   199 instead.  INCOMPATIBILITY.
   200 
   201 * Discontinued old "prems" fact, which used to refer to the accidental
   202 collection of foundational premises in the context (already marked as
   203 legacy since Isabelle2011).
   204 
   205 
   206 *** HOL ***
   207 
   208 * Type 'a set is now a proper type constructor (just as before
   209 Isabelle2008).  Definitions mem_def and Collect_def have disappeared.
   210 Non-trivial INCOMPATIBILITY.  For developments keeping predicates and
   211 sets separate, it is often sufficient to rephrase some set S that has
   212 been accidentally used as predicates by "%x. x : S", and some
   213 predicate P that has been accidentally used as set by "{x. P x}".
   214 Corresponding proofs in a first step should be pruned from any
   215 tinkering with former theorems mem_def and Collect_def as far as
   216 possible.
   217 
   218 For developments which deliberately mix predicates and sets, a
   219 planning step is necessary to determine what should become a predicate
   220 and what a set.  It can be helpful to carry out that step in
   221 Isabelle2011-1 before jumping right into the current release.
   222 
   223 * Code generation by default implements sets as container type rather
   224 than predicates.  INCOMPATIBILITY.
   225 
   226 * New type synonym 'a rel = ('a * 'a) set
   227 
   228 * The representation of numerals has changed.  Datatype "num"
   229 represents strictly positive binary numerals, along with functions
   230 "numeral :: num => 'a" and "neg_numeral :: num => 'a" to represent
   231 positive and negated numeric literals, respectively.  See also
   232 definitions in ~~/src/HOL/Num.thy.  Potential INCOMPATIBILITY, some
   233 user theories may require adaptations as follows:
   234 
   235   - Theorems with number_ring or number_semiring constraints: These
   236     classes are gone; use comm_ring_1 or comm_semiring_1 instead.
   237 
   238   - Theories defining numeric types: Remove number, number_semiring,
   239     and number_ring instances. Defer all theorems about numerals until
   240     after classes one and semigroup_add have been instantiated.
   241 
   242   - Numeral-only simp rules: Replace each rule having a "number_of v"
   243     pattern with two copies, one for numeral and one for neg_numeral.
   244 
   245   - Theorems about subclasses of semiring_1 or ring_1: These classes
   246     automatically support numerals now, so more simp rules and
   247     simprocs may now apply within the proof.
   248 
   249   - Definitions and theorems using old constructors Pls/Min/Bit0/Bit1:
   250     Redefine using other integer operations.
   251 
   252 * Transfer: New package intended to generalize the existing
   253 "descending" method and related theorem attributes from the Quotient
   254 package.  (Not all functionality is implemented yet, but future
   255 development will focus on Transfer as an eventual replacement for the
   256 corresponding parts of the Quotient package.)
   257 
   258   - transfer_rule attribute: Maintains a collection of transfer rules,
   259     which relate constants at two different types. Transfer rules may
   260     relate different type instances of the same polymorphic constant,
   261     or they may relate an operation on a raw type to a corresponding
   262     operation on an abstract type (quotient or subtype). For example:
   263 
   264     ((A ===> B) ===> list_all2 A ===> list_all2 B) map map
   265     (cr_int ===> cr_int ===> cr_int) (%(x,y) (u,v). (x+u, y+v)) plus_int
   266 
   267   - transfer method: Replaces a subgoal on abstract types with an
   268     equivalent subgoal on the corresponding raw types. Constants are
   269     replaced with corresponding ones according to the transfer rules.
   270     Goals are generalized over all free variables by default; this is
   271     necessary for variables whose types change, but can be overridden
   272     for specific variables with e.g. "transfer fixing: x y z".  The
   273     variant transfer' method allows replacing a subgoal with one that
   274     is logically stronger (rather than equivalent).
   275 
   276   - relator_eq attribute: Collects identity laws for relators of
   277     various type constructors, e.g. "list_all2 (op =) = (op =)".  The
   278     transfer method uses these lemmas to infer transfer rules for
   279     non-polymorphic constants on the fly.
   280 
   281   - transfer_prover method: Assists with proving a transfer rule for a
   282     new constant, provided the constant is defined in terms of other
   283     constants that already have transfer rules. It should be applied
   284     after unfolding the constant definitions.
   285 
   286   - HOL/ex/Transfer_Int_Nat.thy: Example theory demonstrating transfer
   287     from type nat to type int.
   288 
   289 * Lifting: New package intended to generalize the quotient_definition
   290 facility of the Quotient package; designed to work with Transfer.
   291 
   292   - lift_definition command: Defines operations on an abstract type in
   293     terms of a corresponding operation on a representation
   294     type.  Example syntax:
   295 
   296     lift_definition dlist_insert :: "'a => 'a dlist => 'a dlist"
   297       is List.insert
   298 
   299     Users must discharge a respectfulness proof obligation when each
   300     constant is defined. (For a type copy, i.e. a typedef with UNIV,
   301     the proof is discharged automatically.) The obligation is
   302     presented in a user-friendly, readable form; a respectfulness
   303     theorem in the standard format and a transfer rule are generated
   304     by the package.
   305 
   306   - Integration with code_abstype: For typedefs (e.g. subtypes
   307     corresponding to a datatype invariant, such as dlist),
   308     lift_definition generates a code certificate theorem and sets up
   309     code generation for each constant.
   310 
   311   - setup_lifting command: Sets up the Lifting package to work with a
   312     user-defined type. The user must provide either a quotient theorem
   313     or a type_definition theorem.  The package configures transfer
   314     rules for equality and quantifiers on the type, and sets up the
   315     lift_definition command to work with the type.
   316 
   317   - Usage examples: See Quotient_Examples/Lift_DList.thy,
   318     Quotient_Examples/Lift_RBT.thy, Quotient_Examples/Lift_FSet.thy,
   319     Word/Word.thy and Library/Float.thy.
   320 
   321 * Quotient package:
   322 
   323   - The 'quotient_type' command now supports a 'morphisms' option with
   324     rep and abs functions, similar to typedef.
   325 
   326   - 'quotient_type' sets up new types to work with the Lifting and
   327     Transfer packages, as with 'setup_lifting'.
   328 
   329   - The 'quotient_definition' command now requires the user to prove a
   330     respectfulness property at the point where the constant is
   331     defined, similar to lift_definition; INCOMPATIBILITY.
   332 
   333   - Renamed predicate 'Quotient' to 'Quotient3', and renamed theorems
   334     accordingly, INCOMPATIBILITY.
   335 
   336 * New diagnostic command 'find_unused_assms' to find potentially
   337 superfluous assumptions in theorems using Quickcheck.
   338 
   339 * Quickcheck:
   340 
   341   - Quickcheck returns variable assignments as counterexamples, which
   342     allows to reveal the underspecification of functions under test.
   343     For example, refuting "hd xs = x", it presents the variable
   344     assignment xs = [] and x = a1 as a counterexample, assuming that
   345     any property is false whenever "hd []" occurs in it.
   346 
   347     These counterexample are marked as potentially spurious, as
   348     Quickcheck also returns "xs = []" as a counterexample to the
   349     obvious theorem "hd xs = hd xs".
   350 
   351     After finding a potentially spurious counterexample, Quickcheck
   352     continues searching for genuine ones.
   353 
   354     By default, Quickcheck shows potentially spurious and genuine
   355     counterexamples. The option "genuine_only" sets quickcheck to only
   356     show genuine counterexamples.
   357 
   358   - The command 'quickcheck_generator' creates random and exhaustive
   359     value generators for a given type and operations.
   360 
   361     It generates values by using the operations as if they were
   362     constructors of that type.
   363 
   364   - Support for multisets.
   365 
   366   - Added "use_subtype" options.
   367 
   368   - Added "quickcheck_locale" configuration to specify how to process
   369     conjectures in a locale context.
   370 
   371 * Nitpick: Fixed infinite loop caused by the 'peephole_optim' option
   372 and affecting 'rat' and 'real'.
   373 
   374 * Sledgehammer:
   375   - Integrated more tightly with SPASS, as described in the ITP 2012
   376     paper "More SPASS with Isabelle".
   377   - Made it try "smt" as a fallback if "metis" fails or times out.
   378   - Added support for the following provers: Alt-Ergo (via Why3 and
   379     TFF1), iProver, iProver-Eq.
   380   - Sped up the minimizer.
   381   - Added "lam_trans", "uncurry_aliases", and "minimize" options.
   382   - Renamed "slicing" ("no_slicing") option to "slice" ("dont_slice").
   383   - Renamed "sound" option to "strict".
   384 
   385 * Metis: Added possibility to specify lambda translations scheme as a
   386 parenthesized argument (e.g., "by (metis (lifting) ...)").
   387 
   388 * SMT: Renamed "smt_fixed" option to "smt_read_only_certificates".
   389 
   390 * Command 'try0': Renamed from 'try_methods'. INCOMPATIBILITY.
   391 
   392 * New "case_product" attribute to generate a case rule doing multiple
   393 case distinctions at the same time.  E.g.
   394 
   395   list.exhaust [case_product nat.exhaust]
   396 
   397 produces a rule which can be used to perform case distinction on both
   398 a list and a nat.
   399 
   400 * New "eventually_elim" method as a generalized variant of the
   401 eventually_elim* rules.  Supports structured proofs.
   402 
   403 * Typedef with implicit set definition is considered legacy.  Use
   404 "typedef (open)" form instead, which will eventually become the
   405 default.
   406 
   407 * Record: code generation can be switched off manually with
   408 
   409   declare [[record_coden = false]]  -- "default true"
   410 
   411 * Datatype: type parameters allow explicit sort constraints.
   412 
   413 * Concrete syntax for case expressions includes constraints for source
   414 positions, and thus produces Prover IDE markup for its bindings.
   415 INCOMPATIBILITY for old-style syntax translations that augment the
   416 pattern notation; e.g. see src/HOL/HOLCF/One.thy for translations of
   417 one_case.
   418 
   419 * Clarified attribute "mono_set": pure declaration without modifying
   420 the result of the fact expression.
   421 
   422 * More default pred/set conversions on a couple of relation operations
   423 and predicates.  Added powers of predicate relations.  Consolidation
   424 of some relation theorems:
   425 
   426   converse_def ~> converse_unfold
   427   rel_comp_def ~> relcomp_unfold
   428   symp_def ~> (modified, use symp_def and sym_def instead)
   429   transp_def ~> transp_trans
   430   Domain_def ~> Domain_unfold
   431   Range_def ~> Domain_converse [symmetric]
   432 
   433 Generalized theorems INF_INT_eq, INF_INT_eq2, SUP_UN_eq, SUP_UN_eq2.
   434 
   435 See theory "Relation" for examples for making use of pred/set
   436 conversions by means of attributes "to_set" and "to_pred".
   437 
   438 INCOMPATIBILITY.
   439 
   440 * Renamed facts about the power operation on relations, i.e., relpow
   441 to match the constant's name:
   442 
   443   rel_pow_1 ~> relpow_1
   444   rel_pow_0_I ~> relpow_0_I
   445   rel_pow_Suc_I ~> relpow_Suc_I
   446   rel_pow_Suc_I2 ~> relpow_Suc_I2
   447   rel_pow_0_E ~> relpow_0_E
   448   rel_pow_Suc_E ~> relpow_Suc_E
   449   rel_pow_E ~> relpow_E
   450   rel_pow_Suc_D2 ~> relpow_Suc_D2
   451   rel_pow_Suc_E2 ~> relpow_Suc_E2
   452   rel_pow_Suc_D2' ~> relpow_Suc_D2'
   453   rel_pow_E2 ~> relpow_E2
   454   rel_pow_add ~> relpow_add
   455   rel_pow_commute ~> relpow
   456   rel_pow_empty ~> relpow_empty:
   457   rtrancl_imp_UN_rel_pow ~> rtrancl_imp_UN_relpow
   458   rel_pow_imp_rtrancl ~> relpow_imp_rtrancl
   459   rtrancl_is_UN_rel_pow ~> rtrancl_is_UN_relpow
   460   rtrancl_imp_rel_pow ~> rtrancl_imp_relpow
   461   rel_pow_fun_conv ~> relpow_fun_conv
   462   rel_pow_finite_bounded1 ~> relpow_finite_bounded1
   463   rel_pow_finite_bounded ~> relpow_finite_bounded
   464   rtrancl_finite_eq_rel_pow ~> rtrancl_finite_eq_relpow
   465   trancl_finite_eq_rel_pow ~> trancl_finite_eq_relpow
   466   single_valued_rel_pow ~> single_valued_relpow
   467 
   468 INCOMPATIBILITY.
   469 
   470 * Theory Relation: Consolidated constant name for relation composition
   471 and corresponding theorem names:
   472 
   473   - Renamed constant rel_comp to relcomp.
   474 
   475   - Dropped abbreviation pred_comp. Use relcompp instead.
   476 
   477   - Renamed theorems:
   478 
   479     rel_compI ~> relcompI
   480     rel_compEpair ~> relcompEpair
   481     rel_compE ~> relcompE
   482     pred_comp_rel_comp_eq ~> relcompp_relcomp_eq
   483     rel_comp_empty1 ~> relcomp_empty1
   484     rel_comp_mono ~> relcomp_mono
   485     rel_comp_subset_Sigma ~> relcomp_subset_Sigma
   486     rel_comp_distrib ~> relcomp_distrib
   487     rel_comp_distrib2 ~> relcomp_distrib2
   488     rel_comp_UNION_distrib ~> relcomp_UNION_distrib
   489     rel_comp_UNION_distrib2 ~> relcomp_UNION_distrib2
   490     single_valued_rel_comp ~> single_valued_relcomp
   491     rel_comp_def ~> relcomp_unfold
   492     converse_rel_comp ~> converse_relcomp
   493     pred_compI ~> relcomppI
   494     pred_compE ~> relcomppE
   495     pred_comp_bot1 ~> relcompp_bot1
   496     pred_comp_bot2 ~> relcompp_bot2
   497     transp_pred_comp_less_eq ~> transp_relcompp_less_eq
   498     pred_comp_mono ~> relcompp_mono
   499     pred_comp_distrib ~> relcompp_distrib
   500     pred_comp_distrib2 ~> relcompp_distrib2
   501     converse_pred_comp ~> converse_relcompp
   502 
   503     finite_rel_comp ~> finite_relcomp
   504 
   505     set_rel_comp ~> set_relcomp
   506 
   507 INCOMPATIBILITY.
   508 
   509 * Theory Divides: Discontinued redundant theorems about div and mod.
   510 INCOMPATIBILITY, use the corresponding generic theorems instead.
   511 
   512   DIVISION_BY_ZERO ~> div_by_0, mod_by_0
   513   zdiv_self ~> div_self
   514   zmod_self ~> mod_self
   515   zdiv_zero ~> div_0
   516   zmod_zero ~> mod_0
   517   zdiv_zmod_equality ~> div_mod_equality2
   518   zdiv_zmod_equality2 ~> div_mod_equality
   519   zmod_zdiv_trivial ~> mod_div_trivial
   520   zdiv_zminus_zminus ~> div_minus_minus
   521   zmod_zminus_zminus ~> mod_minus_minus
   522   zdiv_zminus2 ~> div_minus_right
   523   zmod_zminus2 ~> mod_minus_right
   524   zdiv_minus1_right ~> div_minus1_right
   525   zmod_minus1_right ~> mod_minus1_right
   526   zdvd_mult_div_cancel ~> dvd_mult_div_cancel
   527   zmod_zmult1_eq ~> mod_mult_right_eq
   528   zpower_zmod ~> power_mod
   529   zdvd_zmod ~> dvd_mod
   530   zdvd_zmod_imp_zdvd ~> dvd_mod_imp_dvd
   531   mod_mult_distrib ~> mult_mod_left
   532   mod_mult_distrib2 ~> mult_mod_right
   533 
   534 * Removed redundant theorems nat_mult_2 and nat_mult_2_right; use
   535 generic mult_2 and mult_2_right instead. INCOMPATIBILITY.
   536 
   537 * Finite_Set.fold now qualified.  INCOMPATIBILITY.
   538 
   539 * Consolidated theorem names concerning fold combinators:
   540 
   541   inf_INFI_fold_inf ~> inf_INF_fold_inf
   542   sup_SUPR_fold_sup ~> sup_SUP_fold_sup
   543   INFI_fold_inf ~> INF_fold_inf
   544   SUPR_fold_sup ~> SUP_fold_sup
   545   union_set ~> union_set_fold
   546   minus_set ~> minus_set_fold
   547   INFI_set_fold ~> INF_set_fold
   548   SUPR_set_fold ~> SUP_set_fold
   549   INF_code ~> INF_set_foldr
   550   SUP_code ~> SUP_set_foldr
   551   foldr.simps ~> foldr.simps (in point-free formulation)
   552   foldr_fold_rev ~> foldr_conv_fold
   553   foldl_fold ~> foldl_conv_fold
   554   foldr_foldr ~> foldr_conv_foldl
   555   foldl_foldr ~> foldl_conv_foldr
   556   fold_set_remdups ~> fold_set_fold_remdups
   557   fold_set ~> fold_set_fold
   558   fold1_set ~> fold1_set_fold
   559 
   560 INCOMPATIBILITY.
   561 
   562 * Dropped rarely useful theorems concerning fold combinators:
   563 foldl_apply, foldl_fun_comm, foldl_rev, fold_weak_invariant,
   564 rev_foldl_cons, fold_set_remdups, fold_set, fold_set1,
   565 concat_conv_foldl, foldl_weak_invariant, foldl_invariant,
   566 foldr_invariant, foldl_absorb0, foldl_foldr1_lemma, foldl_foldr1,
   567 listsum_conv_fold, listsum_foldl, sort_foldl_insort, foldl_assoc,
   568 foldr_conv_foldl, start_le_sum, elem_le_sum, sum_eq_0_conv.
   569 INCOMPATIBILITY.  For the common phrases "%xs. List.foldr plus xs 0"
   570 and "List.foldl plus 0", prefer "List.listsum".  Otherwise it can be
   571 useful to boil down "List.foldr" and "List.foldl" to "List.fold" by
   572 unfolding "foldr_conv_fold" and "foldl_conv_fold".
   573 
   574 * Dropped lemmas minus_set_foldr, union_set_foldr, union_coset_foldr,
   575 inter_coset_foldr, Inf_fin_set_foldr, Sup_fin_set_foldr,
   576 Min_fin_set_foldr, Max_fin_set_foldr, Inf_set_foldr, Sup_set_foldr,
   577 INF_set_foldr, SUP_set_foldr.  INCOMPATIBILITY.  Prefer corresponding
   578 lemmas over fold rather than foldr, or make use of lemmas
   579 fold_conv_foldr and fold_rev.
   580 
   581 * Congruence rules Option.map_cong and Option.bind_cong for recursion
   582 through option types.
   583 
   584 * "Transitive_Closure.ntrancl": bounded transitive closure on
   585 relations.
   586 
   587 * Constant "Set.not_member" now qualified.  INCOMPATIBILITY.
   588 
   589 * Theory Int: Discontinued many legacy theorems specific to type int.
   590 INCOMPATIBILITY, use the corresponding generic theorems instead.
   591 
   592   zminus_zminus ~> minus_minus
   593   zminus_0 ~> minus_zero
   594   zminus_zadd_distrib ~> minus_add_distrib
   595   zadd_commute ~> add_commute
   596   zadd_assoc ~> add_assoc
   597   zadd_left_commute ~> add_left_commute
   598   zadd_ac ~> add_ac
   599   zmult_ac ~> mult_ac
   600   zadd_0 ~> add_0_left
   601   zadd_0_right ~> add_0_right
   602   zadd_zminus_inverse2 ~> left_minus
   603   zmult_zminus ~> mult_minus_left
   604   zmult_commute ~> mult_commute
   605   zmult_assoc ~> mult_assoc
   606   zadd_zmult_distrib ~> left_distrib
   607   zadd_zmult_distrib2 ~> right_distrib
   608   zdiff_zmult_distrib ~> left_diff_distrib
   609   zdiff_zmult_distrib2 ~> right_diff_distrib
   610   zmult_1 ~> mult_1_left
   611   zmult_1_right ~> mult_1_right
   612   zle_refl ~> order_refl
   613   zle_trans ~> order_trans
   614   zle_antisym ~> order_antisym
   615   zle_linear ~> linorder_linear
   616   zless_linear ~> linorder_less_linear
   617   zadd_left_mono ~> add_left_mono
   618   zadd_strict_right_mono ~> add_strict_right_mono
   619   zadd_zless_mono ~> add_less_le_mono
   620   int_0_less_1 ~> zero_less_one
   621   int_0_neq_1 ~> zero_neq_one
   622   zless_le ~> less_le
   623   zpower_zadd_distrib ~> power_add
   624   zero_less_zpower_abs_iff ~> zero_less_power_abs_iff
   625   zero_le_zpower_abs ~> zero_le_power_abs
   626 
   627 * Theory Deriv: Renamed
   628 
   629   DERIV_nonneg_imp_nonincreasing ~> DERIV_nonneg_imp_nondecreasing
   630 
   631 * Theory Library/Multiset: Improved code generation of multisets.
   632 
   633 * Theory HOL/Library/Set_Algebras: Addition and multiplication on sets
   634 are expressed via type classes again. The special syntax
   635 \<oplus>/\<otimes> has been replaced by plain +/*. Removed constant
   636 setsum_set, which is now subsumed by Big_Operators.setsum.
   637 INCOMPATIBILITY.
   638 
   639 * Theory HOL/Library/Diagonalize has been removed. INCOMPATIBILITY,
   640 use theory HOL/Library/Nat_Bijection instead.
   641 
   642 * Theory HOL/Library/RBT_Impl: Backing implementation of red-black
   643 trees is now inside a type class context.  Names of affected
   644 operations and lemmas have been prefixed by rbt_.  INCOMPATIBILITY for
   645 theories working directly with raw red-black trees, adapt the names as
   646 follows:
   647 
   648   Operations:
   649   bulkload -> rbt_bulkload
   650   del_from_left -> rbt_del_from_left
   651   del_from_right -> rbt_del_from_right
   652   del -> rbt_del
   653   delete -> rbt_delete
   654   ins -> rbt_ins
   655   insert -> rbt_insert
   656   insertw -> rbt_insert_with
   657   insert_with_key -> rbt_insert_with_key
   658   map_entry -> rbt_map_entry
   659   lookup -> rbt_lookup
   660   sorted -> rbt_sorted
   661   tree_greater -> rbt_greater
   662   tree_less -> rbt_less
   663   tree_less_symbol -> rbt_less_symbol
   664   union -> rbt_union
   665   union_with -> rbt_union_with
   666   union_with_key -> rbt_union_with_key
   667 
   668   Lemmas:
   669   balance_left_sorted -> balance_left_rbt_sorted
   670   balance_left_tree_greater -> balance_left_rbt_greater
   671   balance_left_tree_less -> balance_left_rbt_less
   672   balance_right_sorted -> balance_right_rbt_sorted
   673   balance_right_tree_greater -> balance_right_rbt_greater
   674   balance_right_tree_less -> balance_right_rbt_less
   675   balance_sorted -> balance_rbt_sorted
   676   balance_tree_greater -> balance_rbt_greater
   677   balance_tree_less -> balance_rbt_less
   678   bulkload_is_rbt -> rbt_bulkload_is_rbt
   679   combine_sorted -> combine_rbt_sorted
   680   combine_tree_greater -> combine_rbt_greater
   681   combine_tree_less -> combine_rbt_less
   682   delete_in_tree -> rbt_delete_in_tree
   683   delete_is_rbt -> rbt_delete_is_rbt
   684   del_from_left_tree_greater -> rbt_del_from_left_rbt_greater
   685   del_from_left_tree_less -> rbt_del_from_left_rbt_less
   686   del_from_right_tree_greater -> rbt_del_from_right_rbt_greater
   687   del_from_right_tree_less -> rbt_del_from_right_rbt_less
   688   del_in_tree -> rbt_del_in_tree
   689   del_inv1_inv2 -> rbt_del_inv1_inv2
   690   del_sorted -> rbt_del_rbt_sorted
   691   del_tree_greater -> rbt_del_rbt_greater
   692   del_tree_less -> rbt_del_rbt_less
   693   dom_lookup_Branch -> dom_rbt_lookup_Branch
   694   entries_lookup -> entries_rbt_lookup
   695   finite_dom_lookup -> finite_dom_rbt_lookup
   696   insert_sorted -> rbt_insert_rbt_sorted
   697   insertw_is_rbt -> rbt_insertw_is_rbt
   698   insertwk_is_rbt -> rbt_insertwk_is_rbt
   699   insertwk_sorted -> rbt_insertwk_rbt_sorted
   700   insertw_sorted -> rbt_insertw_rbt_sorted
   701   ins_sorted -> ins_rbt_sorted
   702   ins_tree_greater -> ins_rbt_greater
   703   ins_tree_less -> ins_rbt_less
   704   is_rbt_sorted -> is_rbt_rbt_sorted
   705   lookup_balance -> rbt_lookup_balance
   706   lookup_bulkload -> rbt_lookup_rbt_bulkload
   707   lookup_delete -> rbt_lookup_rbt_delete
   708   lookup_Empty -> rbt_lookup_Empty
   709   lookup_from_in_tree -> rbt_lookup_from_in_tree
   710   lookup_in_tree -> rbt_lookup_in_tree
   711   lookup_ins -> rbt_lookup_ins
   712   lookup_insert -> rbt_lookup_rbt_insert
   713   lookup_insertw -> rbt_lookup_rbt_insertw
   714   lookup_insertwk -> rbt_lookup_rbt_insertwk
   715   lookup_keys -> rbt_lookup_keys
   716   lookup_map -> rbt_lookup_map
   717   lookup_map_entry -> rbt_lookup_rbt_map_entry
   718   lookup_tree_greater -> rbt_lookup_rbt_greater
   719   lookup_tree_less -> rbt_lookup_rbt_less
   720   lookup_union -> rbt_lookup_rbt_union
   721   map_entry_color_of -> rbt_map_entry_color_of
   722   map_entry_inv1 -> rbt_map_entry_inv1
   723   map_entry_inv2 -> rbt_map_entry_inv2
   724   map_entry_is_rbt -> rbt_map_entry_is_rbt
   725   map_entry_sorted -> rbt_map_entry_rbt_sorted
   726   map_entry_tree_greater -> rbt_map_entry_rbt_greater
   727   map_entry_tree_less -> rbt_map_entry_rbt_less
   728   map_tree_greater -> map_rbt_greater
   729   map_tree_less -> map_rbt_less
   730   map_sorted -> map_rbt_sorted
   731   paint_sorted -> paint_rbt_sorted
   732   paint_lookup -> paint_rbt_lookup
   733   paint_tree_greater -> paint_rbt_greater
   734   paint_tree_less -> paint_rbt_less
   735   sorted_entries -> rbt_sorted_entries
   736   tree_greater_eq_trans -> rbt_greater_eq_trans
   737   tree_greater_nit -> rbt_greater_nit
   738   tree_greater_prop -> rbt_greater_prop
   739   tree_greater_simps -> rbt_greater_simps
   740   tree_greater_trans -> rbt_greater_trans
   741   tree_less_eq_trans -> rbt_less_eq_trans
   742   tree_less_nit -> rbt_less_nit
   743   tree_less_prop -> rbt_less_prop
   744   tree_less_simps -> rbt_less_simps
   745   tree_less_trans -> rbt_less_trans
   746   tree_ord_props -> rbt_ord_props
   747   union_Branch -> rbt_union_Branch
   748   union_is_rbt -> rbt_union_is_rbt
   749   unionw_is_rbt -> rbt_unionw_is_rbt
   750   unionwk_is_rbt -> rbt_unionwk_is_rbt
   751   unionwk_sorted -> rbt_unionwk_rbt_sorted
   752 
   753 * Theory HOL/Library/Float: Floating point numbers are now defined as
   754 a subset of the real numbers.  All operations are defined using the
   755 lifing-framework and proofs use the transfer method.  INCOMPATIBILITY.
   756 
   757   Changed Operations:
   758   float_abs -> abs
   759   float_nprt -> nprt
   760   float_pprt -> pprt
   761   pow2 -> use powr
   762   round_down -> float_round_down
   763   round_up -> float_round_up
   764   scale -> exponent
   765 
   766   Removed Operations:
   767   ceiling_fl, lb_mult, lb_mod, ub_mult, ub_mod
   768 
   769   Renamed Lemmas:
   770   abs_float_def -> Float.compute_float_abs
   771   bitlen_ge0 -> bitlen_nonneg
   772   bitlen.simps -> Float.compute_bitlen
   773   float_components -> Float_mantissa_exponent
   774   float_divl.simps -> Float.compute_float_divl
   775   float_divr.simps -> Float.compute_float_divr
   776   float_eq_odd -> mult_powr_eq_mult_powr_iff
   777   float_power -> real_of_float_power
   778   lapprox_posrat_def -> Float.compute_lapprox_posrat
   779   lapprox_rat.simps -> Float.compute_lapprox_rat
   780   le_float_def' -> Float.compute_float_le
   781   le_float_def -> less_eq_float.rep_eq
   782   less_float_def' -> Float.compute_float_less
   783   less_float_def -> less_float.rep_eq
   784   normfloat_def -> Float.compute_normfloat
   785   normfloat_imp_odd_or_zero -> mantissa_not_dvd and mantissa_noteq_0
   786   normfloat -> normfloat_def
   787   normfloat_unique -> use normfloat_def
   788   number_of_float_Float -> Float.compute_float_numeral, Float.compute_float_neg_numeral
   789   one_float_def -> Float.compute_float_one
   790   plus_float_def -> Float.compute_float_plus
   791   rapprox_posrat_def -> Float.compute_rapprox_posrat
   792   rapprox_rat.simps -> Float.compute_rapprox_rat
   793   real_of_float_0 -> zero_float.rep_eq
   794   real_of_float_1 -> one_float.rep_eq
   795   real_of_float_abs -> abs_float.rep_eq
   796   real_of_float_add -> plus_float.rep_eq
   797   real_of_float_minus -> uminus_float.rep_eq
   798   real_of_float_mult -> times_float.rep_eq
   799   real_of_float_simp -> Float.rep_eq
   800   real_of_float_sub -> minus_float.rep_eq
   801   round_down.simps -> Float.compute_float_round_down
   802   round_up.simps -> Float.compute_float_round_up
   803   times_float_def -> Float.compute_float_times
   804   uminus_float_def -> Float.compute_float_uminus
   805   zero_float_def -> Float.compute_float_zero
   806 
   807   Lemmas not necessary anymore, use the transfer method:
   808   bitlen_B0, bitlen_B1, bitlen_ge1, bitlen_Min, bitlen_Pls, float_divl,
   809   float_divr, float_le_simp, float_less1_mantissa_bound,
   810   float_less_simp, float_less_zero, float_le_zero,
   811   float_pos_less1_e_neg, float_pos_m_pos, float_split, float_split2,
   812   floor_pos_exp, lapprox_posrat, lapprox_posrat_bottom, lapprox_rat,
   813   lapprox_rat_bottom, normalized_float, rapprox_posrat,
   814   rapprox_posrat_le1, rapprox_rat, real_of_float_ge0_exp,
   815   real_of_float_neg_exp, real_of_float_nge0_exp, round_down floor_fl,
   816   round_up, zero_le_float, zero_less_float
   817 
   818 * New theory HOL/Library/DAList provides an abstract type for
   819 association lists with distinct keys.
   820 
   821 * Session HOL/IMP: Added new theory of abstract interpretation of
   822 annotated commands.
   823 
   824 * Session HOL-Import: Re-implementation from scratch is faster,
   825 simpler, and more scalable.  Requires a proof bundle, which is
   826 available as an external component.  Discontinued old (and mostly
   827 dead) Importer for HOL4 and HOL Light.  INCOMPATIBILITY.
   828 
   829 * Session HOL-Word: Discontinued many redundant theorems specific to
   830 type 'a word. INCOMPATIBILITY, use the corresponding generic theorems
   831 instead.
   832 
   833   word_sub_alt ~> word_sub_wi
   834   word_add_alt ~> word_add_def
   835   word_mult_alt ~> word_mult_def
   836   word_minus_alt ~> word_minus_def
   837   word_0_alt ~> word_0_wi
   838   word_1_alt ~> word_1_wi
   839   word_add_0 ~> add_0_left
   840   word_add_0_right ~> add_0_right
   841   word_mult_1 ~> mult_1_left
   842   word_mult_1_right ~> mult_1_right
   843   word_add_commute ~> add_commute
   844   word_add_assoc ~> add_assoc
   845   word_add_left_commute ~> add_left_commute
   846   word_mult_commute ~> mult_commute
   847   word_mult_assoc ~> mult_assoc
   848   word_mult_left_commute ~> mult_left_commute
   849   word_left_distrib ~> left_distrib
   850   word_right_distrib ~> right_distrib
   851   word_left_minus ~> left_minus
   852   word_diff_0_right ~> diff_0_right
   853   word_diff_self ~> diff_self
   854   word_sub_def ~> diff_minus
   855   word_diff_minus ~> diff_minus
   856   word_add_ac ~> add_ac
   857   word_mult_ac ~> mult_ac
   858   word_plus_ac0 ~> add_0_left add_0_right add_ac
   859   word_times_ac1 ~> mult_1_left mult_1_right mult_ac
   860   word_order_trans ~> order_trans
   861   word_order_refl ~> order_refl
   862   word_order_antisym ~> order_antisym
   863   word_order_linear ~> linorder_linear
   864   lenw1_zero_neq_one ~> zero_neq_one
   865   word_number_of_eq ~> number_of_eq
   866   word_of_int_add_hom ~> wi_hom_add
   867   word_of_int_sub_hom ~> wi_hom_sub
   868   word_of_int_mult_hom ~> wi_hom_mult
   869   word_of_int_minus_hom ~> wi_hom_neg
   870   word_of_int_succ_hom ~> wi_hom_succ
   871   word_of_int_pred_hom ~> wi_hom_pred
   872   word_of_int_0_hom ~> word_0_wi
   873   word_of_int_1_hom ~> word_1_wi
   874 
   875 * Session HOL-Word: New proof method "word_bitwise" for splitting
   876 machine word equalities and inequalities into logical circuits,
   877 defined in HOL/Word/WordBitwise.thy.  Supports addition, subtraction,
   878 multiplication, shifting by constants, bitwise operators and numeric
   879 constants.  Requires fixed-length word types, not 'a word.  Solves
   880 many standard word identities outright and converts more into first
   881 order problems amenable to blast or similar.  See also examples in
   882 HOL/Word/Examples/WordExamples.thy.
   883 
   884 * Session HOL-Probability: Introduced the type "'a measure" to
   885 represent measures, this replaces the records 'a algebra and 'a
   886 measure_space.  The locales based on subset_class now have two
   887 locale-parameters the space \<Omega> and the set of measurable sets M.
   888 The product of probability spaces uses now the same constant as the
   889 finite product of sigma-finite measure spaces "PiM :: ('i => 'a)
   890 measure".  Most constants are defined now outside of locales and gain
   891 an additional parameter, like null_sets, almost_eventually or \<mu>'.
   892 Measure space constructions for distributions and densities now got
   893 their own constants distr and density.  Instead of using locales to
   894 describe measure spaces with a finite space, the measure count_space
   895 and point_measure is introduced.  INCOMPATIBILITY.
   896 
   897   Renamed constants:
   898   measure -> emeasure
   899   finite_measure.\<mu>' -> measure
   900   product_algebra_generator -> prod_algebra
   901   product_prob_space.emb -> prod_emb
   902   product_prob_space.infprod_algebra -> PiM
   903 
   904   Removed locales:
   905   completeable_measure_space
   906   finite_measure_space
   907   finite_prob_space
   908   finite_product_finite_prob_space
   909   finite_product_sigma_algebra
   910   finite_sigma_algebra
   911   measure_space
   912   pair_finite_prob_space
   913   pair_finite_sigma_algebra
   914   pair_finite_space
   915   pair_sigma_algebra
   916   product_sigma_algebra
   917 
   918   Removed constants:
   919   conditional_space
   920   distribution -> use distr measure, or distributed predicate
   921   image_space
   922   joint_distribution -> use distr measure, or distributed predicate
   923   pair_measure_generator
   924   product_prob_space.infprod_algebra -> use PiM
   925   subvimage
   926 
   927   Replacement theorems:
   928   finite_additivity_sufficient -> ring_of_sets.countably_additiveI_finite
   929   finite_measure.empty_measure -> measure_empty
   930   finite_measure.finite_continuity_from_above -> finite_measure.finite_Lim_measure_decseq
   931   finite_measure.finite_continuity_from_below -> finite_measure.finite_Lim_measure_incseq
   932   finite_measure.finite_measure_countably_subadditive -> finite_measure.finite_measure_subadditive_countably
   933   finite_measure.finite_measure_eq -> finite_measure.emeasure_eq_measure
   934   finite_measure.finite_measure -> finite_measure.emeasure_finite
   935   finite_measure.finite_measure_finite_singleton -> finite_measure.finite_measure_eq_setsum_singleton
   936   finite_measure.positive_measure' -> measure_nonneg
   937   finite_measure.real_measure -> finite_measure.emeasure_real
   938   finite_product_prob_space.finite_measure_times -> finite_product_prob_space.finite_measure_PiM_emb
   939   finite_product_sigma_algebra.in_P -> sets_PiM_I_finite
   940   finite_product_sigma_algebra.P_empty -> space_PiM_empty, sets_PiM_empty
   941   information_space.conditional_entropy_eq -> information_space.conditional_entropy_simple_distributed
   942   information_space.conditional_entropy_positive -> information_space.conditional_entropy_nonneg_simple
   943   information_space.conditional_mutual_information_eq_mutual_information -> information_space.conditional_mutual_information_eq_mutual_information_simple
   944   information_space.conditional_mutual_information_generic_positive -> information_space.conditional_mutual_information_nonneg_simple
   945   information_space.conditional_mutual_information_positive -> information_space.conditional_mutual_information_nonneg_simple
   946   information_space.entropy_commute -> information_space.entropy_commute_simple
   947   information_space.entropy_eq -> information_space.entropy_simple_distributed
   948   information_space.entropy_generic_eq -> information_space.entropy_simple_distributed
   949   information_space.entropy_positive -> information_space.entropy_nonneg_simple
   950   information_space.entropy_uniform_max -> information_space.entropy_uniform
   951   information_space.KL_eq_0_imp -> information_space.KL_eq_0_iff_eq
   952   information_space.KL_eq_0 -> information_space.KL_same_eq_0
   953   information_space.KL_ge_0 -> information_space.KL_nonneg
   954   information_space.mutual_information_eq -> information_space.mutual_information_simple_distributed
   955   information_space.mutual_information_positive -> information_space.mutual_information_nonneg_simple
   956   Int_stable_cuboids -> Int_stable_atLeastAtMost
   957   Int_stable_product_algebra_generator -> positive_integral
   958   measure_preserving -> equality "distr M N f = N" "f : measurable M N"
   959   measure_space.additive -> emeasure_additive
   960   measure_space.AE_iff_null_set -> AE_iff_null
   961   measure_space.almost_everywhere_def -> eventually_ae_filter
   962   measure_space.almost_everywhere_vimage -> AE_distrD
   963   measure_space.continuity_from_above -> INF_emeasure_decseq
   964   measure_space.continuity_from_above_Lim -> Lim_emeasure_decseq
   965   measure_space.continuity_from_below_Lim -> Lim_emeasure_incseq
   966   measure_space.continuity_from_below -> SUP_emeasure_incseq
   967   measure_space_density -> emeasure_density
   968   measure_space.density_is_absolutely_continuous -> absolutely_continuousI_density
   969   measure_space.integrable_vimage -> integrable_distr
   970   measure_space.integral_translated_density -> integral_density
   971   measure_space.integral_vimage -> integral_distr
   972   measure_space.measure_additive -> plus_emeasure
   973   measure_space.measure_compl -> emeasure_compl
   974   measure_space.measure_countable_increasing -> emeasure_countable_increasing
   975   measure_space.measure_countably_subadditive -> emeasure_subadditive_countably
   976   measure_space.measure_decseq -> decseq_emeasure
   977   measure_space.measure_Diff -> emeasure_Diff
   978   measure_space.measure_Diff_null_set -> emeasure_Diff_null_set
   979   measure_space.measure_eq_0 -> emeasure_eq_0
   980   measure_space.measure_finitely_subadditive -> emeasure_subadditive_finite
   981   measure_space.measure_finite_singleton -> emeasure_eq_setsum_singleton
   982   measure_space.measure_incseq -> incseq_emeasure
   983   measure_space.measure_insert -> emeasure_insert
   984   measure_space.measure_mono -> emeasure_mono
   985   measure_space.measure_not_negative -> emeasure_not_MInf
   986   measure_space.measure_preserving_Int_stable -> measure_eqI_generator_eq
   987   measure_space.measure_setsum -> setsum_emeasure
   988   measure_space.measure_setsum_split -> setsum_emeasure_cover
   989   measure_space.measure_space_vimage -> emeasure_distr
   990   measure_space.measure_subadditive_finite -> emeasure_subadditive_finite
   991   measure_space.measure_subadditive -> subadditive
   992   measure_space.measure_top -> emeasure_space
   993   measure_space.measure_UN_eq_0 -> emeasure_UN_eq_0
   994   measure_space.measure_Un_null_set -> emeasure_Un_null_set
   995   measure_space.positive_integral_translated_density -> positive_integral_density
   996   measure_space.positive_integral_vimage -> positive_integral_distr
   997   measure_space.real_continuity_from_above -> Lim_measure_decseq
   998   measure_space.real_continuity_from_below -> Lim_measure_incseq
   999   measure_space.real_measure_countably_subadditive -> measure_subadditive_countably
  1000   measure_space.real_measure_Diff -> measure_Diff
  1001   measure_space.real_measure_finite_Union -> measure_finite_Union
  1002   measure_space.real_measure_setsum_singleton -> measure_eq_setsum_singleton
  1003   measure_space.real_measure_subadditive -> measure_subadditive
  1004   measure_space.real_measure_Union -> measure_Union
  1005   measure_space.real_measure_UNION -> measure_UNION
  1006   measure_space.simple_function_vimage -> simple_function_comp
  1007   measure_space.simple_integral_vimage -> simple_integral_distr
  1008   measure_space.simple_integral_vimage -> simple_integral_distr
  1009   measure_unique_Int_stable -> measure_eqI_generator_eq
  1010   measure_unique_Int_stable_vimage -> measure_eqI_generator_eq
  1011   pair_sigma_algebra.measurable_cut_fst -> sets_Pair1
  1012   pair_sigma_algebra.measurable_cut_snd -> sets_Pair2
  1013   pair_sigma_algebra.measurable_pair_image_fst -> measurable_Pair1
  1014   pair_sigma_algebra.measurable_pair_image_snd -> measurable_Pair2
  1015   pair_sigma_algebra.measurable_product_swap -> measurable_pair_swap_iff
  1016   pair_sigma_algebra.pair_sigma_algebra_measurable -> measurable_pair_swap
  1017   pair_sigma_algebra.pair_sigma_algebra_swap_measurable -> measurable_pair_swap'
  1018   pair_sigma_algebra.sets_swap -> sets_pair_swap
  1019   pair_sigma_finite.measure_cut_measurable_fst -> pair_sigma_finite.measurable_emeasure_Pair1
  1020   pair_sigma_finite.measure_cut_measurable_snd -> pair_sigma_finite.measurable_emeasure_Pair2
  1021   pair_sigma_finite.measure_preserving_swap -> pair_sigma_finite.distr_pair_swap
  1022   pair_sigma_finite.pair_measure_alt2 -> pair_sigma_finite.emeasure_pair_measure_alt2
  1023   pair_sigma_finite.pair_measure_alt -> pair_sigma_finite.emeasure_pair_measure_alt
  1024   pair_sigma_finite.pair_measure_times -> pair_sigma_finite.emeasure_pair_measure_Times
  1025   prob_space.indep_distribution_eq_measure -> prob_space.indep_vars_iff_distr_eq_PiM
  1026   prob_space.indep_var_distributionD -> prob_space.indep_var_distribution_eq
  1027   prob_space.measure_space_1 -> prob_space.emeasure_space_1
  1028   prob_space.prob_space_vimage -> prob_space_distr
  1029   prob_space.random_variable_restrict -> measurable_restrict
  1030   prob_space_unique_Int_stable -> measure_eqI_prob_space
  1031   product_algebraE -> prod_algebraE_all
  1032   product_algebra_generator_der -> prod_algebra_eq_finite
  1033   product_algebra_generator_into_space -> prod_algebra_sets_into_space
  1034   product_algebraI -> sets_PiM_I_finite
  1035   product_measure_exists -> product_sigma_finite.sigma_finite
  1036   product_prob_space.finite_index_eq_finite_product -> product_prob_space.sets_PiM_generator
  1037   product_prob_space.finite_measure_infprod_emb_Pi -> product_prob_space.measure_PiM_emb
  1038   product_prob_space.infprod_spec -> product_prob_space.emeasure_PiM_emb_not_empty
  1039   product_prob_space.measurable_component -> measurable_component_singleton
  1040   product_prob_space.measurable_emb -> measurable_prod_emb
  1041   product_prob_space.measurable_into_infprod_algebra -> measurable_PiM_single
  1042   product_prob_space.measurable_singleton_infprod -> measurable_component_singleton
  1043   product_prob_space.measure_emb -> emeasure_prod_emb
  1044   product_prob_space.measure_preserving_restrict -> product_prob_space.distr_restrict
  1045   product_sigma_algebra.product_algebra_into_space -> space_closed
  1046   product_sigma_finite.measure_fold -> product_sigma_finite.distr_merge
  1047   product_sigma_finite.measure_preserving_component_singelton -> product_sigma_finite.distr_singleton
  1048   product_sigma_finite.measure_preserving_merge -> product_sigma_finite.distr_merge
  1049   sequence_space.measure_infprod -> sequence_space.measure_PiM_countable
  1050   sets_product_algebra -> sets_PiM
  1051   sigma_algebra.measurable_sigma -> measurable_measure_of
  1052   sigma_finite_measure.disjoint_sigma_finite -> sigma_finite_disjoint
  1053   sigma_finite_measure.RN_deriv_vimage -> sigma_finite_measure.RN_deriv_distr
  1054   sigma_product_algebra_sigma_eq -> sigma_prod_algebra_sigma_eq
  1055   space_product_algebra -> space_PiM
  1056 
  1057 * Session HOL-TPTP: support to parse and import TPTP problems (all
  1058 languages) into Isabelle/HOL.
  1059 
  1060 
  1061 *** FOL ***
  1062 
  1063 * New "case_product" attribute (see HOL).
  1064 
  1065 
  1066 *** ZF ***
  1067 
  1068 * Greater support for structured proofs involving induction or case
  1069 analysis.
  1070 
  1071 * Much greater use of mathematical symbols.
  1072 
  1073 * Removal of many ML theorem bindings.  INCOMPATIBILITY.
  1074 
  1075 
  1076 *** ML ***
  1077 
  1078 * Antiquotation @{keyword "name"} produces a parser for outer syntax
  1079 from a minor keyword introduced via theory header declaration.
  1080 
  1081 * Antiquotation @{command_spec "name"} produces the
  1082 Outer_Syntax.command_spec from a major keyword introduced via theory
  1083 header declaration; it can be passed to Outer_Syntax.command etc.
  1084 
  1085 * Local_Theory.define no longer hard-wires default theorem name
  1086 "foo_def", but retains the binding as given.  If that is Binding.empty
  1087 / Attrib.empty_binding, the result is not registered as user-level
  1088 fact.  The Local_Theory.define_internal variant allows to specify a
  1089 non-empty name (used for the foundation in the background theory),
  1090 while omitting the fact binding in the user-context.  Potential
  1091 INCOMPATIBILITY for derived definitional packages: need to specify
  1092 naming policy for primitive definitions more explicitly.
  1093 
  1094 * Renamed Thm.capply to Thm.apply, and Thm.cabs to Thm.lambda in
  1095 conformance with similar operations in structure Term and Logic.
  1096 
  1097 * Antiquotation @{attributes [...]} embeds attribute source
  1098 representation into the ML text, which is particularly useful with
  1099 declarations like Local_Theory.note.
  1100 
  1101 * Structure Proof_Context follows standard naming scheme.  Old
  1102 ProofContext has been discontinued.  INCOMPATIBILITY.
  1103 
  1104 * Refined Local_Theory.declaration {syntax, pervasive}, with subtle
  1105 change of semantics: update is applied to auxiliary local theory
  1106 context as well.
  1107 
  1108 * Modernized some old-style infix operations:
  1109 
  1110   addeqcongs    ~> Simplifier.add_eqcong
  1111   deleqcongs    ~> Simplifier.del_eqcong
  1112   addcongs      ~> Simplifier.add_cong
  1113   delcongs      ~> Simplifier.del_cong
  1114   setmksimps    ~> Simplifier.set_mksimps
  1115   setmkcong     ~> Simplifier.set_mkcong
  1116   setmksym      ~> Simplifier.set_mksym
  1117   setmkeqTrue   ~> Simplifier.set_mkeqTrue
  1118   settermless   ~> Simplifier.set_termless
  1119   setsubgoaler  ~> Simplifier.set_subgoaler
  1120   addsplits     ~> Splitter.add_split
  1121   delsplits     ~> Splitter.del_split
  1122 
  1123 
  1124 *** System ***
  1125 
  1126 * USER_HOME settings variable points to cross-platform user home
  1127 directory, which coincides with HOME on POSIX systems only.  Likewise,
  1128 the Isabelle path specification "~" now expands to $USER_HOME, instead
  1129 of former $HOME.  A different default for USER_HOME may be set
  1130 explicitly in shell environment, before Isabelle settings are
  1131 evaluated.  Minor INCOMPATIBILITY: need to adapt Isabelle path where
  1132 the generic user home was intended.
  1133 
  1134 * ISABELLE_HOME_WINDOWS refers to ISABELLE_HOME in windows file name
  1135 notation, which is useful for the jEdit file browser, for example.
  1136 
  1137 * ISABELLE_JDK_HOME settings variable points to JDK with javac and jar
  1138 (not just JRE).
  1139 
  1140 
  1141 
  1142 New in Isabelle2011-1 (October 2011)
  1143 ------------------------------------
  1144 
  1145 *** General ***
  1146 
  1147 * Improved Isabelle/jEdit Prover IDE (PIDE), which can be invoked as
  1148 "isabelle jedit" or "ISABELLE_HOME/Isabelle" on the command line.
  1149 
  1150   - Management of multiple theory files directly from the editor
  1151     buffer store -- bypassing the file-system (no requirement to save
  1152     files for checking).
  1153 
  1154   - Markup of formal entities within the text buffer, with semantic
  1155     highlighting, tooltips and hyperlinks to jump to defining source
  1156     positions.
  1157 
  1158   - Improved text rendering, with sub/superscripts in the source
  1159     buffer (including support for copy/paste wrt. output panel, HTML
  1160     theory output and other non-Isabelle text boxes).
  1161 
  1162   - Refined scheduling of proof checking and printing of results,
  1163     based on interactive editor view.  (Note: jEdit folding and
  1164     narrowing allows to restrict buffer perspectives explicitly.)
  1165 
  1166   - Reduced CPU performance requirements, usable on machines with few
  1167     cores.
  1168 
  1169   - Reduced memory requirements due to pruning of unused document
  1170     versions (garbage collection).
  1171 
  1172 See also ~~/src/Tools/jEdit/README.html for further information,
  1173 including some remaining limitations.
  1174 
  1175 * Theory loader: source files are exclusively located via the master
  1176 directory of each theory node (where the .thy file itself resides).
  1177 The global load path (such as src/HOL/Library) has been discontinued.
  1178 Note that the path element ~~ may be used to reference theories in the
  1179 Isabelle home folder -- for instance, "~~/src/HOL/Library/FuncSet".
  1180 INCOMPATIBILITY.
  1181 
  1182 * Theory loader: source files are identified by content via SHA1
  1183 digests.  Discontinued former path/modtime identification and optional
  1184 ISABELLE_FILE_IDENT plugin scripts.
  1185 
  1186 * Parallelization of nested Isar proofs is subject to
  1187 Goal.parallel_proofs_threshold (default 100).  See also isabelle
  1188 usedir option -Q.
  1189 
  1190 * Name space: former unsynchronized references are now proper
  1191 configuration options, with more conventional names:
  1192 
  1193   long_names   ~> names_long
  1194   short_names  ~> names_short
  1195   unique_names ~> names_unique
  1196 
  1197 Minor INCOMPATIBILITY, need to declare options in context like this:
  1198 
  1199   declare [[names_unique = false]]
  1200 
  1201 * Literal facts `prop` may contain dummy patterns, e.g. `_ = _`.  Note
  1202 that the result needs to be unique, which means fact specifications
  1203 may have to be refined after enriching a proof context.
  1204 
  1205 * Attribute "case_names" has been refined: the assumptions in each case
  1206 can be named now by following the case name with [name1 name2 ...].
  1207 
  1208 * Isabelle/Isar reference manual has been updated and extended:
  1209   - "Synopsis" provides a catalog of main Isar language concepts.
  1210   - Formal references in syntax diagrams, via @{rail} antiquotation.
  1211   - Updated material from classic "ref" manual, notably about
  1212     "Classical Reasoner".
  1213 
  1214 
  1215 *** HOL ***
  1216 
  1217 * Class bot and top require underlying partial order rather than
  1218 preorder: uniqueness of bot and top is guaranteed.  INCOMPATIBILITY.
  1219 
  1220 * Class complete_lattice: generalized a couple of lemmas from sets;
  1221 generalized theorems INF_cong and SUP_cong.  New type classes for
  1222 complete boolean algebras and complete linear orders.  Lemmas
  1223 Inf_less_iff, less_Sup_iff, INF_less_iff, less_SUP_iff now reside in
  1224 class complete_linorder.
  1225 
  1226 Changed proposition of lemmas Inf_bool_def, Sup_bool_def, Inf_fun_def,
  1227 Sup_fun_def, Inf_apply, Sup_apply.
  1228 
  1229 Removed redundant lemmas (the right hand side gives hints how to
  1230 replace them for (metis ...), or (simp only: ...) proofs):
  1231 
  1232   Inf_singleton ~> Inf_insert [where A="{}", unfolded Inf_empty inf_top_right]
  1233   Sup_singleton ~> Sup_insert [where A="{}", unfolded Sup_empty sup_bot_right]
  1234   Inf_binary ~> Inf_insert, Inf_empty, and inf_top_right
  1235   Sup_binary ~> Sup_insert, Sup_empty, and sup_bot_right
  1236   Int_eq_Inter ~> Inf_insert, Inf_empty, and inf_top_right
  1237   Un_eq_Union ~> Sup_insert, Sup_empty, and sup_bot_right
  1238   Inter_def ~> INF_def, image_def
  1239   Union_def ~> SUP_def, image_def
  1240   INT_eq ~> INF_def, and image_def
  1241   UN_eq ~> SUP_def, and image_def
  1242   INF_subset ~> INF_superset_mono [OF _ order_refl]
  1243 
  1244 More consistent and comprehensive names:
  1245 
  1246   INTER_eq_Inter_image ~> INF_def
  1247   UNION_eq_Union_image ~> SUP_def
  1248   INFI_def ~> INF_def
  1249   SUPR_def ~> SUP_def
  1250   INF_leI ~> INF_lower
  1251   INF_leI2 ~> INF_lower2
  1252   le_INFI ~> INF_greatest
  1253   le_SUPI ~> SUP_upper
  1254   le_SUPI2 ~> SUP_upper2
  1255   SUP_leI ~> SUP_least
  1256   INFI_bool_eq ~> INF_bool_eq
  1257   SUPR_bool_eq ~> SUP_bool_eq
  1258   INFI_apply ~> INF_apply
  1259   SUPR_apply ~> SUP_apply
  1260   INTER_def ~> INTER_eq
  1261   UNION_def ~> UNION_eq
  1262 
  1263 INCOMPATIBILITY.
  1264 
  1265 * Renamed theory Complete_Lattice to Complete_Lattices.
  1266 INCOMPATIBILITY.
  1267 
  1268 * Theory Complete_Lattices: lemmas Inf_eq_top_iff, INF_eq_top_iff,
  1269 INF_image, Inf_insert, INF_top, Inf_top_conv, INF_top_conv, SUP_bot,
  1270 Sup_bot_conv, SUP_bot_conv, Sup_eq_top_iff, SUP_eq_top_iff, SUP_image,
  1271 Sup_insert are now declared as [simp].  INCOMPATIBILITY.
  1272 
  1273 * Theory Lattice: lemmas compl_inf_bot, compl_le_comp_iff,
  1274 compl_sup_top, inf_idem, inf_left_idem, inf_sup_absorb, sup_idem,
  1275 sup_inf_absob, sup_left_idem are now declared as [simp].  Minor
  1276 INCOMPATIBILITY.
  1277 
  1278 * Added syntactic classes "inf" and "sup" for the respective
  1279 constants.  INCOMPATIBILITY: Changes in the argument order of the
  1280 (mostly internal) locale predicates for some derived classes.
  1281 
  1282 * Theorem collections ball_simps and bex_simps do not contain theorems
  1283 referring to UNION any longer; these have been moved to collection
  1284 UN_ball_bex_simps.  INCOMPATIBILITY.
  1285 
  1286 * Theory Archimedean_Field: floor now is defined as parameter of a
  1287 separate type class floor_ceiling.
  1288 
  1289 * Theory Finite_Set: more coherent development of fold_set locales:
  1290 
  1291     locale fun_left_comm ~> locale comp_fun_commute
  1292     locale fun_left_comm_idem ~> locale comp_fun_idem
  1293 
  1294 Both use point-free characterization; interpretation proofs may need
  1295 adjustment.  INCOMPATIBILITY.
  1296 
  1297 * Theory Limits: Type "'a net" has been renamed to "'a filter", in
  1298 accordance with standard mathematical terminology. INCOMPATIBILITY.
  1299 
  1300 * Theory Complex_Main: The locale interpretations for the
  1301 bounded_linear and bounded_bilinear locales have been removed, in
  1302 order to reduce the number of duplicate lemmas. Users must use the
  1303 original names for distributivity theorems, potential INCOMPATIBILITY.
  1304 
  1305   divide.add ~> add_divide_distrib
  1306   divide.diff ~> diff_divide_distrib
  1307   divide.setsum ~> setsum_divide_distrib
  1308   mult.add_right ~> right_distrib
  1309   mult.diff_right ~> right_diff_distrib
  1310   mult_right.setsum ~> setsum_right_distrib
  1311   mult_left.diff ~> left_diff_distrib
  1312 
  1313 * Theory Complex_Main: Several redundant theorems have been removed or
  1314 replaced by more general versions. INCOMPATIBILITY.
  1315 
  1316   real_diff_def ~> minus_real_def
  1317   real_divide_def ~> divide_real_def
  1318   real_less_def ~> less_le
  1319   real_abs_def ~> abs_real_def
  1320   real_sgn_def ~> sgn_real_def
  1321   real_mult_commute ~> mult_commute
  1322   real_mult_assoc ~> mult_assoc
  1323   real_mult_1 ~> mult_1_left
  1324   real_add_mult_distrib ~> left_distrib
  1325   real_zero_not_eq_one ~> zero_neq_one
  1326   real_mult_inverse_left ~> left_inverse
  1327   INVERSE_ZERO ~> inverse_zero
  1328   real_le_refl ~> order_refl
  1329   real_le_antisym ~> order_antisym
  1330   real_le_trans ~> order_trans
  1331   real_le_linear ~> linear
  1332   real_le_eq_diff ~> le_iff_diff_le_0
  1333   real_add_left_mono ~> add_left_mono
  1334   real_mult_order ~> mult_pos_pos
  1335   real_mult_less_mono2 ~> mult_strict_left_mono
  1336   real_of_int_real_of_nat ~> real_of_int_of_nat_eq
  1337   real_0_le_divide_iff ~> zero_le_divide_iff
  1338   realpow_two_disj ~> power2_eq_iff
  1339   real_squared_diff_one_factored ~> square_diff_one_factored
  1340   realpow_two_diff ~> square_diff_square_factored
  1341   reals_complete2 ~> complete_real
  1342   real_sum_squared_expand ~> power2_sum
  1343   exp_ln_eq ~> ln_unique
  1344   expi_add ~> exp_add
  1345   expi_zero ~> exp_zero
  1346   lemma_DERIV_subst ~> DERIV_cong
  1347   LIMSEQ_Zfun_iff ~> tendsto_Zfun_iff
  1348   LIMSEQ_const ~> tendsto_const
  1349   LIMSEQ_norm ~> tendsto_norm
  1350   LIMSEQ_add ~> tendsto_add
  1351   LIMSEQ_minus ~> tendsto_minus
  1352   LIMSEQ_minus_cancel ~> tendsto_minus_cancel
  1353   LIMSEQ_diff ~> tendsto_diff
  1354   bounded_linear.LIMSEQ ~> bounded_linear.tendsto
  1355   bounded_bilinear.LIMSEQ ~> bounded_bilinear.tendsto
  1356   LIMSEQ_mult ~> tendsto_mult
  1357   LIMSEQ_inverse ~> tendsto_inverse
  1358   LIMSEQ_divide ~> tendsto_divide
  1359   LIMSEQ_pow ~> tendsto_power
  1360   LIMSEQ_setsum ~> tendsto_setsum
  1361   LIMSEQ_setprod ~> tendsto_setprod
  1362   LIMSEQ_norm_zero ~> tendsto_norm_zero_iff
  1363   LIMSEQ_rabs_zero ~> tendsto_rabs_zero_iff
  1364   LIMSEQ_imp_rabs ~> tendsto_rabs
  1365   LIMSEQ_add_minus ~> tendsto_add [OF _ tendsto_minus]
  1366   LIMSEQ_add_const ~> tendsto_add [OF _ tendsto_const]
  1367   LIMSEQ_diff_const ~> tendsto_diff [OF _ tendsto_const]
  1368   LIMSEQ_Complex ~> tendsto_Complex
  1369   LIM_ident ~> tendsto_ident_at
  1370   LIM_const ~> tendsto_const
  1371   LIM_add ~> tendsto_add
  1372   LIM_add_zero ~> tendsto_add_zero
  1373   LIM_minus ~> tendsto_minus
  1374   LIM_diff ~> tendsto_diff
  1375   LIM_norm ~> tendsto_norm
  1376   LIM_norm_zero ~> tendsto_norm_zero
  1377   LIM_norm_zero_cancel ~> tendsto_norm_zero_cancel
  1378   LIM_norm_zero_iff ~> tendsto_norm_zero_iff
  1379   LIM_rabs ~> tendsto_rabs
  1380   LIM_rabs_zero ~> tendsto_rabs_zero
  1381   LIM_rabs_zero_cancel ~> tendsto_rabs_zero_cancel
  1382   LIM_rabs_zero_iff ~> tendsto_rabs_zero_iff
  1383   LIM_compose ~> tendsto_compose
  1384   LIM_mult ~> tendsto_mult
  1385   LIM_scaleR ~> tendsto_scaleR
  1386   LIM_of_real ~> tendsto_of_real
  1387   LIM_power ~> tendsto_power
  1388   LIM_inverse ~> tendsto_inverse
  1389   LIM_sgn ~> tendsto_sgn
  1390   isCont_LIM_compose ~> isCont_tendsto_compose
  1391   bounded_linear.LIM ~> bounded_linear.tendsto
  1392   bounded_linear.LIM_zero ~> bounded_linear.tendsto_zero
  1393   bounded_bilinear.LIM ~> bounded_bilinear.tendsto
  1394   bounded_bilinear.LIM_prod_zero ~> bounded_bilinear.tendsto_zero
  1395   bounded_bilinear.LIM_left_zero ~> bounded_bilinear.tendsto_left_zero
  1396   bounded_bilinear.LIM_right_zero ~> bounded_bilinear.tendsto_right_zero
  1397   LIM_inverse_fun ~> tendsto_inverse [OF tendsto_ident_at]
  1398 
  1399 * Theory Complex_Main: The definition of infinite series was
  1400 generalized.  Now it is defined on the type class {topological_space,
  1401 comm_monoid_add}.  Hence it is useable also for extended real numbers.
  1402 
  1403 * Theory Complex_Main: The complex exponential function "expi" is now
  1404 a type-constrained abbreviation for "exp :: complex => complex"; thus
  1405 several polymorphic lemmas about "exp" are now applicable to "expi".
  1406 
  1407 * Code generation:
  1408 
  1409   - Theory Library/Code_Char_ord provides native ordering of
  1410     characters in the target language.
  1411 
  1412   - Commands code_module and code_library are legacy, use export_code
  1413     instead.
  1414 
  1415   - Method "evaluation" is legacy, use method "eval" instead.
  1416 
  1417   - Legacy evaluator "SML" is deactivated by default.  May be
  1418     reactivated by the following theory command:
  1419 
  1420       setup {* Value.add_evaluator ("SML", Codegen.eval_term) *}
  1421 
  1422 * Declare ext [intro] by default.  Rare INCOMPATIBILITY.
  1423 
  1424 * New proof method "induction" that gives induction hypotheses the
  1425 name "IH", thus distinguishing them from further hypotheses that come
  1426 from rule induction.  The latter are still called "hyps".  Method
  1427 "induction" is a thin wrapper around "induct" and follows the same
  1428 syntax.
  1429 
  1430 * Method "fastsimp" has been renamed to "fastforce", but "fastsimp" is
  1431 still available as a legacy feature for some time.
  1432 
  1433 * Nitpick:
  1434   - Added "need" and "total_consts" options.
  1435   - Reintroduced "show_skolems" option by popular demand.
  1436   - Renamed attribute: nitpick_def ~> nitpick_unfold.
  1437     INCOMPATIBILITY.
  1438 
  1439 * Sledgehammer:
  1440   - Use quasi-sound (and efficient) translations by default.
  1441   - Added support for the following provers: E-ToFoF, LEO-II,
  1442     Satallax, SNARK, Waldmeister, and Z3 with TPTP syntax.
  1443   - Automatically preplay and minimize proofs before showing them if
  1444     this can be done within reasonable time.
  1445   - sledgehammer available_provers ~> sledgehammer supported_provers.
  1446     INCOMPATIBILITY.
  1447   - Added "preplay_timeout", "slicing", "type_enc", "sound",
  1448     "max_mono_iters", and "max_new_mono_instances" options.
  1449   - Removed "explicit_apply" and "full_types" options as well as "Full
  1450     Types" Proof General menu item. INCOMPATIBILITY.
  1451 
  1452 * Metis:
  1453   - Removed "metisF" -- use "metis" instead. INCOMPATIBILITY.
  1454   - Obsoleted "metisFT" -- use "metis (full_types)" instead.
  1455     INCOMPATIBILITY.
  1456 
  1457 * Command 'try':
  1458   - Renamed 'try_methods' and added "simp:", "intro:", "dest:", and
  1459     "elim:" options. INCOMPATIBILITY.
  1460   - Introduced 'try' that not only runs 'try_methods' but also
  1461     'solve_direct', 'sledgehammer', 'quickcheck', and 'nitpick'.
  1462 
  1463 * Quickcheck:
  1464   - Added "eval" option to evaluate terms for the found counterexample
  1465     (currently only supported by the default (exhaustive) tester).
  1466   - Added post-processing of terms to obtain readable counterexamples
  1467     (currently only supported by the default (exhaustive) tester).
  1468   - New counterexample generator quickcheck[narrowing] enables
  1469     narrowing-based testing.  Requires the Glasgow Haskell compiler
  1470     with its installation location defined in the Isabelle settings
  1471     environment as ISABELLE_GHC.
  1472   - Removed quickcheck tester "SML" based on the SML code generator
  1473     (formly in HOL/Library).
  1474 
  1475 * Function package: discontinued option "tailrec".  INCOMPATIBILITY,
  1476 use 'partial_function' instead.
  1477 
  1478 * Theory Library/Extended_Reals replaces now the positive extended
  1479 reals found in probability theory. This file is extended by
  1480 Multivariate_Analysis/Extended_Real_Limits.
  1481 
  1482 * Theory Library/Old_Recdef: old 'recdef' package has been moved here,
  1483 from where it must be imported explicitly if it is really required.
  1484 INCOMPATIBILITY.
  1485 
  1486 * Theory Library/Wfrec: well-founded recursion combinator "wfrec" has
  1487 been moved here.  INCOMPATIBILITY.
  1488 
  1489 * Theory Library/Saturated provides type of numbers with saturated
  1490 arithmetic.
  1491 
  1492 * Theory Library/Product_Lattice defines a pointwise ordering for the
  1493 product type 'a * 'b, and provides instance proofs for various order
  1494 and lattice type classes.
  1495 
  1496 * Theory Library/Countable now provides the "countable_datatype" proof
  1497 method for proving "countable" class instances for datatypes.
  1498 
  1499 * Theory Library/Cset_Monad allows do notation for computable sets
  1500 (cset) via the generic monad ad-hoc overloading facility.
  1501 
  1502 * Library: Theories of common data structures are split into theories
  1503 for implementation, an invariant-ensuring type, and connection to an
  1504 abstract type. INCOMPATIBILITY.
  1505 
  1506   - RBT is split into RBT and RBT_Mapping.
  1507   - AssocList is split and renamed into AList and AList_Mapping.
  1508   - DList is split into DList_Impl, DList, and DList_Cset.
  1509   - Cset is split into Cset and List_Cset.
  1510 
  1511 * Theory Library/Nat_Infinity has been renamed to
  1512 Library/Extended_Nat, with name changes of the following types and
  1513 constants:
  1514 
  1515   type inat   ~> type enat
  1516   Fin         ~> enat
  1517   Infty       ~> infinity (overloaded)
  1518   iSuc        ~> eSuc
  1519   the_Fin     ~> the_enat
  1520 
  1521 Every theorem name containing "inat", "Fin", "Infty", or "iSuc" has
  1522 been renamed accordingly. INCOMPATIBILITY.
  1523 
  1524 * Session Multivariate_Analysis: The euclidean_space type class now
  1525 fixes a constant "Basis :: 'a set" consisting of the standard
  1526 orthonormal basis for the type. Users now have the option of
  1527 quantifying over this set instead of using the "basis" function, e.g.
  1528 "ALL x:Basis. P x" vs "ALL i<DIM('a). P (basis i)".
  1529 
  1530 * Session Multivariate_Analysis: Type "('a, 'b) cart" has been renamed
  1531 to "('a, 'b) vec" (the syntax "'a ^ 'b" remains unaffected). Constants
  1532 "Cart_nth" and "Cart_lambda" have been respectively renamed to
  1533 "vec_nth" and "vec_lambda"; theorems mentioning those names have
  1534 changed to match. Definition theorems for overloaded constants now use
  1535 the standard "foo_vec_def" naming scheme. A few other theorems have
  1536 been renamed as follows (INCOMPATIBILITY):
  1537 
  1538   Cart_eq          ~> vec_eq_iff
  1539   dist_nth_le_cart ~> dist_vec_nth_le
  1540   tendsto_vector   ~> vec_tendstoI
  1541   Cauchy_vector    ~> vec_CauchyI
  1542 
  1543 * Session Multivariate_Analysis: Several duplicate theorems have been
  1544 removed, and other theorems have been renamed or replaced with more
  1545 general versions. INCOMPATIBILITY.
  1546 
  1547   finite_choice ~> finite_set_choice
  1548   eventually_conjI ~> eventually_conj
  1549   eventually_and ~> eventually_conj_iff
  1550   eventually_false ~> eventually_False
  1551   setsum_norm ~> norm_setsum
  1552   Lim_sequentially ~> LIMSEQ_def
  1553   Lim_ident_at ~> LIM_ident
  1554   Lim_const ~> tendsto_const
  1555   Lim_cmul ~> tendsto_scaleR [OF tendsto_const]
  1556   Lim_neg ~> tendsto_minus
  1557   Lim_add ~> tendsto_add
  1558   Lim_sub ~> tendsto_diff
  1559   Lim_mul ~> tendsto_scaleR
  1560   Lim_vmul ~> tendsto_scaleR [OF _ tendsto_const]
  1561   Lim_null_norm ~> tendsto_norm_zero_iff [symmetric]
  1562   Lim_linear ~> bounded_linear.tendsto
  1563   Lim_component ~> tendsto_euclidean_component
  1564   Lim_component_cart ~> tendsto_vec_nth
  1565   Lim_inner ~> tendsto_inner [OF tendsto_const]
  1566   dot_lsum ~> inner_setsum_left
  1567   dot_rsum ~> inner_setsum_right
  1568   continuous_cmul ~> continuous_scaleR [OF continuous_const]
  1569   continuous_neg ~> continuous_minus
  1570   continuous_sub ~> continuous_diff
  1571   continuous_vmul ~> continuous_scaleR [OF _ continuous_const]
  1572   continuous_mul ~> continuous_scaleR
  1573   continuous_inv ~> continuous_inverse
  1574   continuous_at_within_inv ~> continuous_at_within_inverse
  1575   continuous_at_inv ~> continuous_at_inverse
  1576   continuous_at_norm ~> continuous_norm [OF continuous_at_id]
  1577   continuous_at_infnorm ~> continuous_infnorm [OF continuous_at_id]
  1578   continuous_at_component ~> continuous_component [OF continuous_at_id]
  1579   continuous_on_neg ~> continuous_on_minus
  1580   continuous_on_sub ~> continuous_on_diff
  1581   continuous_on_cmul ~> continuous_on_scaleR [OF continuous_on_const]
  1582   continuous_on_vmul ~> continuous_on_scaleR [OF _ continuous_on_const]
  1583   continuous_on_mul ~> continuous_on_scaleR
  1584   continuous_on_mul_real ~> continuous_on_mult
  1585   continuous_on_inner ~> continuous_on_inner [OF continuous_on_const]
  1586   continuous_on_norm ~> continuous_on_norm [OF continuous_on_id]
  1587   continuous_on_inverse ~> continuous_on_inv
  1588   uniformly_continuous_on_neg ~> uniformly_continuous_on_minus
  1589   uniformly_continuous_on_sub ~> uniformly_continuous_on_diff
  1590   subset_interior ~> interior_mono
  1591   subset_closure ~> closure_mono
  1592   closure_univ ~> closure_UNIV
  1593   real_arch_lt ~> reals_Archimedean2
  1594   real_arch ~> reals_Archimedean3
  1595   real_abs_norm ~> abs_norm_cancel
  1596   real_abs_sub_norm ~> norm_triangle_ineq3
  1597   norm_cauchy_schwarz_abs ~> Cauchy_Schwarz_ineq2
  1598 
  1599 * Session HOL-Probability:
  1600   - Caratheodory's extension lemma is now proved for ring_of_sets.
  1601   - Infinite products of probability measures are now available.
  1602   - Sigma closure is independent, if the generator is independent
  1603   - Use extended reals instead of positive extended
  1604     reals. INCOMPATIBILITY.
  1605 
  1606 * Session HOLCF: Discontinued legacy theorem names, INCOMPATIBILITY.
  1607 
  1608   expand_fun_below ~> fun_below_iff
  1609   below_fun_ext ~> fun_belowI
  1610   expand_cfun_eq ~> cfun_eq_iff
  1611   ext_cfun ~> cfun_eqI
  1612   expand_cfun_below ~> cfun_below_iff
  1613   below_cfun_ext ~> cfun_belowI
  1614   monofun_fun_fun ~> fun_belowD
  1615   monofun_fun_arg ~> monofunE
  1616   monofun_lub_fun ~> adm_monofun [THEN admD]
  1617   cont_lub_fun ~> adm_cont [THEN admD]
  1618   cont2cont_Rep_CFun ~> cont2cont_APP
  1619   cont_Rep_CFun_app ~> cont_APP_app
  1620   cont_Rep_CFun_app_app ~> cont_APP_app_app
  1621   cont_cfun_fun ~> cont_Rep_cfun1 [THEN contE]
  1622   cont_cfun_arg ~> cont_Rep_cfun2 [THEN contE]
  1623   contlub_cfun ~> lub_APP [symmetric]
  1624   contlub_LAM ~> lub_LAM [symmetric]
  1625   thelubI ~> lub_eqI
  1626   UU_I ~> bottomI
  1627   lift_distinct1 ~> lift.distinct(1)
  1628   lift_distinct2 ~> lift.distinct(2)
  1629   Def_not_UU ~> lift.distinct(2)
  1630   Def_inject ~> lift.inject
  1631   below_UU_iff ~> below_bottom_iff
  1632   eq_UU_iff ~> eq_bottom_iff
  1633 
  1634 
  1635 *** Document preparation ***
  1636 
  1637 * Antiquotation @{rail} layouts railroad syntax diagrams, see also
  1638 isar-ref manual, both for description and actual application of the
  1639 same.
  1640 
  1641 * Antiquotation @{value} evaluates the given term and presents its
  1642 result.
  1643 
  1644 * Antiquotations: term style "isub" provides ad-hoc conversion of
  1645 variables x1, y23 into subscripted form x\<^isub>1,
  1646 y\<^isub>2\<^isub>3.
  1647 
  1648 * Predefined LaTeX macros for Isabelle symbols \<bind> and \<then>
  1649 (e.g. see ~~/src/HOL/Library/Monad_Syntax.thy).
  1650 
  1651 * Localized \isabellestyle switch can be used within blocks or groups
  1652 like this:
  1653 
  1654   \isabellestyle{it}  %preferred default
  1655   {\isabellestylett @{text "typewriter stuff"}}
  1656 
  1657 * Discontinued special treatment of hard tabulators.  Implicit
  1658 tab-width is now defined as 1.  Potential INCOMPATIBILITY for visual
  1659 layouts.
  1660 
  1661 
  1662 *** ML ***
  1663 
  1664 * The inner syntax of sort/type/term/prop supports inlined YXML
  1665 representations within quoted string tokens.  By encoding logical
  1666 entities via Term_XML (in ML or Scala) concrete syntax can be
  1667 bypassed, which is particularly useful for producing bits of text
  1668 under external program control.
  1669 
  1670 * Antiquotations for ML and document preparation are managed as theory
  1671 data, which requires explicit setup.
  1672 
  1673 * Isabelle_Process.is_active allows tools to check if the official
  1674 process wrapper is running (Isabelle/Scala/jEdit) or the old TTY loop
  1675 (better known as Proof General).
  1676 
  1677 * Structure Proof_Context follows standard naming scheme.  Old
  1678 ProofContext is still available for some time as legacy alias.
  1679 
  1680 * Structure Timing provides various operations for timing; supersedes
  1681 former start_timing/end_timing etc.
  1682 
  1683 * Path.print is the official way to show file-system paths to users
  1684 (including quotes etc.).
  1685 
  1686 * Inner syntax: identifiers in parse trees of generic categories
  1687 "logic", "aprop", "idt" etc. carry position information (disguised as
  1688 type constraints).  Occasional INCOMPATIBILITY with non-compliant
  1689 translations that choke on unexpected type constraints.  Positions can
  1690 be stripped in ML translations via Syntax.strip_positions /
  1691 Syntax.strip_positions_ast, or via the syntax constant
  1692 "_strip_positions" within parse trees.  As last resort, positions can
  1693 be disabled via the configuration option Syntax.positions, which is
  1694 called "syntax_positions" in Isar attribute syntax.
  1695 
  1696 * Discontinued special status of various ML structures that contribute
  1697 to structure Syntax (Ast, Lexicon, Mixfix, Parser, Printer etc.): less
  1698 pervasive content, no inclusion in structure Syntax.  INCOMPATIBILITY,
  1699 refer directly to Ast.Constant, Lexicon.is_identifier,
  1700 Syntax_Trans.mk_binder_tr etc.
  1701 
  1702 * Typed print translation: discontinued show_sorts argument, which is
  1703 already available via context of "advanced" translation.
  1704 
  1705 * Refined PARALLEL_GOALS tactical: degrades gracefully for schematic
  1706 goal states; body tactic needs to address all subgoals uniformly.
  1707 
  1708 * Slightly more special eq_list/eq_set, with shortcut involving
  1709 pointer equality (assumes that eq relation is reflexive).
  1710 
  1711 * Classical tactics use proper Proof.context instead of historic types
  1712 claset/clasimpset.  Old-style declarations like addIs, addEs, addDs
  1713 operate directly on Proof.context.  Raw type claset retains its use as
  1714 snapshot of the classical context, which can be recovered via
  1715 (put_claset HOL_cs) etc.  Type clasimpset has been discontinued.
  1716 INCOMPATIBILITY, classical tactics and derived proof methods require
  1717 proper Proof.context.
  1718 
  1719 
  1720 *** System ***
  1721 
  1722 * Discontinued support for Poly/ML 5.2, which was the last version
  1723 without proper multithreading and TimeLimit implementation.
  1724 
  1725 * Discontinued old lib/scripts/polyml-platform, which has been
  1726 obsolete since Isabelle2009-2.
  1727 
  1728 * Various optional external tools are referenced more robustly and
  1729 uniformly by explicit Isabelle settings as follows:
  1730 
  1731   ISABELLE_CSDP   (formerly CSDP_EXE)
  1732   ISABELLE_GHC    (formerly EXEC_GHC or GHC_PATH)
  1733   ISABELLE_OCAML  (formerly EXEC_OCAML)
  1734   ISABELLE_SWIPL  (formerly EXEC_SWIPL)
  1735   ISABELLE_YAP    (formerly EXEC_YAP)
  1736 
  1737 Note that automated detection from the file-system or search path has
  1738 been discontinued.  INCOMPATIBILITY.
  1739 
  1740 * Scala layer provides JVM method invocation service for static
  1741 methods of type (String)String, see Invoke_Scala.method in ML.  For
  1742 example:
  1743 
  1744   Invoke_Scala.method "java.lang.System.getProperty" "java.home"
  1745 
  1746 Together with YXML.string_of_body/parse_body and XML.Encode/Decode
  1747 this allows to pass structured values between ML and Scala.
  1748 
  1749 * The IsabelleText fonts includes some further glyphs to support the
  1750 Prover IDE.  Potential INCOMPATIBILITY: users who happen to have
  1751 installed a local copy (which is normally *not* required) need to
  1752 delete or update it from ~~/lib/fonts/.
  1753 
  1754 
  1755 
  1756 New in Isabelle2011 (January 2011)
  1757 ----------------------------------
  1758 
  1759 *** General ***
  1760 
  1761 * Experimental Prover IDE based on Isabelle/Scala and jEdit (see
  1762 src/Tools/jEdit).  This also serves as IDE for Isabelle/ML, with
  1763 useful tooltips and hyperlinks produced from its static analysis.  The
  1764 bundled component provides an executable Isabelle tool that can be run
  1765 like this:
  1766 
  1767   Isabelle2011/bin/isabelle jedit
  1768 
  1769 * Significantly improved Isabelle/Isar implementation manual.
  1770 
  1771 * System settings: ISABELLE_HOME_USER now includes ISABELLE_IDENTIFIER
  1772 (and thus refers to something like $HOME/.isabelle/Isabelle2011),
  1773 while the default heap location within that directory lacks that extra
  1774 suffix.  This isolates multiple Isabelle installations from each
  1775 other, avoiding problems with old settings in new versions.
  1776 INCOMPATIBILITY, need to copy/upgrade old user settings manually.
  1777 
  1778 * Source files are always encoded as UTF-8, instead of old-fashioned
  1779 ISO-Latin-1.  INCOMPATIBILITY.  Isabelle LaTeX documents might require
  1780 the following package declarations:
  1781 
  1782   \usepackage[utf8]{inputenc}
  1783   \usepackage{textcomp}
  1784 
  1785 * Explicit treatment of UTF-8 sequences as Isabelle symbols, such that
  1786 a Unicode character is treated as a single symbol, not a sequence of
  1787 non-ASCII bytes as before.  Since Isabelle/ML string literals may
  1788 contain symbols without further backslash escapes, Unicode can now be
  1789 used here as well.  Recall that Symbol.explode in ML provides a
  1790 consistent view on symbols, while raw explode (or String.explode)
  1791 merely give a byte-oriented representation.
  1792 
  1793 * Theory loader: source files are primarily located via the master
  1794 directory of each theory node (where the .thy file itself resides).
  1795 The global load path is still partially available as legacy feature.
  1796 Minor INCOMPATIBILITY due to subtle change in file lookup: use
  1797 explicit paths, relatively to the theory.
  1798 
  1799 * Special treatment of ML file names has been discontinued.
  1800 Historically, optional extensions .ML or .sml were added on demand --
  1801 at the cost of clarity of file dependencies.  Recall that Isabelle/ML
  1802 files exclusively use the .ML extension.  Minor INCOMPATIBILTY.
  1803 
  1804 * Various options that affect pretty printing etc. are now properly
  1805 handled within the context via configuration options, instead of
  1806 unsynchronized references or print modes.  There are both ML Config.T
  1807 entities and Isar declaration attributes to access these.
  1808 
  1809   ML (Config.T)                 Isar (attribute)
  1810 
  1811   eta_contract                  eta_contract
  1812   show_brackets                 show_brackets
  1813   show_sorts                    show_sorts
  1814   show_types                    show_types
  1815   show_question_marks           show_question_marks
  1816   show_consts                   show_consts
  1817   show_abbrevs                  show_abbrevs
  1818 
  1819   Syntax.ast_trace              syntax_ast_trace
  1820   Syntax.ast_stat               syntax_ast_stat
  1821   Syntax.ambiguity_level        syntax_ambiguity_level
  1822 
  1823   Goal_Display.goals_limit      goals_limit
  1824   Goal_Display.show_main_goal   show_main_goal
  1825 
  1826   Method.rule_trace             rule_trace
  1827 
  1828   Thy_Output.display            thy_output_display
  1829   Thy_Output.quotes             thy_output_quotes
  1830   Thy_Output.indent             thy_output_indent
  1831   Thy_Output.source             thy_output_source
  1832   Thy_Output.break              thy_output_break
  1833 
  1834 Note that corresponding "..._default" references in ML may only be
  1835 changed globally at the ROOT session setup, but *not* within a theory.
  1836 The option "show_abbrevs" supersedes the former print mode
  1837 "no_abbrevs" with inverted meaning.
  1838 
  1839 * More systematic naming of some configuration options.
  1840 INCOMPATIBILITY.
  1841 
  1842   trace_simp  ~>  simp_trace
  1843   debug_simp  ~>  simp_debug
  1844 
  1845 * Support for real valued configuration options, using simplistic
  1846 floating-point notation that coincides with the inner syntax for
  1847 float_token.
  1848 
  1849 * Support for real valued preferences (with approximative PGIP type):
  1850 front-ends need to accept "pgint" values in float notation.
  1851 INCOMPATIBILITY.
  1852 
  1853 * The IsabelleText font now includes Cyrillic, Hebrew, Arabic from
  1854 DejaVu Sans.
  1855 
  1856 * Discontinued support for Poly/ML 5.0 and 5.1 versions.
  1857 
  1858 
  1859 *** Pure ***
  1860 
  1861 * Command 'type_synonym' (with single argument) replaces somewhat
  1862 outdated 'types', which is still available as legacy feature for some
  1863 time.
  1864 
  1865 * Command 'nonterminal' (with 'and' separated list of arguments)
  1866 replaces somewhat outdated 'nonterminals'.  INCOMPATIBILITY.
  1867 
  1868 * Command 'notepad' replaces former 'example_proof' for
  1869 experimentation in Isar without any result.  INCOMPATIBILITY.
  1870 
  1871 * Locale interpretation commands 'interpret' and 'sublocale' accept
  1872 lists of equations to map definitions in a locale to appropriate
  1873 entities in the context of the interpretation.  The 'interpretation'
  1874 command already provided this functionality.
  1875 
  1876 * Diagnostic command 'print_dependencies' prints the locale instances
  1877 that would be activated if the specified expression was interpreted in
  1878 the current context.  Variant "print_dependencies!" assumes a context
  1879 without interpretations.
  1880 
  1881 * Diagnostic command 'print_interps' prints interpretations in proofs
  1882 in addition to interpretations in theories.
  1883 
  1884 * Discontinued obsolete 'global' and 'local' commands to manipulate
  1885 the theory name space.  Rare INCOMPATIBILITY.  The ML functions
  1886 Sign.root_path and Sign.local_path may be applied directly where this
  1887 feature is still required for historical reasons.
  1888 
  1889 * Discontinued obsolete 'constdefs' command.  INCOMPATIBILITY, use
  1890 'definition' instead.
  1891 
  1892 * The "prems" fact, which refers to the accidental collection of
  1893 foundational premises in the context, is now explicitly marked as
  1894 legacy feature and will be discontinued soon.  Consider using "assms"
  1895 of the head statement or reference facts by explicit names.
  1896 
  1897 * Document antiquotations @{class} and @{type} print classes and type
  1898 constructors.
  1899 
  1900 * Document antiquotation @{file} checks file/directory entries within
  1901 the local file system.
  1902 
  1903 
  1904 *** HOL ***
  1905 
  1906 * Coercive subtyping: functions can be declared as coercions and type
  1907 inference will add them as necessary upon input of a term.  Theory
  1908 Complex_Main declares real :: nat => real and real :: int => real as
  1909 coercions. A coercion function f is declared like this:
  1910 
  1911   declare [[coercion f]]
  1912 
  1913 To lift coercions through type constructors (e.g. from nat => real to
  1914 nat list => real list), map functions can be declared, e.g.
  1915 
  1916   declare [[coercion_map map]]
  1917 
  1918 Currently coercion inference is activated only in theories including
  1919 real numbers, i.e. descendants of Complex_Main.  This is controlled by
  1920 the configuration option "coercion_enabled", e.g. it can be enabled in
  1921 other theories like this:
  1922 
  1923   declare [[coercion_enabled]]
  1924 
  1925 * Command 'partial_function' provides basic support for recursive
  1926 function definitions over complete partial orders.  Concrete instances
  1927 are provided for i) the option type, ii) tail recursion on arbitrary
  1928 types, and iii) the heap monad of Imperative_HOL.  See
  1929 src/HOL/ex/Fundefs.thy and src/HOL/Imperative_HOL/ex/Linked_Lists.thy
  1930 for examples.
  1931 
  1932 * Function package: f.psimps rules are no longer implicitly declared
  1933 as [simp].  INCOMPATIBILITY.
  1934 
  1935 * Datatype package: theorems generated for executable equality (class
  1936 "eq") carry proper names and are treated as default code equations.
  1937 
  1938 * Inductive package: now offers command 'inductive_simps' to
  1939 automatically derive instantiated and simplified equations for
  1940 inductive predicates, similar to 'inductive_cases'.
  1941 
  1942 * Command 'enriched_type' allows to register properties of the
  1943 functorial structure of types.
  1944 
  1945 * Improved infrastructure for term evaluation using code generator
  1946 techniques, in particular static evaluation conversions.
  1947 
  1948 * Code generator: Scala (2.8 or higher) has been added to the target
  1949 languages.
  1950 
  1951 * Code generator: globbing constant expressions "*" and "Theory.*"
  1952 have been replaced by the more idiomatic "_" and "Theory._".
  1953 INCOMPATIBILITY.
  1954 
  1955 * Code generator: export_code without explicit file declaration prints
  1956 to standard output.  INCOMPATIBILITY.
  1957 
  1958 * Code generator: do not print function definitions for case
  1959 combinators any longer.
  1960 
  1961 * Code generator: simplification with rules determined with
  1962 src/Tools/Code/code_simp.ML and method "code_simp".
  1963 
  1964 * Code generator for records: more idiomatic representation of record
  1965 types.  Warning: records are not covered by ancient SML code
  1966 generation any longer.  INCOMPATIBILITY.  In cases of need, a suitable
  1967 rep_datatype declaration helps to succeed then:
  1968 
  1969   record 'a foo = ...
  1970   ...
  1971   rep_datatype foo_ext ...
  1972 
  1973 * Records: logical foundation type for records does not carry a
  1974 '_type' suffix any longer (obsolete due to authentic syntax).
  1975 INCOMPATIBILITY.
  1976 
  1977 * Quickcheck now by default uses exhaustive testing instead of random
  1978 testing.  Random testing can be invoked by "quickcheck [random]",
  1979 exhaustive testing by "quickcheck [exhaustive]".
  1980 
  1981 * Quickcheck instantiates polymorphic types with small finite
  1982 datatypes by default. This enables a simple execution mechanism to
  1983 handle quantifiers and function equality over the finite datatypes.
  1984 
  1985 * Quickcheck random generator has been renamed from "code" to
  1986 "random".  INCOMPATIBILITY.
  1987 
  1988 * Quickcheck now has a configurable time limit which is set to 30
  1989 seconds by default. This can be changed by adding [timeout = n] to the
  1990 quickcheck command. The time limit for Auto Quickcheck is still set
  1991 independently.
  1992 
  1993 * Quickcheck in locales considers interpretations of that locale for
  1994 counter example search.
  1995 
  1996 * Sledgehammer:
  1997   - Added "smt" and "remote_smt" provers based on the "smt" proof
  1998     method. See the Sledgehammer manual for details ("isabelle doc
  1999     sledgehammer").
  2000   - Renamed commands:
  2001     sledgehammer atp_info ~> sledgehammer running_provers
  2002     sledgehammer atp_kill ~> sledgehammer kill_provers
  2003     sledgehammer available_atps ~> sledgehammer available_provers
  2004     INCOMPATIBILITY.
  2005   - Renamed options:
  2006     sledgehammer [atps = ...] ~> sledgehammer [provers = ...]
  2007     sledgehammer [atp = ...] ~> sledgehammer [prover = ...]
  2008     sledgehammer [timeout = 77 s] ~> sledgehammer [timeout = 77]
  2009     (and "ms" and "min" are no longer supported)
  2010     INCOMPATIBILITY.
  2011 
  2012 * Nitpick:
  2013   - Renamed options:
  2014     nitpick [timeout = 77 s] ~> nitpick [timeout = 77]
  2015     nitpick [tac_timeout = 777 ms] ~> nitpick [tac_timeout = 0.777]
  2016     INCOMPATIBILITY.
  2017   - Added support for partial quotient types.
  2018   - Added local versions of the "Nitpick.register_xxx" functions.
  2019   - Added "whack" option.
  2020   - Allow registration of quotient types as codatatypes.
  2021   - Improved "merge_type_vars" option to merge more types.
  2022   - Removed unsound "fast_descrs" option.
  2023   - Added custom symmetry breaking for datatypes, making it possible to reach
  2024     higher cardinalities.
  2025   - Prevent the expansion of too large definitions.
  2026 
  2027 * Proof methods "metis" and "meson" now have configuration options
  2028 "meson_trace", "metis_trace", and "metis_verbose" that can be enabled
  2029 to diagnose these tools. E.g.
  2030 
  2031     using [[metis_trace = true]]
  2032 
  2033 * Auto Solve: Renamed "Auto Solve Direct".  The tool is now available
  2034 manually as command 'solve_direct'.
  2035 
  2036 * The default SMT solver Z3 must be enabled explicitly (due to
  2037 licensing issues) by setting the environment variable
  2038 Z3_NON_COMMERCIAL in etc/settings of the component, for example.  For
  2039 commercial applications, the SMT solver CVC3 is provided as fall-back;
  2040 changing the SMT solver is done via the configuration option
  2041 "smt_solver".
  2042 
  2043 * Remote SMT solvers need to be referred to by the "remote_" prefix,
  2044 i.e. "remote_cvc3" and "remote_z3".
  2045 
  2046 * Added basic SMT support for datatypes, records, and typedefs using
  2047 the oracle mode (no proofs).  Direct support of pairs has been dropped
  2048 in exchange (pass theorems fst_conv snd_conv pair_collapse to the SMT
  2049 support for a similar behavior).  Minor INCOMPATIBILITY.
  2050 
  2051 * Changed SMT configuration options:
  2052   - Renamed:
  2053     z3_proofs ~> smt_oracle (with inverted meaning)
  2054     z3_trace_assms ~> smt_trace_used_facts
  2055     INCOMPATIBILITY.
  2056   - Added:
  2057     smt_verbose
  2058     smt_random_seed
  2059     smt_datatypes
  2060     smt_infer_triggers
  2061     smt_monomorph_limit
  2062     cvc3_options
  2063     remote_cvc3_options
  2064     remote_z3_options
  2065     yices_options
  2066 
  2067 * Boogie output files (.b2i files) need to be declared in the theory
  2068 header.
  2069 
  2070 * Simplification procedure "list_to_set_comprehension" rewrites list
  2071 comprehensions applied to List.set to set comprehensions.  Occasional
  2072 INCOMPATIBILITY, may be deactivated like this:
  2073 
  2074   declare [[simproc del: list_to_set_comprehension]]
  2075 
  2076 * Removed old version of primrec package.  INCOMPATIBILITY.
  2077 
  2078 * Removed simplifier congruence rule of "prod_case", as has for long
  2079 been the case with "split".  INCOMPATIBILITY.
  2080 
  2081 * String.literal is a type, but not a datatype.  INCOMPATIBILITY.
  2082 
  2083 * Removed [split_format ... and ... and ...] version of
  2084 [split_format].  Potential INCOMPATIBILITY.
  2085 
  2086 * Predicate "sorted" now defined inductively, with nice induction
  2087 rules.  INCOMPATIBILITY: former sorted.simps now named sorted_simps.
  2088 
  2089 * Constant "contents" renamed to "the_elem", to free the generic name
  2090 contents for other uses.  INCOMPATIBILITY.
  2091 
  2092 * Renamed class eq and constant eq (for code generation) to class
  2093 equal and constant equal, plus renaming of related facts and various
  2094 tuning.  INCOMPATIBILITY.
  2095 
  2096 * Dropped type classes mult_mono and mult_mono1.  INCOMPATIBILITY.
  2097 
  2098 * Removed output syntax "'a ~=> 'b" for "'a => 'b option".
  2099 INCOMPATIBILITY.
  2100 
  2101 * Renamed theory Fset to Cset, type Fset.fset to Cset.set, in order to
  2102 avoid confusion with finite sets.  INCOMPATIBILITY.
  2103 
  2104 * Abandoned locales equiv, congruent and congruent2 for equivalence
  2105 relations.  INCOMPATIBILITY: use equivI rather than equiv_intro (same
  2106 for congruent(2)).
  2107 
  2108 * Some previously unqualified names have been qualified:
  2109 
  2110   types
  2111     bool ~> HOL.bool
  2112     nat ~> Nat.nat
  2113 
  2114   constants
  2115     Trueprop ~> HOL.Trueprop
  2116     True ~> HOL.True
  2117     False ~> HOL.False
  2118     op & ~> HOL.conj
  2119     op | ~> HOL.disj
  2120     op --> ~> HOL.implies
  2121     op = ~> HOL.eq
  2122     Not ~> HOL.Not
  2123     The ~> HOL.The
  2124     All ~> HOL.All
  2125     Ex ~> HOL.Ex
  2126     Ex1 ~> HOL.Ex1
  2127     Let ~> HOL.Let
  2128     If ~> HOL.If
  2129     Ball ~> Set.Ball
  2130     Bex ~> Set.Bex
  2131     Suc ~> Nat.Suc
  2132     Pair ~> Product_Type.Pair
  2133     fst ~> Product_Type.fst
  2134     snd ~> Product_Type.snd
  2135     curry ~> Product_Type.curry
  2136     op : ~> Set.member
  2137     Collect ~> Set.Collect
  2138 
  2139 INCOMPATIBILITY.
  2140 
  2141 * More canonical naming convention for some fundamental definitions:
  2142 
  2143     bot_bool_eq ~> bot_bool_def
  2144     top_bool_eq ~> top_bool_def
  2145     inf_bool_eq ~> inf_bool_def
  2146     sup_bool_eq ~> sup_bool_def
  2147     bot_fun_eq  ~> bot_fun_def
  2148     top_fun_eq  ~> top_fun_def
  2149     inf_fun_eq  ~> inf_fun_def
  2150     sup_fun_eq  ~> sup_fun_def
  2151 
  2152 INCOMPATIBILITY.
  2153 
  2154 * More stylized fact names:
  2155 
  2156   expand_fun_eq ~> fun_eq_iff
  2157   expand_set_eq ~> set_eq_iff
  2158   set_ext       ~> set_eqI
  2159   nat_number    ~> eval_nat_numeral
  2160 
  2161 INCOMPATIBILITY.
  2162 
  2163 * Refactoring of code-generation specific operations in theory List:
  2164 
  2165   constants
  2166     null ~> List.null
  2167 
  2168   facts
  2169     mem_iff ~> member_def
  2170     null_empty ~> null_def
  2171 
  2172 INCOMPATIBILITY.  Note that these were not supposed to be used
  2173 regularly unless for striking reasons; their main purpose was code
  2174 generation.
  2175 
  2176 Various operations from the Haskell prelude are used for generating
  2177 Haskell code.
  2178 
  2179 * Term "bij f" is now an abbreviation of "bij_betw f UNIV UNIV".  Term
  2180 "surj f" is now an abbreviation of "range f = UNIV".  The theorems
  2181 bij_def and surj_def are unchanged.  INCOMPATIBILITY.
  2182 
  2183 * Abolished some non-alphabetic type names: "prod" and "sum" replace
  2184 "*" and "+" respectively.  INCOMPATIBILITY.
  2185 
  2186 * Name "Plus" of disjoint sum operator "<+>" is now hidden.  Write
  2187 "Sum_Type.Plus" instead.
  2188 
  2189 * Constant "split" has been merged with constant "prod_case"; names of
  2190 ML functions, facts etc. involving split have been retained so far,
  2191 though.  INCOMPATIBILITY.
  2192 
  2193 * Dropped old infix syntax "_ mem _" for List.member; use "_ : set _"
  2194 instead.  INCOMPATIBILITY.
  2195 
  2196 * Removed lemma "Option.is_none_none" which duplicates "is_none_def".
  2197 INCOMPATIBILITY.
  2198 
  2199 * Former theory Library/Enum is now part of the HOL-Main image.
  2200 INCOMPATIBILITY: all constants of the Enum theory now have to be
  2201 referred to by its qualified name.
  2202 
  2203   enum    ~>  Enum.enum
  2204   nlists  ~>  Enum.nlists
  2205   product ~>  Enum.product
  2206 
  2207 * Theory Library/Monad_Syntax provides do-syntax for monad types.
  2208 Syntax in Library/State_Monad has been changed to avoid ambiguities.
  2209 INCOMPATIBILITY.
  2210 
  2211 * Theory Library/SetsAndFunctions has been split into
  2212 Library/Function_Algebras and Library/Set_Algebras; canonical names
  2213 for instance definitions for functions; various improvements.
  2214 INCOMPATIBILITY.
  2215 
  2216 * Theory Library/Multiset provides stable quicksort implementation of
  2217 sort_key.
  2218 
  2219 * Theory Library/Multiset: renamed empty_idemp ~> empty_neutral.
  2220 INCOMPATIBILITY.
  2221 
  2222 * Session Multivariate_Analysis: introduced a type class for euclidean
  2223 space.  Most theorems are now stated in terms of euclidean spaces
  2224 instead of finite cartesian products.
  2225 
  2226   types
  2227     real ^ 'n ~>  'a::real_vector
  2228               ~>  'a::euclidean_space
  2229               ~>  'a::ordered_euclidean_space
  2230         (depends on your needs)
  2231 
  2232   constants
  2233      _ $ _        ~> _ $$ _
  2234      \<chi> x. _  ~> \<chi>\<chi> x. _
  2235      CARD('n)     ~> DIM('a)
  2236 
  2237 Also note that the indices are now natural numbers and not from some
  2238 finite type. Finite cartesian products of euclidean spaces, products
  2239 of euclidean spaces the real and complex numbers are instantiated to
  2240 be euclidean_spaces.  INCOMPATIBILITY.
  2241 
  2242 * Session Probability: introduced pextreal as positive extended real
  2243 numbers.  Use pextreal as value for measures.  Introduce the
  2244 Radon-Nikodym derivative, product spaces and Fubini's theorem for
  2245 arbitrary sigma finite measures.  Introduces Lebesgue measure based on
  2246 the integral in Multivariate Analysis.  INCOMPATIBILITY.
  2247 
  2248 * Session Imperative_HOL: revamped, corrected dozens of inadequacies.
  2249 INCOMPATIBILITY.
  2250 
  2251 * Session SPARK (with image HOL-SPARK) provides commands to load and
  2252 prove verification conditions generated by the SPARK Ada program
  2253 verifier.  See also src/HOL/SPARK and src/HOL/SPARK/Examples.
  2254 
  2255 
  2256 *** HOL-Algebra ***
  2257 
  2258 * Theorems for additive ring operations (locale abelian_monoid and
  2259 descendants) are generated by interpretation from their multiplicative
  2260 counterparts.  Names (in particular theorem names) have the mandatory
  2261 qualifier 'add'.  Previous theorem names are redeclared for
  2262 compatibility.
  2263 
  2264 * Structure "int_ring" is now an abbreviation (previously a
  2265 definition).  This fits more natural with advanced interpretations.
  2266 
  2267 
  2268 *** HOLCF ***
  2269 
  2270 * The domain package now runs in definitional mode by default: The
  2271 former command 'new_domain' is now called 'domain'.  To use the domain
  2272 package in its original axiomatic mode, use 'domain (unsafe)'.
  2273 INCOMPATIBILITY.
  2274 
  2275 * The new class "domain" is now the default sort.  Class "predomain"
  2276 is an unpointed version of "domain". Theories can be updated by
  2277 replacing sort annotations as shown below.  INCOMPATIBILITY.
  2278 
  2279   'a::type ~> 'a::countable
  2280   'a::cpo  ~> 'a::predomain
  2281   'a::pcpo ~> 'a::domain
  2282 
  2283 * The old type class "rep" has been superseded by class "domain".
  2284 Accordingly, users of the definitional package must remove any
  2285 "default_sort rep" declarations.  INCOMPATIBILITY.
  2286 
  2287 * The domain package (definitional mode) now supports unpointed
  2288 predomain argument types, as long as they are marked 'lazy'. (Strict
  2289 arguments must be in class "domain".) For example, the following
  2290 domain definition now works:
  2291 
  2292   domain natlist = nil | cons (lazy "nat discr") (lazy "natlist")
  2293 
  2294 * Theory HOLCF/Library/HOL_Cpo provides cpo and predomain class
  2295 instances for types from main HOL: bool, nat, int, char, 'a + 'b,
  2296 'a option, and 'a list.  Additionally, it configures fixrec and the
  2297 domain package to work with these types.  For example:
  2298 
  2299   fixrec isInl :: "('a + 'b) u -> tr"
  2300     where "isInl$(up$(Inl x)) = TT" | "isInl$(up$(Inr y)) = FF"
  2301 
  2302   domain V = VFun (lazy "V -> V") | VCon (lazy "nat") (lazy "V list")
  2303 
  2304 * The "(permissive)" option of fixrec has been replaced with a
  2305 per-equation "(unchecked)" option. See
  2306 src/HOL/HOLCF/Tutorial/Fixrec_ex.thy for examples. INCOMPATIBILITY.
  2307 
  2308 * The "bifinite" class no longer fixes a constant "approx"; the class
  2309 now just asserts that such a function exists.  INCOMPATIBILITY.
  2310 
  2311 * Former type "alg_defl" has been renamed to "defl".  HOLCF no longer
  2312 defines an embedding of type 'a defl into udom by default; instances
  2313 of "bifinite" and "domain" classes are available in
  2314 src/HOL/HOLCF/Library/Defl_Bifinite.thy.
  2315 
  2316 * The syntax "REP('a)" has been replaced with "DEFL('a)".
  2317 
  2318 * The predicate "directed" has been removed.  INCOMPATIBILITY.
  2319 
  2320 * The type class "finite_po" has been removed.  INCOMPATIBILITY.
  2321 
  2322 * The function "cprod_map" has been renamed to "prod_map".
  2323 INCOMPATIBILITY.
  2324 
  2325 * The monadic bind operator on each powerdomain has new binder syntax
  2326 similar to sets, e.g. "\<Union>\<sharp>x\<in>xs. t" represents
  2327 "upper_bind\<cdot>xs\<cdot>(\<Lambda> x. t)".
  2328 
  2329 * The infix syntax for binary union on each powerdomain has changed
  2330 from e.g. "+\<sharp>" to "\<union>\<sharp>", for consistency with set
  2331 syntax.  INCOMPATIBILITY.
  2332 
  2333 * The constant "UU" has been renamed to "bottom".  The syntax "UU" is
  2334 still supported as an input translation.
  2335 
  2336 * Renamed some theorems (the original names are also still available).
  2337 
  2338   expand_fun_below   ~> fun_below_iff
  2339   below_fun_ext      ~> fun_belowI
  2340   expand_cfun_eq     ~> cfun_eq_iff
  2341   ext_cfun           ~> cfun_eqI
  2342   expand_cfun_below  ~> cfun_below_iff
  2343   below_cfun_ext     ~> cfun_belowI
  2344   cont2cont_Rep_CFun ~> cont2cont_APP
  2345 
  2346 * The Abs and Rep functions for various types have changed names.
  2347 Related theorem names have also changed to match. INCOMPATIBILITY.
  2348 
  2349   Rep_CFun  ~> Rep_cfun
  2350   Abs_CFun  ~> Abs_cfun
  2351   Rep_Sprod ~> Rep_sprod
  2352   Abs_Sprod ~> Abs_sprod
  2353   Rep_Ssum  ~> Rep_ssum
  2354   Abs_Ssum  ~> Abs_ssum
  2355 
  2356 * Lemmas with names of the form *_defined_iff or *_strict_iff have
  2357 been renamed to *_bottom_iff.  INCOMPATIBILITY.
  2358 
  2359 * Various changes to bisimulation/coinduction with domain package:
  2360 
  2361   - Definitions of "bisim" constants no longer mention definedness.
  2362   - With mutual recursion, "bisim" predicate is now curried.
  2363   - With mutual recursion, each type gets a separate coind theorem.
  2364   - Variable names in bisim_def and coinduct rules have changed.
  2365 
  2366 INCOMPATIBILITY.
  2367 
  2368 * Case combinators generated by the domain package for type "foo" are
  2369 now named "foo_case" instead of "foo_when".  INCOMPATIBILITY.
  2370 
  2371 * Several theorems have been renamed to more accurately reflect the
  2372 names of constants and types involved.  INCOMPATIBILITY.
  2373 
  2374   thelub_const    ~> lub_const
  2375   lub_const       ~> is_lub_const
  2376   thelubI         ~> lub_eqI
  2377   is_lub_lub      ~> is_lubD2
  2378   lubI            ~> is_lub_lub
  2379   unique_lub      ~> is_lub_unique
  2380   is_ub_lub       ~> is_lub_rangeD1
  2381   lub_bin_chain   ~> is_lub_bin_chain
  2382   lub_fun         ~> is_lub_fun
  2383   thelub_fun      ~> lub_fun
  2384   thelub_cfun     ~> lub_cfun
  2385   thelub_Pair     ~> lub_Pair
  2386   lub_cprod       ~> is_lub_prod
  2387   thelub_cprod    ~> lub_prod
  2388   minimal_cprod   ~> minimal_prod
  2389   inst_cprod_pcpo ~> inst_prod_pcpo
  2390   UU_I            ~> bottomI
  2391   compact_UU      ~> compact_bottom
  2392   deflation_UU    ~> deflation_bottom
  2393   finite_deflation_UU ~> finite_deflation_bottom
  2394 
  2395 * Many legacy theorem names have been discontinued.  INCOMPATIBILITY.
  2396 
  2397   sq_ord_less_eq_trans ~> below_eq_trans
  2398   sq_ord_eq_less_trans ~> eq_below_trans
  2399   refl_less            ~> below_refl
  2400   trans_less           ~> below_trans
  2401   antisym_less         ~> below_antisym
  2402   antisym_less_inverse ~> po_eq_conv [THEN iffD1]
  2403   box_less             ~> box_below
  2404   rev_trans_less       ~> rev_below_trans
  2405   not_less2not_eq      ~> not_below2not_eq
  2406   less_UU_iff          ~> below_UU_iff
  2407   flat_less_iff        ~> flat_below_iff
  2408   adm_less             ~> adm_below
  2409   adm_not_less         ~> adm_not_below
  2410   adm_compact_not_less ~> adm_compact_not_below
  2411   less_fun_def         ~> below_fun_def
  2412   expand_fun_less      ~> fun_below_iff
  2413   less_fun_ext         ~> fun_belowI
  2414   less_discr_def       ~> below_discr_def
  2415   discr_less_eq        ~> discr_below_eq
  2416   less_unit_def        ~> below_unit_def
  2417   less_cprod_def       ~> below_prod_def
  2418   prod_lessI           ~> prod_belowI
  2419   Pair_less_iff        ~> Pair_below_iff
  2420   fst_less_iff         ~> fst_below_iff
  2421   snd_less_iff         ~> snd_below_iff
  2422   expand_cfun_less     ~> cfun_below_iff
  2423   less_cfun_ext        ~> cfun_belowI
  2424   injection_less       ~> injection_below
  2425   less_up_def          ~> below_up_def
  2426   not_Iup_less         ~> not_Iup_below
  2427   Iup_less             ~> Iup_below
  2428   up_less              ~> up_below
  2429   Def_inject_less_eq   ~> Def_below_Def
  2430   Def_less_is_eq       ~> Def_below_iff
  2431   spair_less_iff       ~> spair_below_iff
  2432   less_sprod           ~> below_sprod
  2433   spair_less           ~> spair_below
  2434   sfst_less_iff        ~> sfst_below_iff
  2435   ssnd_less_iff        ~> ssnd_below_iff
  2436   fix_least_less       ~> fix_least_below
  2437   dist_less_one        ~> dist_below_one
  2438   less_ONE             ~> below_ONE
  2439   ONE_less_iff         ~> ONE_below_iff
  2440   less_sinlD           ~> below_sinlD
  2441   less_sinrD           ~> below_sinrD
  2442 
  2443 
  2444 *** FOL and ZF ***
  2445 
  2446 * All constant names are now qualified internally and use proper
  2447 identifiers, e.g. "IFOL.eq" instead of "op =".  INCOMPATIBILITY.
  2448 
  2449 
  2450 *** ML ***
  2451 
  2452 * Antiquotation @{assert} inlines a function bool -> unit that raises
  2453 Fail if the argument is false.  Due to inlining the source position of
  2454 failed assertions is included in the error output.
  2455 
  2456 * Discontinued antiquotation @{theory_ref}, which is obsolete since ML
  2457 text is in practice always evaluated with a stable theory checkpoint.
  2458 Minor INCOMPATIBILITY, use (Theory.check_thy @{theory}) instead.
  2459 
  2460 * Antiquotation @{theory A} refers to theory A from the ancestry of
  2461 the current context, not any accidental theory loader state as before.
  2462 Potential INCOMPATIBILITY, subtle change in semantics.
  2463 
  2464 * Syntax.pretty_priority (default 0) configures the required priority
  2465 of pretty-printed output and thus affects insertion of parentheses.
  2466 
  2467 * Syntax.default_root (default "any") configures the inner syntax
  2468 category (nonterminal symbol) for parsing of terms.
  2469 
  2470 * Former exception Library.UnequalLengths now coincides with
  2471 ListPair.UnequalLengths.
  2472 
  2473 * Renamed structure MetaSimplifier to Raw_Simplifier.  Note that the
  2474 main functionality is provided by structure Simplifier.
  2475 
  2476 * Renamed raw "explode" function to "raw_explode" to emphasize its
  2477 meaning.  Note that internally to Isabelle, Symbol.explode is used in
  2478 almost all situations.
  2479 
  2480 * Discontinued obsolete function sys_error and exception SYS_ERROR.
  2481 See implementation manual for further details on exceptions in
  2482 Isabelle/ML.
  2483 
  2484 * Renamed setmp_noncritical to Unsynchronized.setmp to emphasize its
  2485 meaning.
  2486 
  2487 * Renamed structure PureThy to Pure_Thy and moved most of its
  2488 operations to structure Global_Theory, to emphasize that this is
  2489 rarely-used global-only stuff.
  2490 
  2491 * Discontinued Output.debug.  Minor INCOMPATIBILITY, use plain writeln
  2492 instead (or tracing for high-volume output).
  2493 
  2494 * Configuration option show_question_marks only affects regular pretty
  2495 printing of types and terms, not raw Term.string_of_vname.
  2496 
  2497 * ML_Context.thm and ML_Context.thms are no longer pervasive.  Rare
  2498 INCOMPATIBILITY, superseded by static antiquotations @{thm} and
  2499 @{thms} for most purposes.
  2500 
  2501 * ML structure Unsynchronized is never opened, not even in Isar
  2502 interaction mode as before.  Old Unsynchronized.set etc. have been
  2503 discontinued -- use plain := instead.  This should be *rare* anyway,
  2504 since modern tools always work via official context data, notably
  2505 configuration options.
  2506 
  2507 * Parallel and asynchronous execution requires special care concerning
  2508 interrupts.  Structure Exn provides some convenience functions that
  2509 avoid working directly with raw Interrupt.  User code must not absorb
  2510 interrupts -- intermediate handling (for cleanup etc.) needs to be
  2511 followed by re-raising of the original exception.  Another common
  2512 source of mistakes are "handle _" patterns, which make the meaning of
  2513 the program subject to physical effects of the environment.
  2514 
  2515 
  2516 
  2517 New in Isabelle2009-2 (June 2010)
  2518 ---------------------------------
  2519 
  2520 *** General ***
  2521 
  2522 * Authentic syntax for *all* logical entities (type classes, type
  2523 constructors, term constants): provides simple and robust
  2524 correspondence between formal entities and concrete syntax.  Within
  2525 the parse tree / AST representations, "constants" are decorated by
  2526 their category (class, type, const) and spelled out explicitly with
  2527 their full internal name.
  2528 
  2529 Substantial INCOMPATIBILITY concerning low-level syntax declarations
  2530 and translations (translation rules and translation functions in ML).
  2531 Some hints on upgrading:
  2532 
  2533   - Many existing uses of 'syntax' and 'translations' can be replaced
  2534     by more modern 'type_notation', 'notation' and 'abbreviation',
  2535     which are independent of this issue.
  2536 
  2537   - 'translations' require markup within the AST; the term syntax
  2538     provides the following special forms:
  2539 
  2540       CONST c   -- produces syntax version of constant c from context
  2541       XCONST c  -- literally c, checked as constant from context
  2542       c         -- literally c, if declared by 'syntax'
  2543 
  2544     Plain identifiers are treated as AST variables -- occasionally the
  2545     system indicates accidental variables via the error "rhs contains
  2546     extra variables".
  2547 
  2548     Type classes and type constructors are marked according to their
  2549     concrete syntax.  Some old translations rules need to be written
  2550     for the "type" category, using type constructor application
  2551     instead of pseudo-term application of the default category
  2552     "logic".
  2553 
  2554   - 'parse_translation' etc. in ML may use the following
  2555     antiquotations:
  2556 
  2557       @{class_syntax c}   -- type class c within parse tree / AST
  2558       @{term_syntax c}    -- type constructor c within parse tree / AST
  2559       @{const_syntax c}   -- ML version of "CONST c" above
  2560       @{syntax_const c}   -- literally c (checked wrt. 'syntax' declarations)
  2561 
  2562   - Literal types within 'typed_print_translations', i.e. those *not*
  2563     represented as pseudo-terms are represented verbatim.  Use @{class
  2564     c} or @{type_name c} here instead of the above syntax
  2565     antiquotations.
  2566 
  2567 Note that old non-authentic syntax was based on unqualified base
  2568 names, so all of the above "constant" names would coincide.  Recall
  2569 that 'print_syntax' and ML_command "set Syntax.trace_ast" help to
  2570 diagnose syntax problems.
  2571 
  2572 * Type constructors admit general mixfix syntax, not just infix.
  2573 
  2574 * Concrete syntax may be attached to local entities without a proof
  2575 body, too.  This works via regular mixfix annotations for 'fix',
  2576 'def', 'obtain' etc. or via the explicit 'write' command, which is
  2577 similar to the 'notation' command in theory specifications.
  2578 
  2579 * Discontinued unnamed infix syntax (legacy feature for many years) --
  2580 need to specify constant name and syntax separately.  Internal ML
  2581 datatype constructors have been renamed from InfixName to Infix etc.
  2582 Minor INCOMPATIBILITY.
  2583 
  2584 * Schematic theorem statements need to be explicitly markup as such,
  2585 via commands 'schematic_lemma', 'schematic_theorem',
  2586 'schematic_corollary'.  Thus the relevance of the proof is made
  2587 syntactically clear, which impacts performance in a parallel or
  2588 asynchronous interactive environment.  Minor INCOMPATIBILITY.
  2589 
  2590 * Use of cumulative prems via "!" in some proof methods has been
  2591 discontinued (old legacy feature).
  2592 
  2593 * References 'trace_simp' and 'debug_simp' have been replaced by
  2594 configuration options stored in the context. Enabling tracing (the
  2595 case of debugging is similar) in proofs works via
  2596 
  2597   using [[trace_simp = true]]
  2598 
  2599 Tracing is then active for all invocations of the simplifier in
  2600 subsequent goal refinement steps. Tracing may also still be enabled or
  2601 disabled via the ProofGeneral settings menu.
  2602 
  2603 * Separate commands 'hide_class', 'hide_type', 'hide_const',
  2604 'hide_fact' replace the former 'hide' KIND command.  Minor
  2605 INCOMPATIBILITY.
  2606 
  2607 * Improved parallelism of proof term normalization: usedir -p2 -q0 is
  2608 more efficient than combinations with -q1 or -q2.
  2609 
  2610 
  2611 *** Pure ***
  2612 
  2613 * Proofterms record type-class reasoning explicitly, using the
  2614 "unconstrain" operation internally.  This eliminates all sort
  2615 constraints from a theorem and proof, introducing explicit
  2616 OFCLASS-premises.  On the proof term level, this operation is
  2617 automatically applied at theorem boundaries, such that closed proofs
  2618 are always free of sort constraints.  INCOMPATIBILITY for tools that
  2619 inspect proof terms.
  2620 
  2621 * Local theory specifications may depend on extra type variables that
  2622 are not present in the result type -- arguments TYPE('a) :: 'a itself
  2623 are added internally.  For example:
  2624 
  2625   definition unitary :: bool where "unitary = (ALL (x::'a) y. x = y)"
  2626 
  2627 * Predicates of locales introduced by classes carry a mandatory
  2628 "class" prefix.  INCOMPATIBILITY.
  2629 
  2630 * Vacuous class specifications observe default sort.  INCOMPATIBILITY.
  2631 
  2632 * Old 'axclass' command has been discontinued.  INCOMPATIBILITY, use
  2633 'class' instead.
  2634 
  2635 * Command 'code_reflect' allows to incorporate generated ML code into
  2636 runtime environment; replaces immature code_datatype antiquotation.
  2637 INCOMPATIBILITY.
  2638 
  2639 * Code generator: simple concept for abstract datatypes obeying
  2640 invariants.
  2641 
  2642 * Code generator: details of internal data cache have no impact on the
  2643 user space functionality any longer.
  2644 
  2645 * Methods "unfold_locales" and "intro_locales" ignore non-locale
  2646 subgoals.  This is more appropriate for interpretations with 'where'.
  2647 INCOMPATIBILITY.
  2648 
  2649 * Command 'example_proof' opens an empty proof body.  This allows to
  2650 experiment with Isar, without producing any persistent result.
  2651 
  2652 * Commands 'type_notation' and 'no_type_notation' declare type syntax
  2653 within a local theory context, with explicit checking of the
  2654 constructors involved (in contrast to the raw 'syntax' versions).
  2655 
  2656 * Commands 'types' and 'typedecl' now work within a local theory
  2657 context -- without introducing dependencies on parameters or
  2658 assumptions, which is not possible in Isabelle/Pure.
  2659 
  2660 * Command 'defaultsort' has been renamed to 'default_sort', it works
  2661 within a local theory context.  Minor INCOMPATIBILITY.
  2662 
  2663 
  2664 *** HOL ***
  2665 
  2666 * Command 'typedef' now works within a local theory context -- without
  2667 introducing dependencies on parameters or assumptions, which is not
  2668 possible in Isabelle/Pure/HOL.  Note that the logical environment may
  2669 contain multiple interpretations of local typedefs (with different
  2670 non-emptiness proofs), even in a global theory context.
  2671 
  2672 * New package for quotient types.  Commands 'quotient_type' and
  2673 'quotient_definition' may be used for defining types and constants by
  2674 quotient constructions.  An example is the type of integers created by
  2675 quotienting pairs of natural numbers:
  2676 
  2677   fun
  2678     intrel :: "(nat * nat) => (nat * nat) => bool"
  2679   where
  2680     "intrel (x, y) (u, v) = (x + v = u + y)"
  2681 
  2682   quotient_type int = "nat * nat" / intrel
  2683     by (auto simp add: equivp_def expand_fun_eq)
  2684 
  2685   quotient_definition
  2686     "0::int" is "(0::nat, 0::nat)"
  2687 
  2688 The method "lifting" can be used to lift of theorems from the
  2689 underlying "raw" type to the quotient type.  The example
  2690 src/HOL/Quotient_Examples/FSet.thy includes such a quotient
  2691 construction and provides a reasoning infrastructure for finite sets.
  2692 
  2693 * Renamed Library/Quotient.thy to Library/Quotient_Type.thy to avoid
  2694 clash with new theory Quotient in Main HOL.
  2695 
  2696 * Moved the SMT binding into the main HOL session, eliminating
  2697 separate HOL-SMT session.
  2698 
  2699 * List membership infix mem operation is only an input abbreviation.
  2700 INCOMPATIBILITY.
  2701 
  2702 * Theory Library/Word.thy has been removed.  Use library Word/Word.thy
  2703 for future developements; former Library/Word.thy is still present in
  2704 the AFP entry RSAPPS.
  2705 
  2706 * Theorem Int.int_induct renamed to Int.int_of_nat_induct and is no
  2707 longer shadowed.  INCOMPATIBILITY.
  2708 
  2709 * Dropped theorem duplicate comp_arith; use semiring_norm instead.
  2710 INCOMPATIBILITY.
  2711 
  2712 * Dropped theorem RealPow.real_sq_order; use power2_le_imp_le instead.
  2713 INCOMPATIBILITY.
  2714 
  2715 * Dropped normalizing_semiring etc; use the facts in semiring classes
  2716 instead.  INCOMPATIBILITY.
  2717 
  2718 * Dropped several real-specific versions of lemmas about floor and
  2719 ceiling; use the generic lemmas from theory "Archimedean_Field"
  2720 instead.  INCOMPATIBILITY.
  2721 
  2722   floor_number_of_eq         ~> floor_number_of
  2723   le_floor_eq_number_of      ~> number_of_le_floor
  2724   le_floor_eq_zero           ~> zero_le_floor
  2725   le_floor_eq_one            ~> one_le_floor
  2726   floor_less_eq_number_of    ~> floor_less_number_of
  2727   floor_less_eq_zero         ~> floor_less_zero
  2728   floor_less_eq_one          ~> floor_less_one
  2729   less_floor_eq_number_of    ~> number_of_less_floor
  2730   less_floor_eq_zero         ~> zero_less_floor
  2731   less_floor_eq_one          ~> one_less_floor
  2732   floor_le_eq_number_of      ~> floor_le_number_of
  2733   floor_le_eq_zero           ~> floor_le_zero
  2734   floor_le_eq_one            ~> floor_le_one
  2735   floor_subtract_number_of   ~> floor_diff_number_of
  2736   floor_subtract_one         ~> floor_diff_one
  2737   ceiling_number_of_eq       ~> ceiling_number_of
  2738   ceiling_le_eq_number_of    ~> ceiling_le_number_of
  2739   ceiling_le_zero_eq         ~> ceiling_le_zero
  2740   ceiling_le_eq_one          ~> ceiling_le_one
  2741   less_ceiling_eq_number_of  ~> number_of_less_ceiling
  2742   less_ceiling_eq_zero       ~> zero_less_ceiling
  2743   less_ceiling_eq_one        ~> one_less_ceiling
  2744   ceiling_less_eq_number_of  ~> ceiling_less_number_of
  2745   ceiling_less_eq_zero       ~> ceiling_less_zero
  2746   ceiling_less_eq_one        ~> ceiling_less_one
  2747   le_ceiling_eq_number_of    ~> number_of_le_ceiling
  2748   le_ceiling_eq_zero         ~> zero_le_ceiling
  2749   le_ceiling_eq_one          ~> one_le_ceiling
  2750   ceiling_subtract_number_of ~> ceiling_diff_number_of
  2751   ceiling_subtract_one       ~> ceiling_diff_one
  2752 
  2753 * Theory "Finite_Set": various folding_XXX locales facilitate the
  2754 application of the various fold combinators on finite sets.
  2755 
  2756 * Library theory "RBT" renamed to "RBT_Impl"; new library theory "RBT"
  2757 provides abstract red-black tree type which is backed by "RBT_Impl" as
  2758 implementation.  INCOMPATIBILTY.
  2759 
  2760 * Theory Library/Coinductive_List has been removed -- superseded by
  2761 AFP/thys/Coinductive.
  2762 
  2763 * Theory PReal, including the type "preal" and related operations, has
  2764 been removed.  INCOMPATIBILITY.
  2765 
  2766 * Real: new development using Cauchy Sequences.
  2767 
  2768 * Split off theory "Big_Operators" containing setsum, setprod,
  2769 Inf_fin, Sup_fin, Min, Max from theory Finite_Set.  INCOMPATIBILITY.
  2770 
  2771 * Theory "Rational" renamed to "Rat", for consistency with "Nat",
  2772 "Int" etc.  INCOMPATIBILITY.
  2773 
  2774 * Constant Rat.normalize needs to be qualified.  INCOMPATIBILITY.
  2775 
  2776 * New set of rules "ac_simps" provides combined assoc / commute
  2777 rewrites for all interpretations of the appropriate generic locales.
  2778 
  2779 * Renamed theory "OrderedGroup" to "Groups" and split theory
  2780 "Ring_and_Field" into theories "Rings" and "Fields"; for more
  2781 appropriate and more consistent names suitable for name prefixes
  2782 within the HOL theories.  INCOMPATIBILITY.
  2783 
  2784 * Some generic constants have been put to appropriate theories:
  2785   - less_eq, less: Orderings
  2786   - zero, one, plus, minus, uminus, times, abs, sgn: Groups
  2787   - inverse, divide: Rings
  2788 INCOMPATIBILITY.
  2789 
  2790 * More consistent naming of type classes involving orderings (and
  2791 lattices):
  2792 
  2793     lower_semilattice                   ~> semilattice_inf
  2794     upper_semilattice                   ~> semilattice_sup
  2795 
  2796     dense_linear_order                  ~> dense_linorder
  2797 
  2798     pordered_ab_group_add               ~> ordered_ab_group_add
  2799     pordered_ab_group_add_abs           ~> ordered_ab_group_add_abs
  2800     pordered_ab_semigroup_add           ~> ordered_ab_semigroup_add
  2801     pordered_ab_semigroup_add_imp_le    ~> ordered_ab_semigroup_add_imp_le
  2802     pordered_cancel_ab_semigroup_add    ~> ordered_cancel_ab_semigroup_add
  2803     pordered_cancel_comm_semiring       ~> ordered_cancel_comm_semiring
  2804     pordered_cancel_semiring            ~> ordered_cancel_semiring
  2805     pordered_comm_monoid_add            ~> ordered_comm_monoid_add
  2806     pordered_comm_ring                  ~> ordered_comm_ring
  2807     pordered_comm_semiring              ~> ordered_comm_semiring
  2808     pordered_ring                       ~> ordered_ring
  2809     pordered_ring_abs                   ~> ordered_ring_abs
  2810     pordered_semiring                   ~> ordered_semiring
  2811 
  2812     ordered_ab_group_add                ~> linordered_ab_group_add
  2813     ordered_ab_semigroup_add            ~> linordered_ab_semigroup_add
  2814     ordered_cancel_ab_semigroup_add     ~> linordered_cancel_ab_semigroup_add
  2815     ordered_comm_semiring_strict        ~> linordered_comm_semiring_strict
  2816     ordered_field                       ~> linordered_field
  2817     ordered_field_no_lb                 ~> linordered_field_no_lb
  2818     ordered_field_no_ub                 ~> linordered_field_no_ub
  2819     ordered_field_dense_linear_order    ~> dense_linordered_field
  2820     ordered_idom                        ~> linordered_idom
  2821     ordered_ring                        ~> linordered_ring
  2822     ordered_ring_le_cancel_factor       ~> linordered_ring_le_cancel_factor
  2823     ordered_ring_less_cancel_factor     ~> linordered_ring_less_cancel_factor
  2824     ordered_ring_strict                 ~> linordered_ring_strict
  2825     ordered_semidom                     ~> linordered_semidom
  2826     ordered_semiring                    ~> linordered_semiring
  2827     ordered_semiring_1                  ~> linordered_semiring_1
  2828     ordered_semiring_1_strict           ~> linordered_semiring_1_strict
  2829     ordered_semiring_strict             ~> linordered_semiring_strict
  2830 
  2831   The following slightly odd type classes have been moved to a
  2832   separate theory Library/Lattice_Algebras:
  2833 
  2834     lordered_ab_group_add               ~> lattice_ab_group_add
  2835     lordered_ab_group_add_abs           ~> lattice_ab_group_add_abs
  2836     lordered_ab_group_add_meet          ~> semilattice_inf_ab_group_add
  2837     lordered_ab_group_add_join          ~> semilattice_sup_ab_group_add
  2838     lordered_ring                       ~> lattice_ring
  2839 
  2840 INCOMPATIBILITY.
  2841 
  2842 * Refined field classes:
  2843   - classes division_ring_inverse_zero, field_inverse_zero,
  2844     linordered_field_inverse_zero include rule inverse 0 = 0 --
  2845     subsumes former division_by_zero class;
  2846   - numerous lemmas have been ported from field to division_ring.
  2847 INCOMPATIBILITY.
  2848 
  2849 * Refined algebra theorem collections:
  2850   - dropped theorem group group_simps, use algebra_simps instead;
  2851   - dropped theorem group ring_simps, use field_simps instead;
  2852   - proper theorem collection field_simps subsumes former theorem
  2853     groups field_eq_simps and field_simps;
  2854   - dropped lemma eq_minus_self_iff which is a duplicate for
  2855     equal_neg_zero.
  2856 INCOMPATIBILITY.
  2857 
  2858 * Theory Finite_Set and List: some lemmas have been generalized from
  2859 sets to lattices:
  2860 
  2861   fun_left_comm_idem_inter      ~> fun_left_comm_idem_inf
  2862   fun_left_comm_idem_union      ~> fun_left_comm_idem_sup
  2863   inter_Inter_fold_inter        ~> inf_Inf_fold_inf
  2864   union_Union_fold_union        ~> sup_Sup_fold_sup
  2865   Inter_fold_inter              ~> Inf_fold_inf
  2866   Union_fold_union              ~> Sup_fold_sup
  2867   inter_INTER_fold_inter        ~> inf_INFI_fold_inf
  2868   union_UNION_fold_union        ~> sup_SUPR_fold_sup
  2869   INTER_fold_inter              ~> INFI_fold_inf
  2870   UNION_fold_union              ~> SUPR_fold_sup
  2871 
  2872 * Theory "Complete_Lattice": lemmas top_def and bot_def have been
  2873 replaced by the more convenient lemmas Inf_empty and Sup_empty.
  2874 Dropped lemmas Inf_insert_simp and Sup_insert_simp, which are subsumed
  2875 by Inf_insert and Sup_insert.  Lemmas Inf_UNIV and Sup_UNIV replace
  2876 former Inf_Univ and Sup_Univ.  Lemmas inf_top_right and sup_bot_right
  2877 subsume inf_top and sup_bot respectively.  INCOMPATIBILITY.
  2878 
  2879 * Reorganized theory Multiset: swapped notation of pointwise and
  2880 multiset order:
  2881 
  2882   - pointwise ordering is instance of class order with standard syntax
  2883     <= and <;
  2884   - multiset ordering has syntax <=# and <#; partial order properties
  2885     are provided by means of interpretation with prefix
  2886     multiset_order;
  2887   - less duplication, less historical organization of sections,
  2888     conversion from associations lists to multisets, rudimentary code
  2889     generation;
  2890   - use insert_DiffM2 [symmetric] instead of elem_imp_eq_diff_union,
  2891     if needed.
  2892 
  2893 Renamed:
  2894 
  2895   multiset_eq_conv_count_eq  ~>  multiset_ext_iff
  2896   multi_count_ext  ~>  multiset_ext
  2897   diff_union_inverse2  ~>  diff_union_cancelR
  2898 
  2899 INCOMPATIBILITY.
  2900 
  2901 * Theory Permutation: replaced local "remove" by List.remove1.
  2902 
  2903 * Code generation: ML and OCaml code is decorated with signatures.
  2904 
  2905 * Theory List: added transpose.
  2906 
  2907 * Library/Nat_Bijection.thy is a collection of bijective functions
  2908 between nat and other types, which supersedes the older libraries
  2909 Library/Nat_Int_Bij.thy and HOLCF/NatIso.thy.  INCOMPATIBILITY.
  2910 
  2911   Constants:
  2912   Nat_Int_Bij.nat2_to_nat         ~> prod_encode
  2913   Nat_Int_Bij.nat_to_nat2         ~> prod_decode
  2914   Nat_Int_Bij.int_to_nat_bij      ~> int_encode
  2915   Nat_Int_Bij.nat_to_int_bij      ~> int_decode
  2916   Countable.pair_encode           ~> prod_encode
  2917   NatIso.prod2nat                 ~> prod_encode
  2918   NatIso.nat2prod                 ~> prod_decode
  2919   NatIso.sum2nat                  ~> sum_encode
  2920   NatIso.nat2sum                  ~> sum_decode
  2921   NatIso.list2nat                 ~> list_encode
  2922   NatIso.nat2list                 ~> list_decode
  2923   NatIso.set2nat                  ~> set_encode
  2924   NatIso.nat2set                  ~> set_decode
  2925 
  2926   Lemmas:
  2927   Nat_Int_Bij.bij_nat_to_int_bij  ~> bij_int_decode
  2928   Nat_Int_Bij.nat2_to_nat_inj     ~> inj_prod_encode
  2929   Nat_Int_Bij.nat2_to_nat_surj    ~> surj_prod_encode
  2930   Nat_Int_Bij.nat_to_nat2_inj     ~> inj_prod_decode
  2931   Nat_Int_Bij.nat_to_nat2_surj    ~> surj_prod_decode
  2932   Nat_Int_Bij.i2n_n2i_id          ~> int_encode_inverse
  2933   Nat_Int_Bij.n2i_i2n_id          ~> int_decode_inverse
  2934   Nat_Int_Bij.surj_nat_to_int_bij ~> surj_int_encode
  2935   Nat_Int_Bij.surj_int_to_nat_bij ~> surj_int_decode
  2936   Nat_Int_Bij.inj_nat_to_int_bij  ~> inj_int_encode
  2937   Nat_Int_Bij.inj_int_to_nat_bij  ~> inj_int_decode
  2938   Nat_Int_Bij.bij_nat_to_int_bij  ~> bij_int_encode
  2939   Nat_Int_Bij.bij_int_to_nat_bij  ~> bij_int_decode
  2940 
  2941 * Sledgehammer:
  2942   - Renamed ATP commands:
  2943     atp_info     ~> sledgehammer running_atps
  2944     atp_kill     ~> sledgehammer kill_atps
  2945     atp_messages ~> sledgehammer messages
  2946     atp_minimize ~> sledgehammer minimize
  2947     print_atps   ~> sledgehammer available_atps
  2948     INCOMPATIBILITY.
  2949   - Added user's manual ("isabelle doc sledgehammer").
  2950   - Added option syntax and "sledgehammer_params" to customize
  2951     Sledgehammer's behavior.  See the manual for details.
  2952   - Modified the Isar proof reconstruction code so that it produces
  2953     direct proofs rather than proofs by contradiction.  (This feature
  2954     is still experimental.)
  2955   - Made Isar proof reconstruction work for SPASS, remote ATPs, and in
  2956     full-typed mode.
  2957   - Added support for TPTP syntax for SPASS via the "spass_tptp" ATP.
  2958 
  2959 * Nitpick:
  2960   - Added and implemented "binary_ints" and "bits" options.
  2961   - Added "std" option and implemented support for nonstandard models.
  2962   - Added and implemented "finitize" option to improve the precision
  2963     of infinite datatypes based on a monotonicity analysis.
  2964   - Added support for quotient types.
  2965   - Added support for "specification" and "ax_specification"
  2966     constructs.
  2967   - Added support for local definitions (for "function" and
  2968     "termination" proofs).
  2969   - Added support for term postprocessors.
  2970   - Optimized "Multiset.multiset" and "FinFun.finfun".
  2971   - Improved efficiency of "destroy_constrs" optimization.
  2972   - Fixed soundness bugs related to "destroy_constrs" optimization and
  2973     record getters.
  2974   - Fixed soundness bug related to higher-order constructors.
  2975   - Fixed soundness bug when "full_descrs" is enabled.
  2976   - Improved precision of set constructs.
  2977   - Added "atoms" option.
  2978   - Added cache to speed up repeated Kodkod invocations on the same
  2979     problems.
  2980   - Renamed "MiniSatJNI", "zChaffJNI", "BerkMinAlloy", and
  2981     "SAT4JLight" to "MiniSat_JNI", "zChaff_JNI", "BerkMin_Alloy", and
  2982     "SAT4J_Light".  INCOMPATIBILITY.
  2983   - Removed "skolemize", "uncurry", "sym_break", "flatten_prop",
  2984     "sharing_depth", and "show_skolems" options.  INCOMPATIBILITY.
  2985   - Removed "nitpick_intro" attribute.  INCOMPATIBILITY.
  2986 
  2987 * Method "induct" now takes instantiations of the form t, where t is not
  2988   a variable, as a shorthand for "x == t", where x is a fresh variable.
  2989   If this is not intended, t has to be enclosed in parentheses.
  2990   By default, the equalities generated by definitional instantiations
  2991   are pre-simplified, which may cause parameters of inductive cases
  2992   to disappear, or may even delete some of the inductive cases.
  2993   Use "induct (no_simp)" instead of "induct" to restore the old
  2994   behaviour. The (no_simp) option is also understood by the "cases"
  2995   and "nominal_induct" methods, which now perform pre-simplification, too.
  2996   INCOMPATIBILITY.
  2997 
  2998 
  2999 *** HOLCF ***
  3000 
  3001 * Variable names in lemmas generated by the domain package have
  3002 changed; the naming scheme is now consistent with the HOL datatype
  3003 package.  Some proof scripts may be affected, INCOMPATIBILITY.
  3004 
  3005 * The domain package no longer defines the function "foo_copy" for
  3006 recursive domain "foo".  The reach lemma is now stated directly in
  3007 terms of "foo_take".  Lemmas and proofs that mention "foo_copy" must
  3008 be reformulated in terms of "foo_take", INCOMPATIBILITY.
  3009 
  3010 * Most definedness lemmas generated by the domain package (previously
  3011 of the form "x ~= UU ==> foo$x ~= UU") now have an if-and-only-if form
  3012 like "foo$x = UU <-> x = UU", which works better as a simp rule.
  3013 Proofs that used definedness lemmas as intro rules may break,
  3014 potential INCOMPATIBILITY.
  3015 
  3016 * Induction and casedist rules generated by the domain package now
  3017 declare proper case_names (one called "bottom", and one named for each
  3018 constructor).  INCOMPATIBILITY.
  3019 
  3020 * For mutually-recursive domains, separate "reach" and "take_lemma"
  3021 rules are generated for each domain, INCOMPATIBILITY.
  3022 
  3023   foo_bar.reach       ~> foo.reach  bar.reach
  3024   foo_bar.take_lemmas ~> foo.take_lemma  bar.take_lemma
  3025 
  3026 * Some lemmas generated by the domain package have been renamed for
  3027 consistency with the datatype package, INCOMPATIBILITY.
  3028 
  3029   foo.ind        ~> foo.induct
  3030   foo.finite_ind ~> foo.finite_induct
  3031   foo.coind      ~> foo.coinduct
  3032   foo.casedist   ~> foo.exhaust
  3033   foo.exhaust    ~> foo.nchotomy
  3034 
  3035 * For consistency with other definition packages, the fixrec package
  3036 now generates qualified theorem names, INCOMPATIBILITY.
  3037 
  3038   foo_simps  ~> foo.simps
  3039   foo_unfold ~> foo.unfold
  3040   foo_induct ~> foo.induct
  3041 
  3042 * The "fixrec_simp" attribute has been removed.  The "fixrec_simp"
  3043 method and internal fixrec proofs now use the default simpset instead.
  3044 INCOMPATIBILITY.
  3045 
  3046 * The "contlub" predicate has been removed.  Proof scripts should use
  3047 lemma contI2 in place of monocontlub2cont, INCOMPATIBILITY.
  3048 
  3049 * The "admw" predicate has been removed, INCOMPATIBILITY.
  3050 
  3051 * The constants cpair, cfst, and csnd have been removed in favor of
  3052 Pair, fst, and snd from Isabelle/HOL, INCOMPATIBILITY.
  3053 
  3054 
  3055 *** ML ***
  3056 
  3057 * Antiquotations for basic formal entities:
  3058 
  3059     @{class NAME}         -- type class
  3060     @{class_syntax NAME}  -- syntax representation of the above
  3061 
  3062     @{type_name NAME}     -- logical type
  3063     @{type_abbrev NAME}   -- type abbreviation
  3064     @{nonterminal NAME}   -- type of concrete syntactic category
  3065     @{type_syntax NAME}   -- syntax representation of any of the above
  3066 
  3067     @{const_name NAME}    -- logical constant (INCOMPATIBILITY)
  3068     @{const_abbrev NAME}  -- abbreviated constant
  3069     @{const_syntax NAME}  -- syntax representation of any of the above
  3070 
  3071 * Antiquotation @{syntax_const NAME} ensures that NAME refers to a raw
  3072 syntax constant (cf. 'syntax' command).
  3073 
  3074 * Antiquotation @{make_string} inlines a function to print arbitrary
  3075 values similar to the ML toplevel.  The result is compiler dependent
  3076 and may fall back on "?" in certain situations.
  3077 
  3078 * Diagnostic commands 'ML_val' and 'ML_command' may refer to
  3079 antiquotations @{Isar.state} and @{Isar.goal}.  This replaces impure
  3080 Isar.state() and Isar.goal(), which belong to the old TTY loop and do
  3081 not work with the asynchronous Isar document model.
  3082 
  3083 * Configuration options now admit dynamic default values, depending on
  3084 the context or even global references.
  3085 
  3086 * SHA1.digest digests strings according to SHA-1 (see RFC 3174).  It
  3087 uses an efficient external library if available (for Poly/ML).
  3088 
  3089 * Renamed some important ML structures, while keeping the old names
  3090 for some time as aliases within the structure Legacy:
  3091 
  3092   OuterKeyword  ~>  Keyword
  3093   OuterLex      ~>  Token
  3094   OuterParse    ~>  Parse
  3095   OuterSyntax   ~>  Outer_Syntax
  3096   PrintMode     ~>  Print_Mode
  3097   SpecParse     ~>  Parse_Spec
  3098   ThyInfo       ~>  Thy_Info
  3099   ThyLoad       ~>  Thy_Load
  3100   ThyOutput     ~>  Thy_Output
  3101   TypeInfer     ~>  Type_Infer
  3102 
  3103 Note that "open Legacy" simplifies porting of sources, but forgetting
  3104 to remove it again will complicate porting again in the future.
  3105 
  3106 * Most operations that refer to a global context are named
  3107 accordingly, e.g. Simplifier.global_context or
  3108 ProofContext.init_global.  There are some situations where a global
  3109 context actually works, but under normal circumstances one needs to
  3110 pass the proper local context through the code!
  3111 
  3112 * Discontinued old TheoryDataFun with its copy/init operation -- data
  3113 needs to be pure.  Functor Theory_Data_PP retains the traditional
  3114 Pretty.pp argument to merge, which is absent in the standard
  3115 Theory_Data version.
  3116 
  3117 * Sorts.certify_sort and derived "cert" operations for types and terms
  3118 no longer minimize sorts.  Thus certification at the boundary of the
  3119 inference kernel becomes invariant under addition of class relations,
  3120 which is an important monotonicity principle.  Sorts are now minimized
  3121 in the syntax layer only, at the boundary between the end-user and the
  3122 system.  Subtle INCOMPATIBILITY, may have to use Sign.minimize_sort
  3123 explicitly in rare situations.
  3124 
  3125 * Renamed old-style Drule.standard to Drule.export_without_context, to
  3126 emphasize that this is in no way a standard operation.
  3127 INCOMPATIBILITY.
  3128 
  3129 * Subgoal.FOCUS (and variants): resulting goal state is normalized as
  3130 usual for resolution.  Rare INCOMPATIBILITY.
  3131 
  3132 * Renamed varify/unvarify operations to varify_global/unvarify_global
  3133 to emphasize that these only work in a global situation (which is
  3134 quite rare).
  3135 
  3136 * Curried take and drop in library.ML; negative length is interpreted
  3137 as infinity (as in chop).  Subtle INCOMPATIBILITY.
  3138 
  3139 * Proof terms: type substitutions on proof constants now use canonical
  3140 order of type variables.  INCOMPATIBILITY for tools working with proof
  3141 terms.
  3142 
  3143 * Raw axioms/defs may no longer carry sort constraints, and raw defs
  3144 may no longer carry premises.  User-level specifications are
  3145 transformed accordingly by Thm.add_axiom/add_def.
  3146 
  3147 
  3148 *** System ***
  3149 
  3150 * Discontinued special HOL_USEDIR_OPTIONS for the main HOL image;
  3151 ISABELLE_USEDIR_OPTIONS applies uniformly to all sessions.  Note that
  3152 proof terms are enabled unconditionally in the new HOL-Proofs image.
  3153 
  3154 * Discontinued old ISABELLE and ISATOOL environment settings (legacy
  3155 feature since Isabelle2009).  Use ISABELLE_PROCESS and ISABELLE_TOOL,
  3156 respectively.
  3157 
  3158 * Old lib/scripts/polyml-platform is superseded by the
  3159 ISABELLE_PLATFORM setting variable, which defaults to the 32 bit
  3160 variant, even on a 64 bit machine.  The following example setting
  3161 prefers 64 bit if available:
  3162 
  3163   ML_PLATFORM="${ISABELLE_PLATFORM64:-$ISABELLE_PLATFORM}"
  3164 
  3165 * The preliminary Isabelle/jEdit application demonstrates the emerging
  3166 Isabelle/Scala layer for advanced prover interaction and integration.
  3167 See src/Tools/jEdit or "isabelle jedit" provided by the properly built
  3168 component.
  3169 
  3170 * "IsabelleText" is a Unicode font derived from Bitstream Vera Mono
  3171 and Bluesky TeX fonts.  It provides the usual Isabelle symbols,
  3172 similar to the default assignment of the document preparation system
  3173 (cf. isabellesym.sty).  The Isabelle/Scala class Isabelle_System
  3174 provides some operations for direct access to the font without asking
  3175 the user for manual installation.
  3176 
  3177 
  3178 
  3179 New in Isabelle2009-1 (December 2009)
  3180 -------------------------------------
  3181 
  3182 *** General ***
  3183 
  3184 * Discontinued old form of "escaped symbols" such as \\<forall>.  Only
  3185 one backslash should be used, even in ML sources.
  3186 
  3187 
  3188 *** Pure ***
  3189 
  3190 * Locale interpretation propagates mixins along the locale hierarchy.
  3191 The currently only available mixins are the equations used to map
  3192 local definitions to terms of the target domain of an interpretation.
  3193 
  3194 * Reactivated diagnostic command 'print_interps'.  Use "print_interps
  3195 loc" to print all interpretations of locale "loc" in the theory.
  3196 Interpretations in proofs are not shown.
  3197 
  3198 * Thoroughly revised locales tutorial.  New section on conditional
  3199 interpretation.
  3200 
  3201 * On instantiation of classes, remaining undefined class parameters
  3202 are formally declared.  INCOMPATIBILITY.
  3203 
  3204 
  3205 *** Document preparation ***
  3206 
  3207 * New generalized style concept for printing terms: @{foo (style) ...}
  3208 instead of @{foo_style style ...}  (old form is still retained for
  3209 backward compatibility).  Styles can be also applied for
  3210 antiquotations prop, term_type and typeof.
  3211 
  3212 
  3213 *** HOL ***
  3214 
  3215 * New proof method "smt" for a combination of first-order logic with
  3216 equality, linear and nonlinear (natural/integer/real) arithmetic, and
  3217 fixed-size bitvectors; there is also basic support for higher-order
  3218 features (esp. lambda abstractions).  It is an incomplete decision
  3219 procedure based on external SMT solvers using the oracle mechanism;
  3220 for the SMT solver Z3, this method is proof-producing.  Certificates
  3221 are provided to avoid calling the external solvers solely for
  3222 re-checking proofs.  Due to a remote SMT service there is no need for
  3223 installing SMT solvers locally.  See src/HOL/SMT.
  3224 
  3225 * New commands to load and prove verification conditions generated by
  3226 the Boogie program verifier or derived systems (e.g. the Verifying C
  3227 Compiler (VCC) or Spec#).  See src/HOL/Boogie.
  3228 
  3229 * New counterexample generator tool 'nitpick' based on the Kodkod
  3230 relational model finder.  See src/HOL/Tools/Nitpick and
  3231 src/HOL/Nitpick_Examples.
  3232 
  3233 * New commands 'code_pred' and 'values' to invoke the predicate
  3234 compiler and to enumerate values of inductive predicates.
  3235 
  3236 * A tabled implementation of the reflexive transitive closure.
  3237 
  3238 * New implementation of quickcheck uses generic code generator;
  3239 default generators are provided for all suitable HOL types, records
  3240 and datatypes.  Old quickcheck can be re-activated importing theory
  3241 Library/SML_Quickcheck.
  3242 
  3243 * New testing tool Mirabelle for automated proof tools.  Applies
  3244 several tools and tactics like sledgehammer, metis, or quickcheck, to
  3245 every proof step in a theory.  To be used in batch mode via the
  3246 "mirabelle" utility.
  3247 
  3248 * New proof method "sos" (sum of squares) for nonlinear real
  3249 arithmetic (originally due to John Harison). It requires theory
  3250 Library/Sum_Of_Squares.  It is not a complete decision procedure but
  3251 works well in practice on quantifier-free real arithmetic with +, -,
  3252 *, ^, =, <= and <, i.e. boolean combinations of equalities and
  3253 inequalities between polynomials.  It makes use of external
  3254 semidefinite programming solvers.  Method "sos" generates a
  3255 certificate that can be pasted into the proof thus avoiding the need
  3256 to call an external tool every time the proof is checked.  See
  3257 src/HOL/Library/Sum_Of_Squares.
  3258 
  3259 * New method "linarith" invokes existing linear arithmetic decision
  3260 procedure only.
  3261 
  3262 * New command 'atp_minimal' reduces result produced by Sledgehammer.
  3263 
  3264 * New Sledgehammer option "Full Types" in Proof General settings menu.
  3265 Causes full type information to be output to the ATPs.  This slows
  3266 ATPs down considerably but eliminates a source of unsound "proofs"
  3267 that fail later.
  3268 
  3269 * New method "metisFT": A version of metis that uses full type
  3270 information in order to avoid failures of proof reconstruction.
  3271 
  3272 * New evaluator "approximate" approximates an real valued term using
  3273 the same method as the approximation method.
  3274 
  3275 * Method "approximate" now supports arithmetic expressions as
  3276 boundaries of intervals and implements interval splitting and Taylor
  3277 series expansion.
  3278 
  3279 * ML antiquotation @{code_datatype} inserts definition of a datatype
  3280 generated by the code generator; e.g. see src/HOL/Predicate.thy.
  3281 
  3282 * New theory SupInf of the supremum and infimum operators for sets of
  3283 reals.
  3284 
  3285 * New theory Probability, which contains a development of measure
  3286 theory, eventually leading to Lebesgue integration and probability.
  3287 
  3288 * Extended Multivariate Analysis to include derivation and Brouwer's
  3289 fixpoint theorem.
  3290 
  3291 * Reorganization of number theory, INCOMPATIBILITY:
  3292   - new number theory development for nat and int, in theories Divides
  3293     and GCD as well as in new session Number_Theory
  3294   - some constants and facts now suffixed with _nat and _int
  3295     accordingly
  3296   - former session NumberTheory now named Old_Number_Theory, including
  3297     theories Legacy_GCD and Primes (prefer Number_Theory if possible)
  3298   - moved theory Pocklington from src/HOL/Library to
  3299     src/HOL/Old_Number_Theory
  3300 
  3301 * Theory GCD includes functions Gcd/GCD and Lcm/LCM for the gcd and
  3302 lcm of finite and infinite sets. It is shown that they form a complete
  3303 lattice.
  3304 
  3305 * Class semiring_div requires superclass no_zero_divisors and proof of
  3306 div_mult_mult1; theorems div_mult_mult1, div_mult_mult2,
  3307 div_mult_mult1_if, div_mult_mult1 and div_mult_mult2 have been
  3308 generalized to class semiring_div, subsuming former theorems
  3309 zdiv_zmult_zmult1, zdiv_zmult_zmult1_if, zdiv_zmult_zmult1 and
  3310 zdiv_zmult_zmult2.  div_mult_mult1 is now [simp] by default.
  3311 INCOMPATIBILITY.
  3312 
  3313 * Refinements to lattice classes and sets:
  3314   - less default intro/elim rules in locale variant, more default
  3315     intro/elim rules in class variant: more uniformity
  3316   - lemma ge_sup_conv renamed to le_sup_iff, in accordance with
  3317     le_inf_iff
  3318   - dropped lemma alias inf_ACI for inf_aci (same for sup_ACI and
  3319     sup_aci)
  3320   - renamed ACI to inf_sup_aci
  3321   - new class "boolean_algebra"
  3322   - class "complete_lattice" moved to separate theory
  3323     "Complete_Lattice"; corresponding constants (and abbreviations)
  3324     renamed and with authentic syntax:
  3325     Set.Inf ~>    Complete_Lattice.Inf
  3326     Set.Sup ~>    Complete_Lattice.Sup
  3327     Set.INFI ~>   Complete_Lattice.INFI
  3328     Set.SUPR ~>   Complete_Lattice.SUPR
  3329     Set.Inter ~>  Complete_Lattice.Inter
  3330     Set.Union ~>  Complete_Lattice.Union
  3331     Set.INTER ~>  Complete_Lattice.INTER
  3332     Set.UNION ~>  Complete_Lattice.UNION
  3333   - authentic syntax for
  3334     Set.Pow
  3335     Set.image
  3336   - mere abbreviations:
  3337     Set.empty               (for bot)
  3338     Set.UNIV                (for top)
  3339     Set.inter               (for inf, formerly Set.Int)
  3340     Set.union               (for sup, formerly Set.Un)
  3341     Complete_Lattice.Inter  (for Inf)
  3342     Complete_Lattice.Union  (for Sup)
  3343     Complete_Lattice.INTER  (for INFI)
  3344     Complete_Lattice.UNION  (for SUPR)
  3345   - object-logic definitions as far as appropriate
  3346 
  3347 INCOMPATIBILITY.  Care is required when theorems Int_subset_iff or
  3348 Un_subset_iff are explicitly deleted as default simp rules; then also
  3349 their lattice counterparts le_inf_iff and le_sup_iff have to be
  3350 deleted to achieve the desired effect.
  3351 
  3352 * Rules inf_absorb1, inf_absorb2, sup_absorb1, sup_absorb2 are no simp
  3353 rules by default any longer; the same applies to min_max.inf_absorb1
  3354 etc.  INCOMPATIBILITY.
  3355 
  3356 * Rules sup_Int_eq and sup_Un_eq are no longer declared as
  3357 pred_set_conv by default.  INCOMPATIBILITY.
  3358 
  3359 * Power operations on relations and functions are now one dedicated
  3360 constant "compow" with infix syntax "^^".  Power operation on
  3361 multiplicative monoids retains syntax "^" and is now defined generic
  3362 in class power.  INCOMPATIBILITY.
  3363 
  3364 * Relation composition "R O S" now has a more standard argument order:
  3365 "R O S = {(x, z). EX y. (x, y) : R & (y, z) : S}".  INCOMPATIBILITY,
  3366 rewrite propositions with "S O R" --> "R O S". Proofs may occasionally
  3367 break, since the O_assoc rule was not rewritten like this.  Fix using
  3368 O_assoc[symmetric].  The same applies to the curried version "R OO S".
  3369 
  3370 * Function "Inv" is renamed to "inv_into" and function "inv" is now an
  3371 abbreviation for "inv_into UNIV".  Lemmas are renamed accordingly.
  3372 INCOMPATIBILITY.
  3373 
  3374 * Most rules produced by inductive and datatype package have mandatory
  3375 prefixes.  INCOMPATIBILITY.
  3376 
  3377 * Changed "DERIV_intros" to a dynamic fact, which can be augmented by
  3378 the attribute of the same name.  Each of the theorems in the list
  3379 DERIV_intros assumes composition with an additional function and
  3380 matches a variable to the derivative, which has to be solved by the
  3381 Simplifier.  Hence (auto intro!: DERIV_intros) computes the derivative
  3382 of most elementary terms.  Former Maclauren.DERIV_tac and
  3383 Maclauren.deriv_tac should be replaced by (auto intro!: DERIV_intros).
  3384 INCOMPATIBILITY.
  3385 
  3386 * Code generator attributes follow the usual underscore convention:
  3387     code_unfold     replaces    code unfold
  3388     code_post       replaces    code post
  3389     etc.
  3390   INCOMPATIBILITY.
  3391 
  3392 * Renamed methods:
  3393     sizechange -> size_change
  3394     induct_scheme -> induction_schema
  3395   INCOMPATIBILITY.
  3396 
  3397 * Discontinued abbreviation "arbitrary" of constant "undefined".
  3398 INCOMPATIBILITY, use "undefined" directly.
  3399 
  3400 * Renamed theorems:
  3401     Suc_eq_add_numeral_1 -> Suc_eq_plus1
  3402     Suc_eq_add_numeral_1_left -> Suc_eq_plus1_left
  3403     Suc_plus1 -> Suc_eq_plus1
  3404     *anti_sym -> *antisym*
  3405     vector_less_eq_def -> vector_le_def
  3406   INCOMPATIBILITY.
  3407 
  3408 * Added theorem List.map_map as [simp].  Removed List.map_compose.
  3409 INCOMPATIBILITY.
  3410 
  3411 * Removed predicate "M hassize n" (<--> card M = n & finite M).
  3412 INCOMPATIBILITY.
  3413 
  3414 
  3415 *** HOLCF ***
  3416 
  3417 * Theory Representable defines a class "rep" of domains that are
  3418 representable (via an ep-pair) in the universal domain type "udom".
  3419 Instances are provided for all type constructors defined in HOLCF.
  3420 
  3421 * The 'new_domain' command is a purely definitional version of the
  3422 domain package, for representable domains.  Syntax is identical to the
  3423 old domain package.  The 'new_domain' package also supports indirect
  3424 recursion using previously-defined type constructors.  See
  3425 src/HOLCF/ex/New_Domain.thy for examples.
  3426 
  3427 * Method "fixrec_simp" unfolds one step of a fixrec-defined constant
  3428 on the left-hand side of an equation, and then performs
  3429 simplification.  Rewriting is done using rules declared with the
  3430 "fixrec_simp" attribute.  The "fixrec_simp" method is intended as a
  3431 replacement for "fixpat"; see src/HOLCF/ex/Fixrec_ex.thy for examples.
  3432 
  3433 * The pattern-match compiler in 'fixrec' can now handle constructors
  3434 with HOL function types.  Pattern-match combinators for the Pair
  3435 constructor are pre-configured.
  3436 
  3437 * The 'fixrec' package now produces better fixed-point induction rules
  3438 for mutually-recursive definitions:  Induction rules have conclusions
  3439 of the form "P foo bar" instead of "P <foo, bar>".
  3440 
  3441 * The constant "sq_le" (with infix syntax "<<" or "\<sqsubseteq>") has
  3442 been renamed to "below".  The name "below" now replaces "less" in many
  3443 theorem names.  (Legacy theorem names using "less" are still supported
  3444 as well.)
  3445 
  3446 * The 'fixrec' package now supports "bottom patterns".  Bottom
  3447 patterns can be used to generate strictness rules, or to make
  3448 functions more strict (much like the bang-patterns supported by the
  3449 Glasgow Haskell Compiler).  See src/HOLCF/ex/Fixrec_ex.thy for
  3450 examples.
  3451 
  3452 
  3453 *** ML ***
  3454 
  3455 * Support for Poly/ML 5.3.0, with improved reporting of compiler
  3456 errors and run-time exceptions, including detailed source positions.
  3457 
  3458 * Structure Name_Space (formerly NameSpace) now manages uniquely
  3459 identified entries, with some additional information such as source
  3460 position, logical grouping etc.
  3461 
  3462 * Theory and context data is now introduced by the simplified and
  3463 modernized functors Theory_Data, Proof_Data, Generic_Data.  Data needs
  3464 to be pure, but the old TheoryDataFun for mutable data (with explicit
  3465 copy operation) is still available for some time.
  3466 
  3467 * Structure Synchronized (cf. src/Pure/Concurrent/synchronized.ML)
  3468 provides a high-level programming interface to synchronized state
  3469 variables with atomic update.  This works via pure function
  3470 application within a critical section -- its runtime should be as
  3471 short as possible; beware of deadlocks if critical code is nested,
  3472 either directly or indirectly via other synchronized variables!
  3473 
  3474 * Structure Unsynchronized (cf. src/Pure/ML-Systems/unsynchronized.ML)
  3475 wraps raw ML references, explicitly indicating their non-thread-safe
  3476 behaviour.  The Isar toplevel keeps this structure open, to
  3477 accommodate Proof General as well as quick and dirty interactive
  3478 experiments with references.
  3479 
  3480 * PARALLEL_CHOICE and PARALLEL_GOALS provide basic support for
  3481 parallel tactical reasoning.
  3482 
  3483 * Tacticals Subgoal.FOCUS, Subgoal.FOCUS_PREMS, Subgoal.FOCUS_PARAMS
  3484 are similar to SUBPROOF, but are slightly more flexible: only the
  3485 specified parts of the subgoal are imported into the context, and the
  3486 body tactic may introduce new subgoals and schematic variables.
  3487 
  3488 * Old tactical METAHYPS, which does not observe the proof context, has
  3489 been renamed to Old_Goals.METAHYPS and awaits deletion.  Use SUBPROOF
  3490 or Subgoal.FOCUS etc.
  3491 
  3492 * Renamed functor TableFun to Table, and GraphFun to Graph.  (Since
  3493 functors have their own ML name space there is no point to mark them
  3494 separately.)  Minor INCOMPATIBILITY.
  3495 
  3496 * Renamed NamedThmsFun to Named_Thms.  INCOMPATIBILITY.
  3497 
  3498 * Renamed several structures FooBar to Foo_Bar.  Occasional,
  3499 INCOMPATIBILITY.
  3500 
  3501 * Operations of structure Skip_Proof no longer require quick_and_dirty
  3502 mode, which avoids critical setmp.
  3503 
  3504 * Eliminated old Attrib.add_attributes, Method.add_methods and related
  3505 combinators for "args".  INCOMPATIBILITY, need to use simplified
  3506 Attrib/Method.setup introduced in Isabelle2009.
  3507 
  3508 * Proper context for simpset_of, claset_of, clasimpset_of.  May fall
  3509 back on global_simpset_of, global_claset_of, global_clasimpset_of as
  3510 last resort.  INCOMPATIBILITY.
  3511 
  3512 * Display.pretty_thm now requires a proper context (cf. former
  3513 ProofContext.pretty_thm).  May fall back on Display.pretty_thm_global
  3514 or even Display.pretty_thm_without_context as last resort.
  3515 INCOMPATIBILITY.
  3516 
  3517 * Discontinued Display.pretty_ctyp/cterm etc.  INCOMPATIBILITY, use
  3518 Syntax.pretty_typ/term directly, preferably with proper context
  3519 instead of global theory.
  3520 
  3521 
  3522 *** System ***
  3523 
  3524 * Further fine tuning of parallel proof checking, scales up to 8 cores
  3525 (max. speedup factor 5.0).  See also Goal.parallel_proofs in ML and
  3526 usedir option -q.
  3527 
  3528 * Support for additional "Isabelle components" via etc/components, see
  3529 also the system manual.
  3530 
  3531 * The isabelle makeall tool now operates on all components with
  3532 IsaMakefile, not just hardwired "logics".
  3533 
  3534 * Removed "compress" option from isabelle-process and isabelle usedir;
  3535 this is always enabled.
  3536 
  3537 * Discontinued support for Poly/ML 4.x versions.
  3538 
  3539 * Isabelle tool "wwwfind" provides web interface for 'find_theorems'
  3540 on a given logic image.  This requires the lighttpd webserver and is
  3541 currently supported on Linux only.
  3542 
  3543 
  3544 
  3545 New in Isabelle2009 (April 2009)
  3546 --------------------------------
  3547 
  3548 *** General ***
  3549 
  3550 * Simplified main Isabelle executables, with less surprises on
  3551 case-insensitive file-systems (such as Mac OS).
  3552 
  3553   - The main Isabelle tool wrapper is now called "isabelle" instead of
  3554     "isatool."
  3555 
  3556   - The former "isabelle" alias for "isabelle-process" has been
  3557     removed (should rarely occur to regular users).
  3558 
  3559   - The former "isabelle-interface" and its alias "Isabelle" have been
  3560     removed (interfaces are now regular Isabelle tools).
  3561 
  3562 Within scripts and make files, the Isabelle environment variables
  3563 ISABELLE_TOOL and ISABELLE_PROCESS replace old ISATOOL and ISABELLE,
  3564 respectively.  (The latter are still available as legacy feature.)
  3565 
  3566 The old isabelle-interface wrapper could react in confusing ways if
  3567 the interface was uninstalled or changed otherwise.  Individual
  3568 interface tool configuration is now more explicit, see also the
  3569 Isabelle system manual.  In particular, Proof General is now available
  3570 via "isabelle emacs".
  3571 
  3572 INCOMPATIBILITY, need to adapt derivative scripts.  Users may need to
  3573 purge installed copies of Isabelle executables and re-run "isabelle
  3574 install -p ...", or use symlinks.
  3575 
  3576 * The default for ISABELLE_HOME_USER is now ~/.isabelle instead of the
  3577 old ~/isabelle, which was slightly non-standard and apt to cause
  3578 surprises on case-insensitive file-systems (such as Mac OS).
  3579 
  3580 INCOMPATIBILITY, need to move existing ~/isabelle/etc,
  3581 ~/isabelle/heaps, ~/isabelle/browser_info to the new place.  Special
  3582 care is required when using older releases of Isabelle.  Note that
  3583 ISABELLE_HOME_USER can be changed in Isabelle/etc/settings of any
  3584 Isabelle distribution, in order to use the new ~/.isabelle uniformly.
  3585 
  3586 * Proofs of fully specified statements are run in parallel on
  3587 multi-core systems.  A speedup factor of 2.5 to 3.2 can be expected on
  3588 a regular 4-core machine, if the initial heap space is made reasonably
  3589 large (cf. Poly/ML option -H).  (Requires Poly/ML 5.2.1 or later.)
  3590 
  3591 * The main reference manuals ("isar-ref", "implementation", and
  3592 "system") have been updated and extended.  Formally checked references
  3593 as hyperlinks are now available uniformly.
  3594 
  3595 
  3596 *** Pure ***
  3597 
  3598 * Complete re-implementation of locales.  INCOMPATIBILITY in several
  3599 respects.  The most important changes are listed below.  See the
  3600 Tutorial on Locales ("locales" manual) for details.
  3601 
  3602 - In locale expressions, instantiation replaces renaming.  Parameters
  3603 must be declared in a for clause.  To aid compatibility with previous
  3604 parameter inheritance, in locale declarations, parameters that are not
  3605 'touched' (instantiation position "_" or omitted) are implicitly added
  3606 with their syntax at the beginning of the for clause.
  3607 
  3608 - Syntax from abbreviations and definitions in locales is available in
  3609 locale expressions and context elements.  The latter is particularly
  3610 useful in locale declarations.
  3611 
  3612 - More flexible mechanisms to qualify names generated by locale
  3613 expressions.  Qualifiers (prefixes) may be specified in locale
  3614 expressions, and can be marked as mandatory (syntax: "name!:") or
  3615 optional (syntax "name?:").  The default depends for plain "name:"
  3616 depends on the situation where a locale expression is used: in
  3617 commands 'locale' and 'sublocale' prefixes are optional, in
  3618 'interpretation' and 'interpret' prefixes are mandatory.  The old
  3619 implicit qualifiers derived from the parameter names of a locale are
  3620 no longer generated.
  3621 
  3622 - Command "sublocale l < e" replaces "interpretation l < e".  The
  3623 instantiation clause in "interpretation" and "interpret" (square
  3624 brackets) is no longer available.  Use locale expressions.
  3625 
  3626 - When converting proof scripts, mandatory qualifiers in
  3627 'interpretation' and 'interpret' should be retained by default, even
  3628 if this is an INCOMPATIBILITY compared to former behavior.  In the
  3629 worst case, use the "name?:" form for non-mandatory ones.  Qualifiers
  3630 in locale expressions range over a single locale instance only.
  3631 
  3632 - Dropped locale element "includes".  This is a major INCOMPATIBILITY.
  3633 In existing theorem specifications replace the includes element by the
  3634 respective context elements of the included locale, omitting those
  3635 that are already present in the theorem specification.  Multiple
  3636 assume elements of a locale should be replaced by a single one
  3637 involving the locale predicate.  In the proof body, declarations (most
  3638 notably theorems) may be regained by interpreting the respective
  3639 locales in the proof context as required (command "interpret").
  3640 
  3641 If using "includes" in replacement of a target solely because the
  3642 parameter types in the theorem are not as general as in the target,
  3643 consider declaring a new locale with additional type constraints on
  3644 the parameters (context element "constrains").
  3645 
  3646 - Discontinued "locale (open)".  INCOMPATIBILITY.
  3647 
  3648 - Locale interpretation commands no longer attempt to simplify goal.
  3649 INCOMPATIBILITY: in rare situations the generated goal differs.  Use
  3650 methods intro_locales and unfold_locales to clarify.
  3651 
  3652 - Locale interpretation commands no longer accept interpretation
  3653 attributes.  INCOMPATIBILITY.
  3654 
  3655 * Class declaration: so-called "base sort" must not be given in import
  3656 list any longer, but is inferred from the specification.  Particularly
  3657 in HOL, write
  3658 
  3659     class foo = ...
  3660 
  3661 instead of
  3662 
  3663     class foo = type + ...
  3664 
  3665 * Class target: global versions of theorems stemming do not carry a
  3666 parameter prefix any longer.  INCOMPATIBILITY.
  3667 
  3668 * Class 'instance' command no longer accepts attached definitions.
  3669 INCOMPATIBILITY, use proper 'instantiation' target instead.
  3670 
  3671 * Recovered hiding of consts, which was accidentally broken in
  3672 Isabelle2007.  Potential INCOMPATIBILITY, ``hide const c'' really
  3673 makes c inaccessible; consider using ``hide (open) const c'' instead.
  3674 
  3675 * Slightly more coherent Pure syntax, with updated documentation in
  3676 isar-ref manual.  Removed locales meta_term_syntax and
  3677 meta_conjunction_syntax: TERM and &&& (formerly &&) are now permanent,
  3678 INCOMPATIBILITY in rare situations.  Note that &&& should not be used
  3679 directly in regular applications.
  3680 
  3681 * There is a new syntactic category "float_const" for signed decimal
  3682 fractions (e.g. 123.45 or -123.45).
  3683 
  3684 * Removed exotic 'token_translation' command.  INCOMPATIBILITY, use ML
  3685 interface with 'setup' command instead.
  3686 
  3687 * Command 'local_setup' is similar to 'setup', but operates on a local
  3688 theory context.
  3689 
  3690 * The 'axiomatization' command now only works within a global theory
  3691 context.  INCOMPATIBILITY.
  3692 
  3693 * Goal-directed proof now enforces strict proof irrelevance wrt. sort
  3694 hypotheses.  Sorts required in the course of reasoning need to be
  3695 covered by the constraints in the initial statement, completed by the
  3696 type instance information of the background theory.  Non-trivial sort
  3697 hypotheses, which rarely occur in practice, may be specified via
  3698 vacuous propositions of the form SORT_CONSTRAINT('a::c).  For example:
  3699 
  3700   lemma assumes "SORT_CONSTRAINT('a::empty)" shows False ...
  3701 
  3702 The result contains an implicit sort hypotheses as before --
  3703 SORT_CONSTRAINT premises are eliminated as part of the canonical rule
  3704 normalization.
  3705 
  3706 * Generalized Isar history, with support for linear undo, direct state
  3707 addressing etc.
  3708 
  3709 * Changed defaults for unify configuration options:
  3710 
  3711   unify_trace_bound = 50 (formerly 25)
  3712   unify_search_bound = 60 (formerly 30)
  3713 
  3714 * Different bookkeeping for code equations (INCOMPATIBILITY):
  3715 
  3716   a) On theory merge, the last set of code equations for a particular
  3717      constant is taken (in accordance with the policy applied by other
  3718      parts of the code generator framework).
  3719 
  3720   b) Code equations stemming from explicit declarations (e.g. code
  3721      attribute) gain priority over default code equations stemming
  3722      from definition, primrec, fun etc.
  3723 
  3724 * Keyword 'code_exception' now named 'code_abort'.  INCOMPATIBILITY.
  3725 
  3726 * Unified theorem tables for both code generators.  Thus [code
  3727 func] has disappeared and only [code] remains.  INCOMPATIBILITY.
  3728 
  3729 * Command 'find_consts' searches for constants based on type and name
  3730 patterns, e.g.
  3731 
  3732     find_consts "_ => bool"
  3733 
  3734 By default, matching is against subtypes, but it may be restricted to
  3735 the whole type.  Searching by name is possible.  Multiple queries are
  3736 conjunctive and queries may be negated by prefixing them with a
  3737 hyphen:
  3738 
  3739     find_consts strict: "_ => bool" name: "Int" -"int => int"
  3740 
  3741 * New 'find_theorems' criterion "solves" matches theorems that
  3742 directly solve the current goal (modulo higher-order unification).
  3743 
  3744 * Auto solve feature for main theorem statements: whenever a new goal
  3745 is stated, "find_theorems solves" is called; any theorems that could
  3746 solve the lemma directly are listed as part of the goal state.
  3747 Cf. associated options in Proof General Isabelle settings menu,
  3748 enabled by default, with reasonable timeout for pathological cases of
  3749 higher-order unification.
  3750 
  3751 
  3752 *** Document preparation ***
  3753 
  3754 * Antiquotation @{lemma} now imitates a regular terminal proof,
  3755 demanding keyword 'by' and supporting the full method expression
  3756 syntax just like the Isar command 'by'.
  3757 
  3758 
  3759 *** HOL ***
  3760 
  3761 * Integrated main parts of former image HOL-Complex with HOL.  Entry
  3762 points Main and Complex_Main remain as before.
  3763 
  3764 * Logic image HOL-Plain provides a minimal HOL with the most important
  3765 tools available (inductive, datatype, primrec, ...).  This facilitates
  3766 experimentation and tool development.  Note that user applications
  3767 (and library theories) should never refer to anything below theory
  3768 Main, as before.
  3769 
  3770 * Logic image HOL-Main stops at theory Main, and thus facilitates
  3771 experimentation due to shorter build times.
  3772 
  3773 * Logic image HOL-NSA contains theories of nonstandard analysis which
  3774 were previously part of former HOL-Complex.  Entry point Hyperreal
  3775 remains valid, but theories formerly using Complex_Main should now use
  3776 new entry point Hypercomplex.
  3777 
  3778 * Generic ATP manager for Sledgehammer, based on ML threads instead of
  3779 Posix processes.  Avoids potentially expensive forking of the ML
  3780 process.  New thread-based implementation also works on non-Unix
  3781 platforms (Cygwin).  Provers are no longer hardwired, but defined
  3782 within the theory via plain ML wrapper functions.  Basic Sledgehammer
  3783 commands are covered in the isar-ref manual.
  3784 
  3785 * Wrapper scripts for remote SystemOnTPTP service allows to use
  3786 sledgehammer without local ATP installation (Vampire etc.). Other
  3787 provers may be included via suitable ML wrappers, see also
  3788 src/HOL/ATP_Linkup.thy.
  3789 
  3790 * ATP selection (E/Vampire/Spass) is now via Proof General's settings
  3791 menu.
  3792 
  3793 * The metis method no longer fails because the theorem is too trivial
  3794 (contains the empty clause).
  3795 
  3796 * The metis method now fails in the usual manner, rather than raising
  3797 an exception, if it determines that it cannot prove the theorem.
  3798 
  3799 * Method "coherent" implements a prover for coherent logic (see also
  3800 src/Tools/coherent.ML).
  3801 
  3802 * Constants "undefined" and "default" replace "arbitrary".  Usually
  3803 "undefined" is the right choice to replace "arbitrary", though
  3804 logically there is no difference.  INCOMPATIBILITY.
  3805 
  3806 * Command "value" now integrates different evaluation mechanisms.  The
  3807 result of the first successful evaluation mechanism is printed.  In
  3808 square brackets a particular named evaluation mechanisms may be
  3809 specified (currently, [SML], [code] or [nbe]).  See further
  3810 src/HOL/ex/Eval_Examples.thy.
  3811 
  3812 * Normalization by evaluation now allows non-leftlinear equations.
  3813 Declare with attribute [code nbe].
  3814 
  3815 * Methods "case_tac" and "induct_tac" now refer to the very same rules
  3816 as the structured Isar versions "cases" and "induct", cf. the
  3817 corresponding "cases" and "induct" attributes.  Mutual induction rules
  3818 are now presented as a list of individual projections
  3819 (e.g. foo_bar.inducts for types foo and bar); the old format with
  3820 explicit HOL conjunction is no longer supported.  INCOMPATIBILITY, in
  3821 rare situations a different rule is selected --- notably nested tuple
  3822 elimination instead of former prod.exhaust: use explicit (case_tac t
  3823 rule: prod.exhaust) here.
  3824 
  3825 * Attributes "cases", "induct", "coinduct" support "del" option.
  3826 
  3827 * Removed fact "case_split_thm", which duplicates "case_split".
  3828 
  3829 * The option datatype has been moved to a new theory Option.  Renamed
  3830 option_map to Option.map, and o2s to Option.set, INCOMPATIBILITY.
  3831 
  3832 * New predicate "strict_mono" classifies strict functions on partial
  3833 orders.  With strict functions on linear orders, reasoning about
  3834 (in)equalities is facilitated by theorems "strict_mono_eq",
  3835 "strict_mono_less_eq" and "strict_mono_less".
  3836 
  3837 * Some set operations are now proper qualified constants with
  3838 authentic syntax.  INCOMPATIBILITY:
  3839 
  3840     op Int ~>   Set.Int
  3841     op Un ~>    Set.Un
  3842     INTER ~>    Set.INTER
  3843     UNION ~>    Set.UNION
  3844     Inter ~>    Set.Inter
  3845     Union ~>    Set.Union
  3846     {} ~>       Set.empty
  3847     UNIV ~>     Set.UNIV
  3848 
  3849 * Class complete_lattice with operations Inf, Sup, INFI, SUPR now in
  3850 theory Set.
  3851 
  3852 * Auxiliary class "itself" has disappeared -- classes without any
  3853 parameter are treated as expected by the 'class' command.
  3854 
  3855 * Leibnitz's Series for Pi and the arcus tangens and logarithm series.
  3856 
  3857 * Common decision procedures (Cooper, MIR, Ferrack, Approximation,
  3858 Dense_Linear_Order) are now in directory HOL/Decision_Procs.
  3859 
  3860 * Theory src/HOL/Decision_Procs/Approximation provides the new proof
  3861 method "approximation".  It proves formulas on real values by using
  3862 interval arithmetic.  In the formulas are also the transcendental
  3863 functions sin, cos, tan, atan, ln, exp and the constant pi are
  3864 allowed. For examples see
  3865 src/HOL/Descision_Procs/ex/Approximation_Ex.thy.
  3866 
  3867 * Theory "Reflection" now resides in HOL/Library.
  3868 
  3869 * Entry point to Word library now simply named "Word".
  3870 INCOMPATIBILITY.
  3871 
  3872 * Made source layout more coherent with logical distribution
  3873 structure:
  3874 
  3875     src/HOL/Library/RType.thy ~> src/HOL/Typerep.thy
  3876     src/HOL/Library/Code_Message.thy ~> src/HOL/
  3877     src/HOL/Library/GCD.thy ~> src/HOL/
  3878     src/HOL/Library/Order_Relation.thy ~> src/HOL/
  3879     src/HOL/Library/Parity.thy ~> src/HOL/
  3880     src/HOL/Library/Univ_Poly.thy ~> src/HOL/
  3881     src/HOL/Real/ContNotDenum.thy ~> src/HOL/Library/
  3882     src/HOL/Real/Lubs.thy ~> src/HOL/
  3883     src/HOL/Real/PReal.thy ~> src/HOL/
  3884     src/HOL/Real/Rational.thy ~> src/HOL/
  3885     src/HOL/Real/RComplete.thy ~> src/HOL/
  3886     src/HOL/Real/RealDef.thy ~> src/HOL/
  3887     src/HOL/Real/RealPow.thy ~> src/HOL/
  3888     src/HOL/Real/Real.thy ~> src/HOL/
  3889     src/HOL/Complex/Complex_Main.thy ~> src/HOL/
  3890     src/HOL/Complex/Complex.thy ~> src/HOL/
  3891     src/HOL/Complex/FrechetDeriv.thy ~> src/HOL/Library/
  3892     src/HOL/Complex/Fundamental_Theorem_Algebra.thy ~> src/HOL/Library/
  3893     src/HOL/Hyperreal/Deriv.thy ~> src/HOL/
  3894     src/HOL/Hyperreal/Fact.thy ~> src/HOL/
  3895     src/HOL/Hyperreal/Integration.thy ~> src/HOL/
  3896     src/HOL/Hyperreal/Lim.thy ~> src/HOL/
  3897     src/HOL/Hyperreal/Ln.thy ~> src/HOL/
  3898     src/HOL/Hyperreal/Log.thy ~> src/HOL/
  3899     src/HOL/Hyperreal/MacLaurin.thy ~> src/HOL/
  3900     src/HOL/Hyperreal/NthRoot.thy ~> src/HOL/
  3901     src/HOL/Hyperreal/Series.thy ~> src/HOL/
  3902     src/HOL/Hyperreal/SEQ.thy ~> src/HOL/
  3903     src/HOL/Hyperreal/Taylor.thy ~> src/HOL/
  3904     src/HOL/Hyperreal/Transcendental.thy ~> src/HOL/
  3905     src/HOL/Real/Float ~> src/HOL/Library/
  3906     src/HOL/Real/HahnBanach ~> src/HOL/HahnBanach
  3907     src/HOL/Real/RealVector.thy ~> src/HOL/
  3908 
  3909     src/HOL/arith_data.ML ~> src/HOL/Tools
  3910     src/HOL/hologic.ML ~> src/HOL/Tools
  3911     src/HOL/simpdata.ML ~> src/HOL/Tools
  3912     src/HOL/int_arith1.ML ~> src/HOL/Tools/int_arith.ML
  3913     src/HOL/int_factor_simprocs.ML ~> src/HOL/Tools
  3914     src/HOL/nat_simprocs.ML ~> src/HOL/Tools
  3915     src/HOL/Real/float_arith.ML ~> src/HOL/Tools
  3916     src/HOL/Real/float_syntax.ML ~> src/HOL/Tools
  3917     src/HOL/Real/rat_arith.ML ~> src/HOL/Tools
  3918     src/HOL/Real/real_arith.ML ~> src/HOL/Tools
  3919 
  3920     src/HOL/Library/Array.thy ~> src/HOL/Imperative_HOL
  3921     src/HOL/Library/Heap_Monad.thy ~> src/HOL/Imperative_HOL
  3922     src/HOL/Library/Heap.thy ~> src/HOL/Imperative_HOL
  3923     src/HOL/Library/Imperative_HOL.thy ~> src/HOL/Imperative_HOL
  3924     src/HOL/Library/Ref.thy ~> src/HOL/Imperative_HOL
  3925     src/HOL/Library/Relational.thy ~> src/HOL/Imperative_HOL
  3926 
  3927 * If methods "eval" and "evaluation" encounter a structured proof
  3928 state with !!/==>, only the conclusion is evaluated to True (if
  3929 possible), avoiding strange error messages.
  3930 
  3931 * Method "sizechange" automates termination proofs using (a
  3932 modification of) the size-change principle.  Requires SAT solver.  See
  3933 src/HOL/ex/Termination.thy for examples.
  3934 
  3935 * Simplifier: simproc for let expressions now unfolds if bound
  3936 variable occurs at most once in let expression body.  INCOMPATIBILITY.
  3937 
  3938 * Method "arith": Linear arithmetic now ignores all inequalities when
  3939 fast_arith_neq_limit is exceeded, instead of giving up entirely.
  3940 
  3941 * New attribute "arith" for facts that should always be used
  3942 automatically by arithmetic. It is intended to be used locally in
  3943 proofs, e.g.
  3944 
  3945   assumes [arith]: "x > 0"
  3946 
  3947 Global usage is discouraged because of possible performance impact.
  3948 
  3949 * New classes "top" and "bot" with corresponding operations "top" and
  3950 "bot" in theory Orderings; instantiation of class "complete_lattice"
  3951 requires instantiation of classes "top" and "bot".  INCOMPATIBILITY.
  3952 
  3953 * Changed definition lemma "less_fun_def" in order to provide an
  3954 instance for preorders on functions; use lemma "less_le" instead.
  3955 INCOMPATIBILITY.
  3956 
  3957 * Theory Orderings: class "wellorder" moved here, with explicit
  3958 induction rule "less_induct" as assumption.  For instantiation of
  3959 "wellorder" by means of predicate "wf", use rule wf_wellorderI.
  3960 INCOMPATIBILITY.
  3961 
  3962 * Theory Orderings: added class "preorder" as superclass of "order".
  3963 INCOMPATIBILITY: Instantiation proofs for order, linorder
  3964 etc. slightly changed.  Some theorems named order_class.* now named
  3965 preorder_class.*.
  3966 
  3967 * Theory Relation: renamed "refl" to "refl_on", "reflexive" to "refl,
  3968 "diag" to "Id_on".
  3969 
  3970 * Theory Finite_Set: added a new fold combinator of type
  3971 
  3972   ('a => 'b => 'b) => 'b => 'a set => 'b
  3973 
  3974 Occasionally this is more convenient than the old fold combinator
  3975 which is now defined in terms of the new one and renamed to
  3976 fold_image.
  3977 
  3978 * Theories Ring_and_Field and OrderedGroup: The lemmas "group_simps"
  3979 and "ring_simps" have been replaced by "algebra_simps" (which can be
  3980 extended with further lemmas!).  At the moment both still exist but
  3981 the former will disappear at some point.
  3982 
  3983 * Theory Power: Lemma power_Suc is now declared as a simp rule in
  3984 class recpower.  Type-specific simp rules for various recpower types
  3985 have been removed.  INCOMPATIBILITY, rename old lemmas as follows:
  3986 
  3987 rat_power_0    -> power_0
  3988 rat_power_Suc  -> power_Suc
  3989 realpow_0      -> power_0
  3990 realpow_Suc    -> power_Suc
  3991 complexpow_0   -> power_0
  3992 complexpow_Suc -> power_Suc
  3993 power_poly_0   -> power_0
  3994 power_poly_Suc -> power_Suc
  3995 
  3996 * Theories Ring_and_Field and Divides: Definition of "op dvd" has been
  3997 moved to separate class dvd in Ring_and_Field; a couple of lemmas on
  3998 dvd has been generalized to class comm_semiring_1.  Likewise a bunch
  3999 of lemmas from Divides has been generalized from nat to class
  4000 semiring_div.  INCOMPATIBILITY.  This involves the following theorem
  4001 renames resulting from duplicate elimination:
  4002 
  4003     dvd_def_mod ~>          dvd_eq_mod_eq_0
  4004     zero_dvd_iff ~>         dvd_0_left_iff
  4005     dvd_0 ~>                dvd_0_right
  4006     DIVISION_BY_ZERO_DIV ~> div_by_0
  4007     DIVISION_BY_ZERO_MOD ~> mod_by_0
  4008     mult_div ~>             div_mult_self2_is_id
  4009     mult_mod ~>             mod_mult_self2_is_0
  4010 
  4011 * Theory IntDiv: removed many lemmas that are instances of class-based
  4012 generalizations (from Divides and Ring_and_Field).  INCOMPATIBILITY,
  4013 rename old lemmas as follows:
  4014 
  4015 dvd_diff               -> nat_dvd_diff
  4016 dvd_zminus_iff         -> dvd_minus_iff
  4017 mod_add1_eq            -> mod_add_eq
  4018 mod_mult1_eq           -> mod_mult_right_eq
  4019 mod_mult1_eq'          -> mod_mult_left_eq
  4020 mod_mult_distrib_mod   -> mod_mult_eq
  4021 nat_mod_add_left_eq    -> mod_add_left_eq
  4022 nat_mod_add_right_eq   -> mod_add_right_eq
  4023 nat_mod_div_trivial    -> mod_div_trivial
  4024 nat_mod_mod_trivial    -> mod_mod_trivial
  4025 zdiv_zadd_self1        -> div_add_self1
  4026 zdiv_zadd_self2        -> div_add_self2
  4027 zdiv_zmult_self1       -> div_mult_self2_is_id
  4028 zdiv_zmult_self2       -> div_mult_self1_is_id
  4029 zdvd_triv_left         -> dvd_triv_left
  4030 zdvd_triv_right        -> dvd_triv_right
  4031 zdvd_zmult_cancel_disj -> dvd_mult_cancel_left
  4032 zmod_eq0_zdvd_iff      -> dvd_eq_mod_eq_0[symmetric]
  4033 zmod_zadd_left_eq      -> mod_add_left_eq
  4034 zmod_zadd_right_eq     -> mod_add_right_eq
  4035 zmod_zadd_self1        -> mod_add_self1
  4036 zmod_zadd_self2        -> mod_add_self2
  4037 zmod_zadd1_eq          -> mod_add_eq
  4038 zmod_zdiff1_eq         -> mod_diff_eq
  4039 zmod_zdvd_zmod         -> mod_mod_cancel
  4040 zmod_zmod_cancel       -> mod_mod_cancel
  4041 zmod_zmult_self1       -> mod_mult_self2_is_0
  4042 zmod_zmult_self2       -> mod_mult_self1_is_0
  4043 zmod_1                 -> mod_by_1
  4044 zdiv_1                 -> div_by_1
  4045 zdvd_abs1              -> abs_dvd_iff
  4046 zdvd_abs2              -> dvd_abs_iff
  4047 zdvd_refl              -> dvd_refl
  4048 zdvd_trans             -> dvd_trans
  4049 zdvd_zadd              -> dvd_add
  4050 zdvd_zdiff             -> dvd_diff
  4051 zdvd_zminus_iff        -> dvd_minus_iff
  4052 zdvd_zminus2_iff       -> minus_dvd_iff
  4053 zdvd_zmultD            -> dvd_mult_right
  4054 zdvd_zmultD2           -> dvd_mult_left
  4055 zdvd_zmult_mono        -> mult_dvd_mono
  4056 zdvd_0_right           -> dvd_0_right
  4057 zdvd_0_left            -> dvd_0_left_iff
  4058 zdvd_1_left            -> one_dvd
  4059 zminus_dvd_iff         -> minus_dvd_iff
  4060 
  4061 * Theory Rational: 'Fract k 0' now equals '0'.  INCOMPATIBILITY.
  4062 
  4063 * The real numbers offer decimal input syntax: 12.34 is translated
  4064 into 1234/10^2. This translation is not reversed upon output.
  4065 
  4066 * Theory Library/Polynomial defines an abstract type 'a poly of
  4067 univariate polynomials with coefficients of type 'a.  In addition to
  4068 the standard ring operations, it also supports div and mod.  Code
  4069 generation is also supported, using list-style constructors.
  4070 
  4071 * Theory Library/Inner_Product defines a class of real_inner for real
  4072 inner product spaces, with an overloaded operation inner :: 'a => 'a
  4073 => real.  Class real_inner is a subclass of real_normed_vector from
  4074 theory RealVector.
  4075 
  4076 * Theory Library/Product_Vector provides instances for the product
  4077 type 'a * 'b of several classes from RealVector and Inner_Product.
  4078 Definitions of addition, subtraction, scalar multiplication, norms,
  4079 and inner products are included.
  4080 
  4081 * Theory Library/Bit defines the field "bit" of integers modulo 2.  In
  4082 addition to the field operations, numerals and case syntax are also
  4083 supported.
  4084 
  4085 * Theory Library/Diagonalize provides constructive version of Cantor's
  4086 first diagonalization argument.
  4087 
  4088 * Theory Library/GCD: Curried operations gcd, lcm (for nat) and zgcd,
  4089 zlcm (for int); carried together from various gcd/lcm developements in
  4090 the HOL Distribution.  Constants zgcd and zlcm replace former igcd and
  4091 ilcm; corresponding theorems renamed accordingly.  INCOMPATIBILITY,
  4092 may recover tupled syntax as follows:
  4093 
  4094     hide (open) const gcd
  4095     abbreviation gcd where
  4096       "gcd == (%(a, b). GCD.gcd a b)"
  4097     notation (output)
  4098       GCD.gcd ("gcd '(_, _')")
  4099 
  4100 The same works for lcm, zgcd, zlcm.
  4101 
  4102 * Theory Library/Nat_Infinity: added addition, numeral syntax and more
  4103 instantiations for algebraic structures.  Removed some duplicate
  4104 theorems.  Changes in simp rules.  INCOMPATIBILITY.
  4105 
  4106 * ML antiquotation @{code} takes a constant as argument and generates
  4107 corresponding code in background and inserts name of the corresponding
  4108 resulting ML value/function/datatype constructor binding in place.
  4109 All occurrences of @{code} with a single ML block are generated
  4110 simultaneously.  Provides a generic and safe interface for
  4111 instrumentalizing code generation.  See
  4112 src/HOL/Decision_Procs/Ferrack.thy for a more ambitious application.
  4113 In future you ought to refrain from ad-hoc compiling generated SML
  4114 code on the ML toplevel.  Note that (for technical reasons) @{code}
  4115 cannot refer to constants for which user-defined serializations are
  4116 set.  Refer to the corresponding ML counterpart directly in that
  4117 cases.
  4118 
  4119 * Command 'rep_datatype': instead of theorem names the command now
  4120 takes a list of terms denoting the constructors of the type to be
  4121 represented as datatype.  The characteristic theorems have to be
  4122 proven.  INCOMPATIBILITY.  Also observe that the following theorems
  4123 have disappeared in favour of existing ones:
  4124 
  4125     unit_induct                 ~> unit.induct
  4126     prod_induct                 ~> prod.induct
  4127     sum_induct                  ~> sum.induct
  4128     Suc_Suc_eq                  ~> nat.inject
  4129     Suc_not_Zero Zero_not_Suc   ~> nat.distinct
  4130 
  4131 
  4132 *** HOL-Algebra ***
  4133 
  4134 * New locales for orders and lattices where the equivalence relation
  4135 is not restricted to equality.  INCOMPATIBILITY: all order and lattice
  4136 locales use a record structure with field eq for the equivalence.
  4137 
  4138 * New theory of factorial domains.
  4139 
  4140 * Units_l_inv and Units_r_inv are now simp rules by default.
  4141 INCOMPATIBILITY.  Simplifier proof that require deletion of l_inv
  4142 and/or r_inv will now also require deletion of these lemmas.
  4143 
  4144 * Renamed the following theorems, INCOMPATIBILITY:
  4145 
  4146 UpperD ~> Upper_memD
  4147 LowerD ~> Lower_memD
  4148 least_carrier ~> least_closed
  4149 greatest_carrier ~> greatest_closed
  4150 greatest_Lower_above ~> greatest_Lower_below
  4151 one_zero ~> carrier_one_zero
  4152 one_not_zero ~> carrier_one_not_zero  (collision with assumption)
  4153 
  4154 
  4155 *** HOL-Nominal ***
  4156 
  4157 * Nominal datatypes can now contain type-variables.
  4158 
  4159 * Commands 'nominal_inductive' and 'equivariance' work with local
  4160 theory targets.
  4161 
  4162 * Nominal primrec can now works with local theory targets and its
  4163 specification syntax now conforms to the general format as seen in
  4164 'inductive' etc.
  4165 
  4166 * Method "perm_simp" honours the standard simplifier attributes
  4167 (no_asm), (no_asm_use) etc.
  4168 
  4169 * The new predicate #* is defined like freshness, except that on the
  4170 left hand side can be a set or list of atoms.
  4171 
  4172 * Experimental command 'nominal_inductive2' derives strong induction
  4173 principles for inductive definitions.  In contrast to
  4174 'nominal_inductive', which can only deal with a fixed number of
  4175 binders, it can deal with arbitrary expressions standing for sets of
  4176 atoms to be avoided.  The only inductive definition we have at the
  4177 moment that needs this generalisation is the typing rule for Lets in
  4178 the algorithm W:
  4179 
  4180  Gamma |- t1 : T1   (x,close Gamma T1)::Gamma |- t2 : T2   x#Gamma
  4181  -----------------------------------------------------------------
  4182          Gamma |- Let x be t1 in t2 : T2
  4183 
  4184 In this rule one wants to avoid all the binders that are introduced by
  4185 "close Gamma T1".  We are looking for other examples where this
  4186 feature might be useful.  Please let us know.
  4187 
  4188 
  4189 *** HOLCF ***
  4190 
  4191 * Reimplemented the simplification procedure for proving continuity
  4192 subgoals.  The new simproc is extensible; users can declare additional
  4193 continuity introduction rules with the attribute [cont2cont].
  4194 
  4195 * The continuity simproc now uses a different introduction rule for
  4196 solving continuity subgoals on terms with lambda abstractions.  In
  4197 some rare cases the new simproc may fail to solve subgoals that the
  4198 old one could solve, and "simp add: cont2cont_LAM" may be necessary.
  4199 Potential INCOMPATIBILITY.
  4200 
  4201 * Command 'fixrec': specification syntax now conforms to the general
  4202 format as seen in 'inductive' etc.  See src/HOLCF/ex/Fixrec_ex.thy for
  4203 examples.  INCOMPATIBILITY.
  4204 
  4205 
  4206 *** ZF ***
  4207 
  4208 * Proof of Zorn's Lemma for partial orders.
  4209 
  4210 
  4211 *** ML ***
  4212 
  4213 * Multithreading for Poly/ML 5.1/5.2 is no longer supported, only for
  4214 Poly/ML 5.2.1 or later.  Important note: the TimeLimit facility
  4215 depends on multithreading, so timouts will not work before Poly/ML
  4216 5.2.1!
  4217 
  4218 * High-level support for concurrent ML programming, see
  4219 src/Pure/Cuncurrent.  The data-oriented model of "future values" is
  4220 particularly convenient to organize independent functional
  4221 computations.  The concept of "synchronized variables" provides a
  4222 higher-order interface for components with shared state, avoiding the
  4223 delicate details of mutexes and condition variables.  (Requires
  4224 Poly/ML 5.2.1 or later.)
  4225 
  4226 * ML bindings produced via Isar commands are stored within the Isar
  4227 context (theory or proof).  Consequently, commands like 'use' and 'ML'
  4228 become thread-safe and work with undo as expected (concerning
  4229 top-level bindings, not side-effects on global references).
  4230 INCOMPATIBILITY, need to provide proper Isar context when invoking the
  4231 compiler at runtime; really global bindings need to be given outside a
  4232 theory.  (Requires Poly/ML 5.2 or later.)
  4233 
  4234 * Command 'ML_prf' is analogous to 'ML' but works within a proof
  4235 context.  Top-level ML bindings are stored within the proof context in
  4236 a purely sequential fashion, disregarding the nested proof structure.
  4237 ML bindings introduced by 'ML_prf' are discarded at the end of the
  4238 proof.  (Requires Poly/ML 5.2 or later.)
  4239 
  4240 * Simplified ML attribute and method setup, cf. functions Attrib.setup
  4241 and Method.setup, as well as Isar commands 'attribute_setup' and
  4242 'method_setup'.  INCOMPATIBILITY for 'method_setup', need to simplify
  4243 existing code accordingly, or use plain 'setup' together with old
  4244 Method.add_method.
  4245 
  4246 * Simplified ML oracle interface Thm.add_oracle promotes 'a -> cterm
  4247 to 'a -> thm, while results are always tagged with an authentic oracle
  4248 name.  The Isar command 'oracle' is now polymorphic, no argument type
  4249 is specified.  INCOMPATIBILITY, need to simplify existing oracle code
  4250 accordingly.  Note that extra performance may be gained by producing
  4251 the cterm carefully, avoiding slow Thm.cterm_of.
  4252 
  4253 * Simplified interface for defining document antiquotations via
  4254 ThyOutput.antiquotation, ThyOutput.output, and optionally
  4255 ThyOutput.maybe_pretty_source.  INCOMPATIBILITY, need to simplify user
  4256 antiquotations accordingly, see src/Pure/Thy/thy_output.ML for common
  4257 examples.
  4258 
  4259 * More systematic treatment of long names, abstract name bindings, and
  4260 name space operations.  Basic operations on qualified names have been
  4261 move from structure NameSpace to Long_Name, e.g. Long_Name.base_name,
  4262 Long_Name.append.  Old type bstring has been mostly replaced by
  4263 abstract type binding (see structure Binding), which supports precise
  4264 qualification by packages and local theory targets, as well as proper
  4265 tracking of source positions.  INCOMPATIBILITY, need to wrap old
  4266 bstring values into Binding.name, or better pass through abstract
  4267 bindings everywhere.  See further src/Pure/General/long_name.ML,
  4268 src/Pure/General/binding.ML and src/Pure/General/name_space.ML
  4269 
  4270 * Result facts (from PureThy.note_thms, ProofContext.note_thms,
  4271 LocalTheory.note etc.) now refer to the *full* internal name, not the
  4272 bstring as before.  INCOMPATIBILITY, not detected by ML type-checking!
  4273 
  4274 * Disposed old type and term read functions (Sign.read_def_typ,
  4275 Sign.read_typ, Sign.read_def_terms, Sign.read_term,
  4276 Thm.read_def_cterms, Thm.read_cterm etc.).  INCOMPATIBILITY, should
  4277 use regular Syntax.read_typ, Syntax.read_term, Syntax.read_typ_global,
  4278 Syntax.read_term_global etc.; see also OldGoals.read_term as last
  4279 resort for legacy applications.
  4280 
  4281 * Disposed old declarations, tactics, tactic combinators that refer to
  4282 the simpset or claset of an implicit theory (such as Addsimps,
  4283 Simp_tac, SIMPSET).  INCOMPATIBILITY, should use @{simpset} etc. in
  4284 embedded ML text, or local_simpset_of with a proper context passed as
  4285 explicit runtime argument.
  4286 
  4287 * Rules and tactics that read instantiations (read_instantiate,
  4288 res_inst_tac, thin_tac, subgoal_tac etc.) now demand a proper proof
  4289 context, which is required for parsing and type-checking.  Moreover,
  4290 the variables are specified as plain indexnames, not string encodings
  4291 thereof.  INCOMPATIBILITY.
  4292 
  4293 * Generic Toplevel.add_hook interface allows to analyze the result of
  4294 transactions.  E.g. see src/Pure/ProofGeneral/proof_general_pgip.ML
  4295 for theorem dependency output of transactions resulting in a new
  4296 theory state.
  4297 
  4298 * ML antiquotations: block-structured compilation context indicated by
  4299 \<lbrace> ... \<rbrace>; additional antiquotation forms:
  4300 
  4301   @{binding name}                         - basic name binding
  4302   @{let ?pat = term}                      - term abbreviation (HO matching)
  4303   @{note name = fact}                     - fact abbreviation
  4304   @{thm fact}                             - singleton fact (with attributes)
  4305   @{thms fact}                            - general fact (with attributes)
  4306   @{lemma prop by method}                 - singleton goal
  4307   @{lemma prop by meth1 meth2}            - singleton goal
  4308   @{lemma prop1 ... propN by method}      - general goal
  4309   @{lemma prop1 ... propN by meth1 meth2} - general goal
  4310   @{lemma (open) ...}                     - open derivation
  4311 
  4312 
  4313 *** System ***
  4314 
  4315 * The Isabelle "emacs" tool provides a specific interface to invoke
  4316 Proof General / Emacs, with more explicit failure if that is not
  4317 installed (the old isabelle-interface script silently falls back on
  4318 isabelle-process).  The PROOFGENERAL_HOME setting determines the
  4319 installation location of the Proof General distribution.
  4320 
  4321 * Isabelle/lib/classes/Pure.jar provides basic support to integrate
  4322 the Isabelle process into a JVM/Scala application.  See
  4323 Isabelle/lib/jedit/plugin for a minimal example.  (The obsolete Java
  4324 process wrapper has been discontinued.)
  4325 
  4326 * Added homegrown Isabelle font with unicode layout, see lib/fonts.
  4327 
  4328 * Various status messages (with exact source position information) are
  4329 emitted, if proper markup print mode is enabled.  This allows
  4330 user-interface components to provide detailed feedback on internal
  4331 prover operations.
  4332 
  4333 
  4334 
  4335 New in Isabelle2008 (June 2008)
  4336 -------------------------------
  4337 
  4338 *** General ***
  4339 
  4340 * The Isabelle/Isar Reference Manual (isar-ref) has been reorganized
  4341 and updated, with formally checked references as hyperlinks.
  4342 
  4343 * Theory loader: use_thy (and similar operations) no longer set the
  4344 implicit ML context, which was occasionally hard to predict and in
  4345 conflict with concurrency.  INCOMPATIBILITY, use ML within Isar which
  4346 provides a proper context already.
  4347 
  4348 * Theory loader: old-style ML proof scripts being *attached* to a thy
  4349 file are no longer supported.  INCOMPATIBILITY, regular 'uses' and
  4350 'use' within a theory file will do the job.
  4351 
  4352 * Name space merge now observes canonical order, i.e. the second space
  4353 is inserted into the first one, while existing entries in the first
  4354 space take precedence.  INCOMPATIBILITY in rare situations, may try to
  4355 swap theory imports.
  4356 
  4357 * Syntax: symbol \<chi> is now considered a letter.  Potential
  4358 INCOMPATIBILITY in identifier syntax etc.
  4359 
  4360 * Outer syntax: string tokens no longer admit escaped white space,
  4361 which was an accidental (undocumented) feature.  INCOMPATIBILITY, use
  4362 white space without escapes.
  4363 
  4364 * Outer syntax: string tokens may contain arbitrary character codes
  4365 specified via 3 decimal digits (as in SML).  E.g. "foo\095bar" for
  4366 "foo_bar".
  4367 
  4368 
  4369 *** Pure ***
  4370 
  4371 * Context-dependent token translations.  Default setup reverts locally
  4372 fixed variables, and adds hilite markup for undeclared frees.
  4373 
  4374 * Unused theorems can be found using the new command 'unused_thms'.
  4375 There are three ways of invoking it:
  4376 
  4377 (1) unused_thms
  4378      Only finds unused theorems in the current theory.
  4379 
  4380 (2) unused_thms thy_1 ... thy_n -
  4381      Finds unused theorems in the current theory and all of its ancestors,
  4382      excluding the theories thy_1 ... thy_n and all of their ancestors.
  4383 
  4384 (3) unused_thms thy_1 ... thy_n - thy'_1 ... thy'_m
  4385      Finds unused theorems in the theories thy'_1 ... thy'_m and all of
  4386      their ancestors, excluding the theories thy_1 ... thy_n and all of
  4387      their ancestors.
  4388 
  4389 In order to increase the readability of the list produced by
  4390 unused_thms, theorems that have been created by a particular instance
  4391 of a theory command such as 'inductive' or 'function' are considered
  4392 to belong to the same "group", meaning that if at least one theorem in
  4393 this group is used, the other theorems in the same group are no longer
  4394 reported as unused.  Moreover, if all theorems in the group are
  4395 unused, only one theorem in the group is displayed.
  4396 
  4397 Note that proof objects have to be switched on in order for
  4398 unused_thms to work properly (i.e. !proofs must be >= 1, which is
  4399 usually the case when using Proof General with the default settings).
  4400 
  4401 * Authentic naming of facts disallows ad-hoc overwriting of previous
  4402 theorems within the same name space.  INCOMPATIBILITY, need to remove
  4403 duplicate fact bindings, or even accidental fact duplications.  Note
  4404 that tools may maintain dynamically scoped facts systematically, using
  4405 PureThy.add_thms_dynamic.
  4406 
  4407 * Command 'hide' now allows to hide from "fact" name space as well.
  4408 
  4409 * Eliminated destructive theorem database, simpset, claset, and
  4410 clasimpset.  Potential INCOMPATIBILITY, really need to observe linear
  4411 update of theories within ML code.
  4412 
  4413 * Eliminated theory ProtoPure and CPure, leaving just one Pure theory.
  4414 INCOMPATIBILITY, object-logics depending on former Pure require
  4415 additional setup PureThy.old_appl_syntax_setup; object-logics
  4416 depending on former CPure need to refer to Pure.
  4417 
  4418 * Commands 'use' and 'ML' are now purely functional, operating on
  4419 theory/local_theory.  Removed former 'ML_setup' (on theory), use 'ML'
  4420 instead.  Added 'ML_val' as mere diagnostic replacement for 'ML'.
  4421 INCOMPATIBILITY.
  4422 
  4423 * Command 'setup': discontinued implicit version with ML reference.
  4424 
  4425 * Instantiation target allows for simultaneous specification of class
  4426 instance operations together with an instantiation proof.
  4427 Type-checking phase allows to refer to class operations uniformly.
  4428 See src/HOL/Complex/Complex.thy for an Isar example and
  4429 src/HOL/Library/Eval.thy for an ML example.
  4430 
  4431 * Indexing of literal facts: be more serious about including only
  4432 facts from the visible specification/proof context, but not the
  4433 background context (locale etc.).  Affects `prop` notation and method
  4434 "fact".  INCOMPATIBILITY: need to name facts explicitly in rare
  4435 situations.
  4436 
  4437 * Method "cases", "induct", "coinduct": removed obsolete/undocumented
  4438 "(open)" option, which used to expose internal bound variables to the
  4439 proof text.
  4440 
  4441 * Isar statements: removed obsolete case "rule_context".
  4442 INCOMPATIBILITY, better use explicit fixes/assumes.
  4443 
  4444 * Locale proofs: default proof step now includes 'unfold_locales';
  4445 hence 'proof' without argument may be used to unfold locale
  4446 predicates.
  4447 
  4448 
  4449 *** Document preparation ***
  4450 
  4451 * Simplified pdfsetup.sty: color/hyperref is used unconditionally for
  4452 both pdf and dvi (hyperlinks usually work in xdvi as well); removed
  4453 obsolete thumbpdf setup (contemporary PDF viewers do this on the
  4454 spot); renamed link color from "darkblue" to "linkcolor" (default
  4455 value unchanged, can be redefined via \definecolor); no longer sets
  4456 "a4paper" option (unnecessary or even intrusive).
  4457 
  4458 * Antiquotation @{lemma A method} proves proposition A by the given
  4459 method (either a method name or a method name plus (optional) method
  4460 arguments in parentheses) and prints A just like @{prop A}.
  4461 
  4462 
  4463 *** HOL ***
  4464 
  4465 * New primrec package.  Specification syntax conforms in style to
  4466 definition/function/....  No separate induction rule is provided.  The
  4467 "primrec" command distinguishes old-style and new-style specifications
  4468 by syntax.  The former primrec package is now named OldPrimrecPackage.
  4469 When adjusting theories, beware: constants stemming from new-style
  4470 primrec specifications have authentic syntax.
  4471 
  4472 * Metis prover is now an order of magnitude faster, and also works
  4473 with multithreading.
  4474 
  4475 * Metis: the maximum number of clauses that can be produced from a
  4476 theorem is now given by the attribute max_clauses.  Theorems that
  4477 exceed this number are ignored, with a warning printed.
  4478 
  4479 * Sledgehammer no longer produces structured proofs by default. To
  4480 enable, declare [[sledgehammer_full = true]].  Attributes
  4481 reconstruction_modulus, reconstruction_sorts renamed
  4482 sledgehammer_modulus, sledgehammer_sorts.  INCOMPATIBILITY.
  4483 
  4484 * Method "induct_scheme" derives user-specified induction rules
  4485 from well-founded induction and completeness of patterns. This factors
  4486 out some operations that are done internally by the function package
  4487 and makes them available separately.  See
  4488 src/HOL/ex/Induction_Scheme.thy for examples.
  4489 
  4490 * More flexible generation of measure functions for termination
  4491 proofs: Measure functions can be declared by proving a rule of the
  4492 form "is_measure f" and giving it the [measure_function] attribute.
  4493 The "is_measure" predicate is logically meaningless (always true), and
  4494 just guides the heuristic.  To find suitable measure functions, the
  4495 termination prover sets up the goal "is_measure ?f" of the appropriate
  4496 type and generates all solutions by prolog-style backwards proof using
  4497 the declared rules.
  4498 
  4499 This setup also deals with rules like 
  4500 
  4501   "is_measure f ==> is_measure (list_size f)"
  4502 
  4503 which accommodates nested datatypes that recurse through lists.
  4504 Similar rules are predeclared for products and option types.
  4505 
  4506 * Turned the type of sets "'a set" into an abbreviation for "'a => bool"
  4507 
  4508   INCOMPATIBILITIES:
  4509 
  4510   - Definitions of overloaded constants on sets have to be replaced by
  4511     definitions on => and bool.
  4512 
  4513   - Some definitions of overloaded operators on sets can now be proved
  4514     using the definitions of the operators on => and bool.  Therefore,
  4515     the following theorems have been renamed:
  4516 
  4517       subset_def   -> subset_eq
  4518       psubset_def  -> psubset_eq
  4519       set_diff_def -> set_diff_eq
  4520       Compl_def    -> Compl_eq
  4521       Sup_set_def  -> Sup_set_eq
  4522       Inf_set_def  -> Inf_set_eq
  4523       sup_set_def  -> sup_set_eq
  4524       inf_set_def  -> inf_set_eq
  4525 
  4526   - Due to the incompleteness of the HO unification algorithm, some
  4527     rules such as subst may require manual instantiation, if some of
  4528     the unknowns in the rule is a set.
  4529 
  4530   - Higher order unification and forward proofs:
  4531     The proof pattern
  4532 
  4533       have "P (S::'a set)" <...>
  4534       then have "EX S. P S" ..
  4535 
  4536     no longer works (due to the incompleteness of the HO unification
  4537     algorithm) and must be replaced by the pattern
  4538 
  4539       have "EX S. P S"
  4540       proof
  4541         show "P S" <...>
  4542       qed
  4543 
  4544   - Calculational reasoning with subst (or similar rules):
  4545     The proof pattern
  4546 
  4547       have "P (S::'a set)" <...>
  4548       also have "S = T" <...>
  4549       finally have "P T" .
  4550 
  4551     no longer works (for similar reasons as the previous example) and
  4552     must be replaced by something like
  4553 
  4554       have "P (S::'a set)" <...>
  4555       moreover have "S = T" <...>
  4556       ultimately have "P T" by simp
  4557 
  4558   - Tactics or packages written in ML code:
  4559     Code performing pattern matching on types via
  4560 
  4561       Type ("set", [T]) => ...
  4562 
  4563     must be rewritten. Moreover, functions like strip_type or
  4564     binder_types no longer return the right value when applied to a
  4565     type of the form
  4566 
  4567       T1 => ... => Tn => U => bool
  4568 
  4569     rather than
  4570 
  4571       T1 => ... => Tn => U set
  4572 
  4573 * Merged theories Wellfounded_Recursion, Accessible_Part and
  4574 Wellfounded_Relations to theory Wellfounded.
  4575 
  4576 * Explicit class "eq" for executable equality.  INCOMPATIBILITY.
  4577 
  4578 * Class finite no longer treats UNIV as class parameter.  Use class
  4579 enum from theory Library/Enum instead to achieve a similar effect.
  4580 INCOMPATIBILITY.
  4581 
  4582 * Theory List: rule list_induct2 now has explicitly named cases "Nil"
  4583 and "Cons".  INCOMPATIBILITY.
  4584 
  4585 * HOL (and FOL): renamed variables in rules imp_elim and swap.
  4586 Potential INCOMPATIBILITY.
  4587 
  4588 * Theory Product_Type: duplicated lemmas split_Pair_apply and
  4589 injective_fst_snd removed, use split_eta and prod_eqI instead.
  4590 Renamed upd_fst to apfst and upd_snd to apsnd.  INCOMPATIBILITY.
  4591 
  4592 * Theory Nat: removed redundant lemmas that merely duplicate lemmas of
  4593 the same name in theory Orderings:
  4594 
  4595   less_trans
  4596   less_linear
  4597   le_imp_less_or_eq
  4598   le_less_trans
  4599   less_le_trans
  4600   less_not_sym
  4601   less_asym
  4602 
  4603 Renamed less_imp_le to less_imp_le_nat, and less_irrefl to
  4604 less_irrefl_nat.  Potential INCOMPATIBILITY due to more general types
  4605 and different variable names.
  4606 
  4607 * Library/Option_ord.thy: Canonical order on option type.
  4608 
  4609 * Library/RBT.thy: Red-black trees, an efficient implementation of
  4610 finite maps.
  4611 
  4612 * Library/Countable.thy: Type class for countable types.
  4613 
  4614 * Theory Int: The representation of numerals has changed.  The infix
  4615 operator BIT and the bit datatype with constructors B0 and B1 have
  4616 disappeared.  INCOMPATIBILITY, use "Int.Bit0 x" and "Int.Bit1 y" in
  4617 place of "x BIT bit.B0" and "y BIT bit.B1", respectively.  Theorems
  4618 involving BIT, B0, or B1 have been renamed with "Bit0" or "Bit1"
  4619 accordingly.
  4620 
  4621 * Theory Nat: definition of <= and < on natural numbers no longer
  4622 depend on well-founded relations.  INCOMPATIBILITY.  Definitions
  4623 le_def and less_def have disappeared.  Consider lemmas not_less
  4624 [symmetric, where ?'a = nat] and less_eq [symmetric] instead.
  4625 
  4626 * Theory Finite_Set: locales ACf, ACe, ACIf, ACIfSL and ACIfSLlin
  4627 (whose purpose mainly is for various fold_set functionals) have been
  4628 abandoned in favor of the existing algebraic classes
  4629 ab_semigroup_mult, comm_monoid_mult, ab_semigroup_idem_mult,
  4630 lower_semilattice (resp. upper_semilattice) and linorder.
  4631 INCOMPATIBILITY.
  4632 
  4633 * Theory Transitive_Closure: induct and cases rules now declare proper
  4634 case_names ("base" and "step").  INCOMPATIBILITY.
  4635 
  4636 * Theorem Inductive.lfp_ordinal_induct generalized to complete
  4637 lattices.  The form set-specific version is available as
  4638 Inductive.lfp_ordinal_induct_set.
  4639 
  4640 * Renamed theorems "power.simps" to "power_int.simps".
  4641 INCOMPATIBILITY.
  4642 
  4643 * Class semiring_div provides basic abstract properties of semirings
  4644 with division and modulo operations.  Subsumes former class dvd_mod.
  4645 
  4646 * Merged theories IntDef, Numeral and IntArith into unified theory
  4647 Int.  INCOMPATIBILITY.
  4648 
  4649 * Theory Library/Code_Index: type "index" now represents natural
  4650 numbers rather than integers.  INCOMPATIBILITY.
  4651 
  4652 * New class "uminus" with operation "uminus" (split of from class
  4653 "minus" which now only has operation "minus", binary).
  4654 INCOMPATIBILITY.
  4655 
  4656 * Constants "card", "internal_split", "option_map" now with authentic
  4657 syntax.  INCOMPATIBILITY.
  4658 
  4659 * Definitions subset_def, psubset_def, set_diff_def, Compl_def,
  4660 le_bool_def, less_bool_def, le_fun_def, less_fun_def, inf_bool_def,
  4661 sup_bool_def, Inf_bool_def, Sup_bool_def, inf_fun_def, sup_fun_def,
  4662 Inf_fun_def, Sup_fun_def, inf_set_def, sup_set_def, Inf_set_def,
  4663 Sup_set_def, le_def, less_def, option_map_def now with object
  4664 equality.  INCOMPATIBILITY.
  4665 
  4666 * Records. Removed K_record, and replaced it by pure lambda term
  4667 %x. c. The simplifier setup is now more robust against eta expansion.
  4668 INCOMPATIBILITY: in cases explicitly referring to K_record.
  4669 
  4670 * Library/Multiset: {#a, b, c#} abbreviates {#a#} + {#b#} + {#c#}.
  4671 
  4672 * Library/ListVector: new theory of arithmetic vector operations.
  4673 
  4674 * Library/Order_Relation: new theory of various orderings as sets of
  4675 pairs.  Defines preorders, partial orders, linear orders and
  4676 well-orders on sets and on types.
  4677 
  4678 
  4679 *** ZF ***
  4680 
  4681 * Renamed some theories to allow to loading both ZF and HOL in the
  4682 same session:
  4683 
  4684   Datatype  -> Datatype_ZF
  4685   Inductive -> Inductive_ZF
  4686   Int       -> Int_ZF
  4687   IntDiv    -> IntDiv_ZF
  4688   Nat       -> Nat_ZF
  4689   List      -> List_ZF
  4690   Main      -> Main_ZF
  4691 
  4692 INCOMPATIBILITY: ZF theories that import individual theories below
  4693 Main might need to be adapted.  Regular theory Main is still
  4694 available, as trivial extension of Main_ZF.
  4695 
  4696 
  4697 *** ML ***
  4698 
  4699 * ML within Isar: antiquotation @{const name} or @{const
  4700 name(typargs)} produces statically-checked Const term.
  4701 
  4702 * Functor NamedThmsFun: data is available to the user as dynamic fact
  4703 (of the same name).  Removed obsolete print command.
  4704 
  4705 * Removed obsolete "use_legacy_bindings" function.
  4706 
  4707 * The ``print mode'' is now a thread-local value derived from a global
  4708 template (the former print_mode reference), thus access becomes
  4709 non-critical.  The global print_mode reference is for session
  4710 management only; user-code should use print_mode_value,
  4711 print_mode_active, PrintMode.setmp etc.  INCOMPATIBILITY.
  4712 
  4713 * Functions system/system_out provide a robust way to invoke external
  4714 shell commands, with propagation of interrupts (requires Poly/ML
  4715 5.2.1).  Do not use OS.Process.system etc. from the basis library!
  4716 
  4717 
  4718 *** System ***
  4719 
  4720 * Default settings: PROOFGENERAL_OPTIONS no longer impose xemacs ---
  4721 in accordance with Proof General 3.7, which prefers GNU emacs.
  4722 
  4723 * isatool tty runs Isabelle process with plain tty interaction;
  4724 optional line editor may be specified via ISABELLE_LINE_EDITOR
  4725 setting, the default settings attempt to locate "ledit" and "rlwrap".
  4726 
  4727 * isatool browser now works with Cygwin as well, using general
  4728 "javapath" function defined in Isabelle process environment.
  4729 
  4730 * YXML notation provides a simple and efficient alternative to
  4731 standard XML transfer syntax.  See src/Pure/General/yxml.ML and
  4732 isatool yxml as described in the Isabelle system manual.
  4733 
  4734 * JVM class isabelle.IsabelleProcess (located in Isabelle/lib/classes)
  4735 provides general wrapper for managing an Isabelle process in a robust
  4736 fashion, with ``cooked'' output from stdin/stderr.
  4737 
  4738 * Rudimentary Isabelle plugin for jEdit (see Isabelle/lib/jedit),
  4739 based on Isabelle/JVM process wrapper (see Isabelle/lib/classes).
  4740 
  4741 * Removed obsolete THIS_IS_ISABELLE_BUILD feature.  NB: the documented
  4742 way of changing the user's settings is via
  4743 ISABELLE_HOME_USER/etc/settings, which is a fully featured bash
  4744 script.
  4745 
  4746 * Multithreading.max_threads := 0 refers to the number of actual CPU
  4747 cores of the underlying machine, which is a good starting point for
  4748 optimal performance tuning.  The corresponding usedir option -M allows
  4749 "max" as an alias for "0".  WARNING: does not work on certain versions
  4750 of Mac OS (with Poly/ML 5.1).
  4751 
  4752 * isabelle-process: non-ML sessions are run with "nice", to reduce the
  4753 adverse effect of Isabelle flooding interactive front-ends (notably
  4754 ProofGeneral / XEmacs).
  4755 
  4756 
  4757 
  4758 New in Isabelle2007 (November 2007)
  4759 -----------------------------------
  4760 
  4761 *** General ***
  4762 
  4763 * More uniform information about legacy features, notably a
  4764 warning/error of "Legacy feature: ...", depending on the state of the
  4765 tolerate_legacy_features flag (default true). FUTURE INCOMPATIBILITY:
  4766 legacy features will disappear eventually.
  4767 
  4768 * Theory syntax: the header format ``theory A = B + C:'' has been
  4769 discontinued in favour of ``theory A imports B C begin''.  Use isatool
  4770 fixheaders to convert existing theory files.  INCOMPATIBILITY.
  4771 
  4772 * Theory syntax: the old non-Isar theory file format has been
  4773 discontinued altogether.  Note that ML proof scripts may still be used
  4774 with Isar theories; migration is usually quite simple with the ML
  4775 function use_legacy_bindings.  INCOMPATIBILITY.
  4776 
  4777 * Theory syntax: some popular names (e.g. 'class', 'declaration',
  4778 'fun', 'help', 'if') are now keywords.  INCOMPATIBILITY, use double
  4779 quotes.
  4780 
  4781 * Theory loader: be more serious about observing the static theory
  4782 header specifications (including optional directories), but not the
  4783 accidental file locations of previously successful loads.  The strict
  4784 update policy of former update_thy is now already performed by
  4785 use_thy, so the former has been removed; use_thys updates several
  4786 theories simultaneously, just as 'imports' within a theory header
  4787 specification, but without merging the results.  Potential
  4788 INCOMPATIBILITY: may need to refine theory headers and commands
  4789 ROOT.ML which depend on load order.
  4790 
  4791 * Theory loader: optional support for content-based file
  4792 identification, instead of the traditional scheme of full physical
  4793 path plus date stamp; configured by the ISABELLE_FILE_IDENT setting
  4794 (cf. the system manual).  The new scheme allows to work with
  4795 non-finished theories in persistent session images, such that source
  4796 files may be moved later on without requiring reloads.
  4797 
  4798 * Theory loader: old-style ML proof scripts being *attached* to a thy
  4799 file (with the same base name as the theory) are considered a legacy
  4800 feature, which will disappear eventually. Even now, the theory loader
  4801 no longer maintains dependencies on such files.
  4802 
  4803 * Syntax: the scope for resolving ambiguities via type-inference is
  4804 now limited to individual terms, instead of whole simultaneous
  4805 specifications as before. This greatly reduces the complexity of the
  4806 syntax module and improves flexibility by separating parsing and
  4807 type-checking. INCOMPATIBILITY: additional type-constraints (explicit
  4808 'fixes' etc.) are required in rare situations.
  4809 
  4810 * Syntax: constants introduced by new-style packages ('definition',
  4811 'abbreviation' etc.) are passed through the syntax module in
  4812 ``authentic mode''. This means that associated mixfix annotations
  4813 really stick to such constants, independently of potential name space
  4814 ambiguities introduced later on. INCOMPATIBILITY: constants in parse
  4815 trees are represented slightly differently, may need to adapt syntax
  4816 translations accordingly. Use CONST marker in 'translations' and
  4817 @{const_syntax} antiquotation in 'parse_translation' etc.
  4818 
  4819 * Legacy goal package: reduced interface to the bare minimum required
  4820 to keep existing proof scripts running.  Most other user-level
  4821 functions are now part of the OldGoals structure, which is *not* open
  4822 by default (consider isatool expandshort before open OldGoals).
  4823 Removed top_sg, prin, printyp, pprint_term/typ altogether, because
  4824 these tend to cause confusion about the actual goal (!) context being
  4825 used here, which is not necessarily the same as the_context().
  4826 
  4827 * Command 'find_theorems': supports "*" wild-card in "name:"
  4828 criterion; "with_dups" option.  Certain ProofGeneral versions might
  4829 support a specific search form (see ProofGeneral/CHANGES).
  4830 
  4831 * The ``prems limit'' option (cf. ProofContext.prems_limit) is now -1
  4832 by default, which means that "prems" (and also "fixed variables") are
  4833 suppressed from proof state output.  Note that the ProofGeneral
  4834 settings mechanism allows to change and save options persistently, but
  4835 older versions of Isabelle will fail to start up if a negative prems
  4836 limit is imposed.
  4837 
  4838 * Local theory targets may be specified by non-nested blocks of
  4839 ``context/locale/class ... begin'' followed by ``end''.  The body may
  4840 contain definitions, theorems etc., including any derived mechanism
  4841 that has been implemented on top of these primitives.  This concept
  4842 generalizes the existing ``theorem (in ...)'' towards more versatility
  4843 and scalability.
  4844 
  4845 * Proof General interface: proper undo of final 'end' command;
  4846 discontinued Isabelle/classic mode (ML proof scripts).
  4847 
  4848 
  4849 *** Document preparation ***
  4850 
  4851 * Added antiquotation @{theory name} which prints the given name,
  4852 after checking that it refers to a valid ancestor theory in the
  4853 current context.
  4854 
  4855 * Added antiquotations @{ML_type text} and @{ML_struct text} which
  4856 check the given source text as ML type/structure, printing verbatim.
  4857 
  4858 * Added antiquotation @{abbrev "c args"} which prints the abbreviation
  4859 "c args == rhs" given in the current context.  (Any number of
  4860 arguments may be given on the LHS.)
  4861 
  4862 
  4863 *** Pure ***
  4864 
  4865 * The 'class' package offers a combination of axclass and locale to
  4866 achieve Haskell-like type classes in Isabelle.  Definitions and
  4867 theorems within a class context produce both relative results (with
  4868 implicit parameters according to the locale context), and polymorphic
  4869 constants with qualified polymorphism (according to the class
  4870 context).  Within the body context of a 'class' target, a separate
  4871 syntax layer ("user space type system") takes care of converting
  4872 between global polymorphic consts and internal locale representation.
  4873 See src/HOL/ex/Classpackage.thy for examples (as well as main HOL).
  4874 "isatool doc classes" provides a tutorial.
  4875 
  4876 * Generic code generator framework allows to generate executable
  4877 code for ML and Haskell (including Isabelle classes).  A short usage
  4878 sketch:
  4879 
  4880     internal compilation:
  4881         export_code <list of constants (term syntax)> in SML
  4882     writing SML code to a file:
  4883         export_code <list of constants (term syntax)> in SML <filename>
  4884     writing OCaml code to a file:
  4885         export_code <list of constants (term syntax)> in OCaml <filename>
  4886     writing Haskell code to a bunch of files:
  4887         export_code <list of constants (term syntax)> in Haskell <filename>
  4888 
  4889     evaluating closed propositions to True/False using code generation:
  4890         method ``eval''
  4891 
  4892 Reasonable default setup of framework in HOL.
  4893 
  4894 Theorem attributs for selecting and transforming function equations theorems:
  4895 
  4896     [code fun]:        select a theorem as function equation for a specific constant
  4897     [code fun del]:    deselect a theorem as function equation for a specific constant
  4898     [code inline]:     select an equation theorem for unfolding (inlining) in place
  4899     [code inline del]: deselect an equation theorem for unfolding (inlining) in place
  4900 
  4901 User-defined serializations (target in {SML, OCaml, Haskell}):
  4902 
  4903     code_const <and-list of constants (term syntax)>
  4904       {(target) <and-list of const target syntax>}+
  4905 
  4906     code_type <and-list of type constructors>
  4907       {(target) <and-list of type target syntax>}+
  4908 
  4909     code_instance <and-list of instances>
  4910       {(target)}+
  4911         where instance ::= <type constructor> :: <class>
  4912 
  4913     code_class <and_list of classes>
  4914       {(target) <and-list of class target syntax>}+
  4915         where class target syntax ::= <class name> {where {<classop> == <target syntax>}+}?
  4916 
  4917 code_instance and code_class only are effective to target Haskell.
  4918 
  4919 For example usage see src/HOL/ex/Codegenerator.thy and
  4920 src/HOL/ex/Codegenerator_Pretty.thy.  A separate tutorial on code
  4921 generation from Isabelle/HOL theories is available via "isatool doc
  4922 codegen".
  4923 
  4924 * Code generator: consts in 'consts_code' Isar commands are now
  4925 referred to by usual term syntax (including optional type
  4926 annotations).
  4927 
  4928 * Command 'no_translations' removes translation rules from theory
  4929 syntax.
  4930 
  4931 * Overloaded definitions are now actually checked for acyclic
  4932 dependencies.  The overloading scheme is slightly more general than
  4933 that of Haskell98, although Isabelle does not demand an exact
  4934 correspondence to type class and instance declarations.
  4935 INCOMPATIBILITY, use ``defs (unchecked overloaded)'' to admit more
  4936 exotic versions of overloading -- at the discretion of the user!
  4937 
  4938 Polymorphic constants are represented via type arguments, i.e. the
  4939 instantiation that matches an instance against the most general
  4940 declaration given in the signature.  For example, with the declaration
  4941 c :: 'a => 'a => 'a, an instance c :: nat => nat => nat is represented
  4942 as c(nat).  Overloading is essentially simultaneous structural
  4943 recursion over such type arguments.  Incomplete specification patterns
  4944 impose global constraints on all occurrences, e.g. c('a * 'a) on the
  4945 LHS means that more general c('a * 'b) will be disallowed on any RHS.
  4946 Command 'print_theory' outputs the normalized system of recursive
  4947 equations, see section "definitions".
  4948 
  4949 * Configuration options are maintained within the theory or proof
  4950 context (with name and type bool/int/string), providing a very simple
  4951 interface to a poor-man's version of general context data.  Tools may
  4952 declare options in ML (e.g. using Attrib.config_int) and then refer to
  4953 these values using Config.get etc.  Users may change options via an
  4954 associated attribute of the same name.  This form of context
  4955 declaration works particularly well with commands 'declare' or
  4956 'using', for example ``declare [[foo = 42]]''.  Thus it has become
  4957 very easy to avoid global references, which would not observe Isar
  4958 toplevel undo/redo and fail to work with multithreading.
  4959 
  4960 Various global ML references of Pure and HOL have been turned into
  4961 configuration options:
  4962 
  4963   Unify.search_bound		unify_search_bound
  4964   Unify.trace_bound		unify_trace_bound
  4965   Unify.trace_simp		unify_trace_simp
  4966   Unify.trace_types		unify_trace_types
  4967   Simplifier.simp_depth_limit	simp_depth_limit
  4968   Blast.depth_limit		blast_depth_limit
  4969   DatatypeProp.dtK		datatype_distinctness_limit
  4970   fast_arith_neq_limit  	fast_arith_neq_limit
  4971   fast_arith_split_limit	fast_arith_split_limit
  4972 
  4973 * Named collections of theorems may be easily installed as context
  4974 data using the functor NamedThmsFun (see also
  4975 src/Pure/Tools/named_thms.ML).  The user may add or delete facts via
  4976 attributes; there is also a toplevel print command.  This facility is
  4977 just a common case of general context data, which is the preferred way
  4978 for anything more complex than just a list of facts in canonical
  4979 order.
  4980 
  4981 * Isar: command 'declaration' augments a local theory by generic
  4982 declaration functions written in ML.  This enables arbitrary content
  4983 being added to the context, depending on a morphism that tells the
  4984 difference of the original declaration context wrt. the application
  4985 context encountered later on.
  4986 
  4987 * Isar: proper interfaces for simplification procedures.  Command
  4988 'simproc_setup' declares named simprocs (with match patterns, and body
  4989 text in ML).  Attribute "simproc" adds/deletes simprocs in the current
  4990 context.  ML antiquotation @{simproc name} retrieves named simprocs.
  4991 
  4992 * Isar: an extra pair of brackets around attribute declarations
  4993 abbreviates a theorem reference involving an internal dummy fact,
  4994 which will be ignored later --- only the effect of the attribute on
  4995 the background context will persist.  This form of in-place
  4996 declarations is particularly useful with commands like 'declare' and
  4997 'using', for example ``have A using [[simproc a]] by simp''.
  4998 
  4999 * Isar: method "assumption" (and implicit closing of subproofs) now
  5000 takes simple non-atomic goal assumptions into account: after applying
  5001 an assumption as a rule the resulting subgoals are solved by atomic
  5002 assumption steps.  This is particularly useful to finish 'obtain'
  5003 goals, such as "!!x. (!!x. P x ==> thesis) ==> P x ==> thesis",
  5004 without referring to the original premise "!!x. P x ==> thesis" in the
  5005 Isar proof context.  POTENTIAL INCOMPATIBILITY: method "assumption" is
  5006 more permissive.
  5007 
  5008 * Isar: implicit use of prems from the Isar proof context is
  5009 considered a legacy feature.  Common applications like ``have A .''
  5010 may be replaced by ``have A by fact'' or ``note `A`''.  In general,
  5011 referencing facts explicitly here improves readability and
  5012 maintainability of proof texts.
  5013 
  5014 * Isar: improper proof element 'guess' is like 'obtain', but derives
  5015 the obtained context from the course of reasoning!  For example:
  5016 
  5017   assume "EX x y. A x & B y"   -- "any previous fact"
  5018   then guess x and y by clarify
  5019 
  5020 This technique is potentially adventurous, depending on the facts and
  5021 proof tools being involved here.
  5022 
  5023 * Isar: known facts from the proof context may be specified as literal
  5024 propositions, using ASCII back-quote syntax.  This works wherever
  5025 named facts used to be allowed so far, in proof commands, proof
  5026 methods, attributes etc.  Literal facts are retrieved from the context
  5027 according to unification of type and term parameters.  For example,
  5028 provided that "A" and "A ==> B" and "!!x. P x ==> Q x" are known
  5029 theorems in the current context, then these are valid literal facts:
  5030 `A` and `A ==> B` and `!!x. P x ==> Q x" as well as `P a ==> Q a` etc.
  5031 
  5032 There is also a proof method "fact" which does the same composition
  5033 for explicit goal states, e.g. the following proof texts coincide with
  5034 certain special cases of literal facts:
  5035 
  5036   have "A" by fact                 ==  note `A`
  5037   have "A ==> B" by fact           ==  note `A ==> B`
  5038   have "!!x. P x ==> Q x" by fact  ==  note `!!x. P x ==> Q x`
  5039   have "P a ==> Q a" by fact       ==  note `P a ==> Q a`
  5040 
  5041 * Isar: ":" (colon) is no longer a symbolic identifier character in
  5042 outer syntax.  Thus symbolic identifiers may be used without
  5043 additional white space in declarations like this: ``assume *: A''.
  5044 
  5045 * Isar: 'print_facts' prints all local facts of the current context,
  5046 both named and unnamed ones.
  5047 
  5048 * Isar: 'def' now admits simultaneous definitions, e.g.:
  5049 
  5050   def x == "t" and y == "u"
  5051 
  5052 * Isar: added command 'unfolding', which is structurally similar to
  5053 'using', but affects both the goal state and facts by unfolding given
  5054 rewrite rules.  Thus many occurrences of the 'unfold' method or
  5055 'unfolded' attribute may be replaced by first-class proof text.
  5056 
  5057 * Isar: methods 'unfold' / 'fold', attributes 'unfolded' / 'folded',
  5058 and command 'unfolding' now all support object-level equalities
  5059 (potentially conditional).  The underlying notion of rewrite rule is
  5060 analogous to the 'rule_format' attribute, but *not* that of the
  5061 Simplifier (which is usually more generous).
  5062 
  5063 * Isar: the new attribute [rotated n] (default n = 1) rotates the
  5064 premises of a theorem by n. Useful in conjunction with drule.
  5065 
  5066 * Isar: the goal restriction operator [N] (default N = 1) evaluates a
  5067 method expression within a sandbox consisting of the first N
  5068 sub-goals, which need to exist.  For example, ``simp_all [3]''
  5069 simplifies the first three sub-goals, while (rule foo, simp_all)[]
  5070 simplifies all new goals that emerge from applying rule foo to the
  5071 originally first one.
  5072 
  5073 * Isar: schematic goals are no longer restricted to higher-order
  5074 patterns; e.g. ``lemma "?P(?x)" by (rule TrueI)'' now works as
  5075 expected.
  5076 
  5077 * Isar: the conclusion of a long theorem statement is now either
  5078 'shows' (a simultaneous conjunction, as before), or 'obtains'
  5079 (essentially a disjunction of cases with local parameters and
  5080 assumptions).  The latter allows to express general elimination rules
  5081 adequately; in this notation common elimination rules look like this:
  5082 
  5083   lemma exE:    -- "EX x. P x ==> (!!x. P x ==> thesis) ==> thesis"
  5084     assumes "EX x. P x"
  5085     obtains x where "P x"
  5086 
  5087   lemma conjE:  -- "A & B ==> (A ==> B ==> thesis) ==> thesis"
  5088     assumes "A & B"
  5089     obtains A and B
  5090 
  5091   lemma disjE:  -- "A | B ==> (A ==> thesis) ==> (B ==> thesis) ==> thesis"
  5092     assumes "A | B"
  5093     obtains
  5094       A
  5095     | B
  5096 
  5097 The subsequent classical rules even refer to the formal "thesis"
  5098 explicitly:
  5099 
  5100   lemma classical:     -- "(~ thesis ==> thesis) ==> thesis"
  5101     obtains "~ thesis"
  5102 
  5103   lemma Peirce's_Law:  -- "((thesis ==> something) ==> thesis) ==> thesis"
  5104     obtains "thesis ==> something"
  5105 
  5106 The actual proof of an 'obtains' statement is analogous to that of the
  5107 Isar proof element 'obtain', only that there may be several cases.
  5108 Optional case names may be specified in parentheses; these will be
  5109 available both in the present proof and as annotations in the
  5110 resulting rule, for later use with the 'cases' method (cf. attribute
  5111 case_names).
  5112 
  5113 * Isar: the assumptions of a long theorem statement are available as
  5114 "assms" fact in the proof context.  This is more appropriate than the
  5115 (historical) "prems", which refers to all assumptions of the current
  5116 context, including those from the target locale, proof body etc.
  5117 
  5118 * Isar: 'print_statement' prints theorems from the current theory or
  5119 proof context in long statement form, according to the syntax of a
  5120 top-level lemma.
  5121 
  5122 * Isar: 'obtain' takes an optional case name for the local context
  5123 introduction rule (default "that").
  5124 
  5125 * Isar: removed obsolete 'concl is' patterns.  INCOMPATIBILITY, use
  5126 explicit (is "_ ==> ?foo") in the rare cases where this still happens
  5127 to occur.
  5128 
  5129 * Pure: syntax "CONST name" produces a fully internalized constant
  5130 according to the current context.  This is particularly useful for
  5131 syntax translations that should refer to internal constant
  5132 representations independently of name spaces.
  5133 
  5134 * Pure: syntax constant for foo (binder "FOO ") is called "foo_binder"
  5135 instead of "FOO ". This allows multiple binder declarations to coexist
  5136 in the same context.  INCOMPATIBILITY.
  5137 
  5138 * Isar/locales: 'notation' provides a robust interface to the 'syntax'
  5139 primitive that also works in a locale context (both for constants and
  5140 fixed variables). Type declaration and internal syntactic representation
  5141 of given constants retrieved from the context. Likewise, the
  5142 'no_notation' command allows to remove given syntax annotations from the
  5143 current context.
  5144 
  5145 * Isar/locales: new derived specification elements 'axiomatization',
  5146 'definition', 'abbreviation', which support type-inference, admit
  5147 object-level specifications (equality, equivalence).  See also the
  5148 isar-ref manual.  Examples:
  5149 
  5150   axiomatization
  5151     eq  (infix "===" 50) where
  5152     eq_refl: "x === x" and eq_subst: "x === y ==> P x ==> P y"
  5153 
  5154   definition "f x y = x + y + 1"
  5155   definition g where "g x = f x x"
  5156 
  5157   abbreviation
  5158     neq  (infix "=!=" 50) where
  5159     "x =!= y == ~ (x === y)"
  5160 
  5161 These specifications may be also used in a locale context.  Then the
  5162 constants being introduced depend on certain fixed parameters, and the
  5163 constant name is qualified by the locale base name.  An internal
  5164 abbreviation takes care for convenient input and output, making the
  5165 parameters implicit and using the original short name.  See also
  5166 src/HOL/ex/Abstract_NAT.thy for an example of deriving polymorphic
  5167 entities from a monomorphic theory.
  5168 
  5169 Presently, abbreviations are only available 'in' a target locale, but
  5170 not inherited by general import expressions.  Also note that
  5171 'abbreviation' may be used as a type-safe replacement for 'syntax' +
  5172 'translations' in common applications.  The "no_abbrevs" print mode
  5173 prevents folding of abbreviations in term output.
  5174 
  5175 Concrete syntax is attached to specified constants in internal form,
  5176 independently of name spaces.  The parse tree representation is
  5177 slightly different -- use 'notation' instead of raw 'syntax', and
  5178 'translations' with explicit "CONST" markup to accommodate this.
  5179 
  5180 * Pure/Isar: unified syntax for new-style specification mechanisms
  5181 (e.g.  'definition', 'abbreviation', or 'inductive' in HOL) admits
  5182 full type inference and dummy patterns ("_").  For example:
  5183 
  5184   definition "K x _ = x"
  5185 
  5186   inductive conj for A B
  5187   where "A ==> B ==> conj A B"
  5188 
  5189 * Pure: command 'print_abbrevs' prints all constant abbreviations of
  5190 the current context.  Print mode "no_abbrevs" prevents inversion of
  5191 abbreviations on output.
  5192 
  5193 * Isar/locales: improved parameter handling: use of locales "var" and
  5194 "struct" no longer necessary; - parameter renamings are no longer
  5195 required to be injective.  For example, this allows to define
  5196 endomorphisms as locale endom = homom mult mult h.
  5197 
  5198 * Isar/locales: changed the way locales with predicates are defined.
  5199 Instead of accumulating the specification, the imported expression is
  5200 now an interpretation.  INCOMPATIBILITY: different normal form of
  5201 locale expressions.  In particular, in interpretations of locales with
  5202 predicates, goals repesenting already interpreted fragments are not
  5203 removed automatically.  Use methods `intro_locales' and
  5204 `unfold_locales'; see below.
  5205 
  5206 * Isar/locales: new methods `intro_locales' and `unfold_locales'
  5207 provide backward reasoning on locales predicates.  The methods are
  5208 aware of interpretations and discharge corresponding goals.
  5209 `intro_locales' is less aggressive then `unfold_locales' and does not
  5210 unfold predicates to assumptions.
  5211 
  5212 * Isar/locales: the order in which locale fragments are accumulated
  5213 has changed.  This enables to override declarations from fragments due
  5214 to interpretations -- for example, unwanted simp rules.
  5215 
  5216 * Isar/locales: interpretation in theories and proof contexts has been
  5217 extended.  One may now specify (and prove) equations, which are
  5218 unfolded in interpreted theorems.  This is useful for replacing
  5219 defined concepts (constants depending on locale parameters) by
  5220 concepts already existing in the target context.  Example:
  5221 
  5222   interpretation partial_order ["op <= :: [int, int] => bool"]
  5223     where "partial_order.less (op <=) (x::int) y = (x < y)"
  5224 
  5225 Typically, the constant `partial_order.less' is created by a
  5226 definition specification element in the context of locale
  5227 partial_order.
  5228 
  5229 * Method "induct": improved internal context management to support
  5230 local fixes and defines on-the-fly. Thus explicit meta-level
  5231 connectives !!  and ==> are rarely required anymore in inductive goals
  5232 (using object-logic connectives for this purpose has been long
  5233 obsolete anyway). Common proof patterns are explained in
  5234 src/HOL/Induct/Common_Patterns.thy, see also
  5235 src/HOL/Isar_examples/Puzzle.thy and src/HOL/Lambda for realistic
  5236 examples.
  5237 
  5238 * Method "induct": improved handling of simultaneous goals. Instead of
  5239 introducing object-level conjunction, the statement is now split into
  5240 several conclusions, while the corresponding symbolic cases are nested
  5241 accordingly. INCOMPATIBILITY, proofs need to be structured explicitly,
  5242 see src/HOL/Induct/Common_Patterns.thy, for example.
  5243 
  5244 * Method "induct": mutual induction rules are now specified as a list
  5245 of rule sharing the same induction cases. HOL packages usually provide
  5246 foo_bar.inducts for mutually defined items foo and bar (e.g. inductive
  5247 predicates/sets or datatypes). INCOMPATIBILITY, users need to specify
  5248 mutual induction rules differently, i.e. like this:
  5249 
  5250   (induct rule: foo_bar.inducts)
  5251   (induct set: foo bar)
  5252   (induct pred: foo bar)
  5253   (induct type: foo bar)
  5254 
  5255 The ML function ProjectRule.projections turns old-style rules into the
  5256 new format.
  5257 
  5258 * Method "coinduct": dual of induction, see
  5259 src/HOL/Library/Coinductive_List.thy for various examples.
  5260 
  5261 * Method "cases", "induct", "coinduct": the ``(open)'' option is
  5262 considered a legacy feature.
  5263 
  5264 * Attribute "symmetric" produces result with standardized schematic
  5265 variables (index 0).  Potential INCOMPATIBILITY.
  5266 
  5267 * Simplifier: by default the simplifier trace only shows top level
  5268 rewrites now. That is, trace_simp_depth_limit is set to 1 by
  5269 default. Thus there is less danger of being flooded by the trace. The
  5270 trace indicates where parts have been suppressed.
  5271   
  5272 * Provers/classical: removed obsolete classical version of elim_format
  5273 attribute; classical elim/dest rules are now treated uniformly when
  5274 manipulating the claset.
  5275 
  5276 * Provers/classical: stricter checks to ensure that supplied intro,
  5277 dest and elim rules are well-formed; dest and elim rules must have at
  5278 least one premise.
  5279 
  5280 * Provers/classical: attributes dest/elim/intro take an optional
  5281 weight argument for the rule (just as the Pure versions).  Weights are
  5282 ignored by automated tools, but determine the search order of single
  5283 rule steps.
  5284 
  5285 * Syntax: input syntax now supports dummy variable binding "%_. b",
  5286 where the body does not mention the bound variable.  Note that dummy
  5287 patterns implicitly depend on their context of bounds, which makes
  5288 "{_. _}" match any set comprehension as expected.  Potential
  5289 INCOMPATIBILITY -- parse translations need to cope with syntactic
  5290 constant "_idtdummy" in the binding position.
  5291 
  5292 * Syntax: removed obsolete syntactic constant "_K" and its associated
  5293 parse translation.  INCOMPATIBILITY -- use dummy abstraction instead,
  5294 for example "A -> B" => "Pi A (%_. B)".
  5295 
  5296 * Pure: 'class_deps' command visualizes the subclass relation, using
  5297 the graph browser tool.
  5298 
  5299 * Pure: 'print_theory' now suppresses certain internal declarations by
  5300 default; use '!' option for full details.
  5301 
  5302 
  5303 *** HOL ***
  5304 
  5305 * Method "metis" proves goals by applying the Metis general-purpose
  5306 resolution prover (see also http://gilith.com/software/metis/).
  5307 Examples are in the directory MetisExamples.  WARNING: the
  5308 Isabelle/HOL-Metis integration does not yet work properly with
  5309 multi-threading.
  5310   
  5311 * Command 'sledgehammer' invokes external automatic theorem provers as
  5312 background processes.  It generates calls to the "metis" method if
  5313 successful. These can be pasted into the proof.  Users do not have to
  5314 wait for the automatic provers to return.  WARNING: does not really
  5315 work with multi-threading.
  5316 
  5317 * New "auto_quickcheck" feature tests outermost goal statements for
  5318 potential counter-examples.  Controlled by ML references
  5319 auto_quickcheck (default true) and auto_quickcheck_time_limit (default
  5320 5000 milliseconds).  Fails silently if statements is outside of
  5321 executable fragment, or any other codgenerator problem occurs.
  5322 
  5323 * New constant "undefined" with axiom "undefined x = undefined".
  5324 
  5325 * Added class "HOL.eq", allowing for code generation with polymorphic
  5326 equality.
  5327 
  5328 * Some renaming of class constants due to canonical name prefixing in
  5329 the new 'class' package:
  5330 
  5331     HOL.abs ~> HOL.abs_class.abs
  5332     HOL.divide ~> HOL.divide_class.divide
  5333     0 ~> HOL.zero_class.zero
  5334     1 ~> HOL.one_class.one
  5335     op + ~> HOL.plus_class.plus
  5336     op - ~> HOL.minus_class.minus
  5337     uminus ~> HOL.minus_class.uminus
  5338     op * ~> HOL.times_class.times
  5339     op < ~> HOL.ord_class.less
  5340     op <= > HOL.ord_class.less_eq
  5341     Nat.power ~> Power.power_class.power
  5342     Nat.size ~> Nat.size_class.size
  5343     Numeral.number_of ~> Numeral.number_class.number_of
  5344     FixedPoint.Inf ~> Lattices.complete_lattice_class.Inf
  5345     FixedPoint.Sup ~> Lattices.complete_lattice_class.Sup
  5346     Orderings.min ~> Orderings.ord_class.min
  5347     Orderings.max ~> Orderings.ord_class.max
  5348     Divides.op div ~> Divides.div_class.div
  5349     Divides.op mod ~> Divides.div_class.mod
  5350     Divides.op dvd ~> Divides.div_class.dvd
  5351 
  5352 INCOMPATIBILITY.  Adaptions may be required in the following cases:
  5353 
  5354 a) User-defined constants using any of the names "plus", "minus",
  5355 "times", "less" or "less_eq". The standard syntax translations for
  5356 "+", "-" and "*" may go wrong.  INCOMPATIBILITY: use more specific
  5357 names.
  5358 
  5359 b) Variables named "plus", "minus", "times", "less", "less_eq"
  5360 INCOMPATIBILITY: use more specific names.
  5361 
  5362 c) Permutative equations (e.g. "a + b = b + a")
  5363 Since the change of names also changes the order of terms, permutative
  5364 rewrite rules may get applied in a different order. Experience shows
  5365 that this is rarely the case (only two adaptions in the whole Isabelle
  5366 distribution).  INCOMPATIBILITY: rewrite proofs
  5367 
  5368 d) ML code directly refering to constant names
  5369 This in general only affects hand-written proof tactics, simprocs and
  5370 so on.  INCOMPATIBILITY: grep your sourcecode and replace names.
  5371 Consider using @{const_name} antiquotation.
  5372 
  5373 * New class "default" with associated constant "default".
  5374 
  5375 * Function "sgn" is now overloaded and available on int, real, complex
  5376 (and other numeric types), using class "sgn".  Two possible defs of
  5377 sgn are given as equational assumptions in the classes sgn_if and
  5378 sgn_div_norm; ordered_idom now also inherits from sgn_if.
  5379 INCOMPATIBILITY.
  5380 
  5381 * Locale "partial_order" now unified with class "order" (cf. theory
  5382 Orderings), added parameter "less".  INCOMPATIBILITY.
  5383 
  5384 * Renamings in classes "order" and "linorder": facts "refl", "trans" and
  5385 "cases" to "order_refl", "order_trans" and "linorder_cases", to avoid
  5386 clashes with HOL "refl" and "trans".  INCOMPATIBILITY.
  5387 
  5388 * Classes "order" and "linorder": potential INCOMPATIBILITY due to
  5389 changed order of proof goals in instance proofs.
  5390 
  5391 * The transitivity reasoner for partial and linear orders is set up
  5392 for classes "order" and "linorder".  Instances of the reasoner are available
  5393 in all contexts importing or interpreting the corresponding locales.
  5394 Method "order" invokes the reasoner separately; the reasoner
  5395 is also integrated with the Simplifier as a solver.  Diagnostic
  5396 command 'print_orders' shows the available instances of the reasoner
  5397 in the current context.
  5398 
  5399 * Localized monotonicity predicate in theory "Orderings"; integrated
  5400 lemmas max_of_mono and min_of_mono with this predicate.
  5401 INCOMPATIBILITY.
  5402 
  5403 * Formulation of theorem "dense" changed slightly due to integration
  5404 with new class dense_linear_order.
  5405 
  5406 * Uniform lattice theory development in HOL.
  5407 
  5408     constants "meet" and "join" now named "inf" and "sup"
  5409     constant "Meet" now named "Inf"
  5410 
  5411     classes "meet_semilorder" and "join_semilorder" now named
  5412       "lower_semilattice" and "upper_semilattice"
  5413     class "lorder" now named "lattice"
  5414     class "comp_lat" now named "complete_lattice"
  5415 
  5416     Instantiation of lattice classes allows explicit definitions
  5417     for "inf" and "sup" operations (or "Inf" and "Sup" for complete lattices).
  5418 
  5419   INCOMPATIBILITY.  Theorem renames:
  5420 
  5421     meet_left_le            ~> inf_le1
  5422     meet_right_le           ~> inf_le2
  5423     join_left_le            ~> sup_ge1
  5424     join_right_le           ~> sup_ge2
  5425     meet_join_le            ~> inf_sup_ord
  5426     le_meetI                ~> le_infI
  5427     join_leI                ~> le_supI
  5428     le_meet                 ~> le_inf_iff
  5429     le_join                 ~> ge_sup_conv
  5430     meet_idempotent         ~> inf_idem
  5431     join_idempotent         ~> sup_idem
  5432     meet_comm               ~> inf_commute
  5433     join_comm               ~> sup_commute
  5434     meet_leI1               ~> le_infI1
  5435     meet_leI2               ~> le_infI2
  5436     le_joinI1               ~> le_supI1
  5437     le_joinI2               ~> le_supI2
  5438     meet_assoc              ~> inf_assoc
  5439     join_assoc              ~> sup_assoc
  5440     meet_left_comm          ~> inf_left_commute
  5441     meet_left_idempotent    ~> inf_left_idem
  5442     join_left_comm          ~> sup_left_commute
  5443     join_left_idempotent    ~> sup_left_idem
  5444     meet_aci                ~> inf_aci
  5445     join_aci                ~> sup_aci
  5446     le_def_meet             ~> le_iff_inf
  5447     le_def_join             ~> le_iff_sup
  5448     join_absorp2            ~> sup_absorb2
  5449     join_absorp1            ~> sup_absorb1
  5450     meet_absorp1            ~> inf_absorb1
  5451     meet_absorp2            ~> inf_absorb2
  5452     meet_join_absorp        ~> inf_sup_absorb
  5453     join_meet_absorp        ~> sup_inf_absorb
  5454     distrib_join_le         ~> distrib_sup_le
  5455     distrib_meet_le         ~> distrib_inf_le
  5456 
  5457     add_meet_distrib_left   ~> add_inf_distrib_left
  5458     add_join_distrib_left   ~> add_sup_distrib_left
  5459     is_join_neg_meet        ~> is_join_neg_inf
  5460     is_meet_neg_join        ~> is_meet_neg_sup
  5461     add_meet_distrib_right  ~> add_inf_distrib_right
  5462     add_join_distrib_right  ~> add_sup_distrib_right
  5463     add_meet_join_distribs  ~> add_sup_inf_distribs
  5464     join_eq_neg_meet        ~> sup_eq_neg_inf
  5465     meet_eq_neg_join        ~> inf_eq_neg_sup
  5466     add_eq_meet_join        ~> add_eq_inf_sup
  5467     meet_0_imp_0            ~> inf_0_imp_0
  5468     join_0_imp_0            ~> sup_0_imp_0
  5469     meet_0_eq_0             ~> inf_0_eq_0
  5470     join_0_eq_0             ~> sup_0_eq_0
  5471     neg_meet_eq_join        ~> neg_inf_eq_sup
  5472     neg_join_eq_meet        ~> neg_sup_eq_inf
  5473     join_eq_if              ~> sup_eq_if
  5474 
  5475     mono_meet               ~> mono_inf
  5476     mono_join               ~> mono_sup
  5477     meet_bool_eq            ~> inf_bool_eq
  5478     join_bool_eq            ~> sup_bool_eq
  5479     meet_fun_eq             ~> inf_fun_eq
  5480     join_fun_eq             ~> sup_fun_eq
  5481     meet_set_eq             ~> inf_set_eq
  5482     join_set_eq             ~> sup_set_eq
  5483     meet1_iff               ~> inf1_iff
  5484     meet2_iff               ~> inf2_iff
  5485     meet1I                  ~> inf1I
  5486     meet2I                  ~> inf2I
  5487     meet1D1                 ~> inf1D1
  5488     meet2D1                 ~> inf2D1
  5489     meet1D2                 ~> inf1D2
  5490     meet2D2                 ~> inf2D2
  5491     meet1E                  ~> inf1E
  5492     meet2E                  ~> inf2E
  5493     join1_iff               ~> sup1_iff
  5494     join2_iff               ~> sup2_iff
  5495     join1I1                 ~> sup1I1
  5496     join2I1                 ~> sup2I1
  5497     join1I1                 ~> sup1I1
  5498     join2I2                 ~> sup1I2
  5499     join1CI                 ~> sup1CI
  5500     join2CI                 ~> sup2CI
  5501     join1E                  ~> sup1E
  5502     join2E                  ~> sup2E
  5503 
  5504     is_meet_Meet            ~> is_meet_Inf
  5505     Meet_bool_def           ~> Inf_bool_def
  5506     Meet_fun_def            ~> Inf_fun_def
  5507     Meet_greatest           ~> Inf_greatest
  5508     Meet_lower              ~> Inf_lower
  5509     Meet_set_def            ~> Inf_set_def
  5510 
  5511     Sup_def                 ~> Sup_Inf
  5512     Sup_bool_eq             ~> Sup_bool_def
  5513     Sup_fun_eq              ~> Sup_fun_def
  5514     Sup_set_eq              ~> Sup_set_def
  5515 
  5516     listsp_meetI            ~> listsp_infI
  5517     listsp_meet_eq          ~> listsp_inf_eq
  5518 
  5519     meet_min                ~> inf_min
  5520     join_max                ~> sup_max
  5521 
  5522 * Added syntactic class "size"; overloaded constant "size" now has
  5523 type "'a::size ==> bool"
  5524 
  5525 * Internal reorganisation of `size' of datatypes: size theorems
  5526 "foo.size" are no longer subsumed by "foo.simps" (but are still
  5527 simplification rules by default!); theorems "prod.size" now named
  5528 "*.size".
  5529 
  5530 * Class "div" now inherits from class "times" rather than "type".
  5531 INCOMPATIBILITY.
  5532 
  5533 * HOL/Finite_Set: "name-space" locales Lattice, Distrib_lattice,
  5534 Linorder etc.  have disappeared; operations defined in terms of
  5535 fold_set now are named Inf_fin, Sup_fin.  INCOMPATIBILITY.
  5536 
  5537 * HOL/Nat: neq0_conv no longer declared as iff.  INCOMPATIBILITY.
  5538 
  5539 * HOL-Word: New extensive library and type for generic, fixed size
  5540 machine words, with arithmetic, bit-wise, shifting and rotating
  5541 operations, reflection into int, nat, and bool lists, automation for
  5542 linear arithmetic (by automatic reflection into nat or int), including
  5543 lemmas on overflow and monotonicity.  Instantiated to all appropriate
  5544 arithmetic type classes, supporting automatic simplification of
  5545 numerals on all operations.
  5546 
  5547 * Library/Boolean_Algebra: locales for abstract boolean algebras.
  5548 
  5549 * Library/Numeral_Type: numbers as types, e.g. TYPE(32).
  5550 
  5551 * Code generator library theories:
  5552   - Code_Integer represents HOL integers by big integer literals in target
  5553     languages.
  5554   - Code_Char represents HOL characters by character literals in target
  5555     languages.
  5556   - Code_Char_chr like Code_Char, but also offers treatment of character
  5557     codes; includes Code_Integer.
  5558   - Executable_Set allows to generate code for finite sets using lists.
  5559   - Executable_Rat implements rational numbers as triples (sign, enumerator,
  5560     denominator).
  5561   - Executable_Real implements a subset of real numbers, namly those
  5562     representable by rational numbers.
  5563   - Efficient_Nat implements natural numbers by integers, which in general will
  5564     result in higher efficency; pattern matching with 0/Suc is eliminated;
  5565     includes Code_Integer.
  5566   - Code_Index provides an additional datatype index which is mapped to
  5567     target-language built-in integers.
  5568   - Code_Message provides an additional datatype message_string which is isomorphic to
  5569     strings; messages are mapped to target-language strings.
  5570 
  5571 * New package for inductive predicates
  5572 
  5573   An n-ary predicate p with m parameters z_1, ..., z_m can now be defined via
  5574 
  5575     inductive
  5576       p :: "U_1 => ... => U_m => T_1 => ... => T_n => bool"
  5577       for z_1 :: U_1 and ... and z_n :: U_m
  5578     where
  5579       rule_1: "... ==> p z_1 ... z_m t_1_1 ... t_1_n"
  5580     | ...
  5581 
  5582   with full support for type-inference, rather than
  5583 
  5584     consts s :: "U_1 => ... => U_m => (T_1 * ... * T_n) set"
  5585 
  5586     abbreviation p :: "U_1 => ... => U_m => T_1 => ... => T_n => bool"
  5587     where "p z_1 ... z_m x_1 ... x_n == (x_1, ..., x_n) : s z_1 ... z_m"
  5588 
  5589     inductive "s z_1 ... z_m"
  5590     intros
  5591       rule_1: "... ==> (t_1_1, ..., t_1_n) : s z_1 ... z_m"
  5592       ...
  5593 
  5594   For backward compatibility, there is a wrapper allowing inductive
  5595   sets to be defined with the new package via
  5596 
  5597     inductive_set
  5598       s :: "U_1 => ... => U_m => (T_1 * ... * T_n) set"
  5599       for z_1 :: U_1 and ... and z_n :: U_m
  5600     where
  5601       rule_1: "... ==> (t_1_1, ..., t_1_n) : s z_1 ... z_m"
  5602     | ...
  5603 
  5604   or
  5605 
  5606     inductive_set
  5607       s :: "U_1 => ... => U_m => (T_1 * ... * T_n) set"
  5608       and p :: "U_1 => ... => U_m => T_1 => ... => T_n => bool"
  5609       for z_1 :: U_1 and ... and z_n :: U_m
  5610     where
  5611       "p z_1 ... z_m x_1 ... x_n == (x_1, ..., x_n) : s z_1 ... z_m"
  5612     | rule_1: "... ==> p z_1 ... z_m t_1_1 ... t_1_n"
  5613     | ...
  5614 
  5615   if the additional syntax "p ..." is required.
  5616 
  5617   Numerous examples can be found in the subdirectories src/HOL/Auth,
  5618   src/HOL/Bali, src/HOL/Induct, and src/HOL/MicroJava.
  5619 
  5620   INCOMPATIBILITIES:
  5621 
  5622   - Since declaration and definition of inductive sets or predicates
  5623     is no longer separated, abbreviations involving the newly
  5624     introduced sets or predicates must be specified together with the
  5625     introduction rules after the 'where' keyword (see above), rather
  5626     than before the actual inductive definition.
  5627 
  5628   - The variables in induction and elimination rules are now
  5629     quantified in the order of their occurrence in the introduction
  5630     rules, rather than in alphabetical order. Since this may break
  5631     some proofs, these proofs either have to be repaired, e.g. by
  5632     reordering the variables a_i_1 ... a_i_{k_i} in Isar 'case'
  5633     statements of the form
  5634 
  5635       case (rule_i a_i_1 ... a_i_{k_i})
  5636 
  5637     or the old order of quantification has to be restored by explicitly adding
  5638     meta-level quantifiers in the introduction rules, i.e.
  5639 
  5640       | rule_i: "!!a_i_1 ... a_i_{k_i}. ... ==> p z_1 ... z_m t_i_1 ... t_i_n"
  5641 
  5642   - The format of the elimination rules is now
  5643 
  5644       p z_1 ... z_m x_1 ... x_n ==>
  5645         (!!a_1_1 ... a_1_{k_1}. x_1 = t_1_1 ==> ... ==> x_n = t_1_n ==> ... ==> P)
  5646         ==> ... ==> P
  5647 
  5648     for predicates and
  5649 
  5650       (x_1, ..., x_n) : s z_1 ... z_m ==>
  5651         (!!a_1_1 ... a_1_{k_1}. x_1 = t_1_1 ==> ... ==> x_n = t_1_n ==> ... ==> P)
  5652         ==> ... ==> P
  5653 
  5654     for sets rather than
  5655 
  5656       x : s z_1 ... z_m ==>
  5657         (!!a_1_1 ... a_1_{k_1}. x = (t_1_1, ..., t_1_n) ==> ... ==> P)
  5658         ==> ... ==> P
  5659 
  5660     This may require terms in goals to be expanded to n-tuples
  5661     (e.g. using case_tac or simplification with the split_paired_all
  5662     rule) before the above elimination rule is applicable.
  5663 
  5664   - The elimination or case analysis rules for (mutually) inductive
  5665     sets or predicates are now called "p_1.cases" ... "p_k.cases". The
  5666     list of rules "p_1_..._p_k.elims" is no longer available.
  5667 
  5668 * New package "function"/"fun" for general recursive functions,
  5669 supporting mutual and nested recursion, definitions in local contexts,
  5670 more general pattern matching and partiality. See HOL/ex/Fundefs.thy
  5671 for small examples, and the separate tutorial on the function
  5672 package. The old recdef "package" is still available as before, but
  5673 users are encouraged to use the new package.
  5674 
  5675 * Method "lexicographic_order" automatically synthesizes termination
  5676 relations as lexicographic combinations of size measures. 
  5677 
  5678 * Case-expressions allow arbitrary constructor-patterns (including
  5679 "_") and take their order into account, like in functional
  5680 programming.  Internally, this is translated into nested
  5681 case-expressions; missing cases are added and mapped to the predefined
  5682 constant "undefined". In complicated cases printing may no longer show
  5683 the original input but the internal form. Lambda-abstractions allow
  5684 the same form of pattern matching: "% pat1 => e1 | ..." is an
  5685 abbreviation for "%x. case x of pat1 => e1 | ..." where x is a new
  5686 variable.
  5687 
  5688 * IntDef: The constant "int :: nat => int" has been removed; now "int"
  5689 is an abbreviation for "of_nat :: nat => int". The simplification
  5690 rules for "of_nat" have been changed to work like "int" did
  5691 previously.  Potential INCOMPATIBILITY:
  5692   - "of_nat (Suc m)" simplifies to "1 + of_nat m" instead of "of_nat m + 1"
  5693   - of_nat_diff and of_nat_mult are no longer default simp rules
  5694 
  5695 * Method "algebra" solves polynomial equations over (semi)rings using
  5696 Groebner bases. The (semi)ring structure is defined by locales and the
  5697 tool setup depends on that generic context. Installing the method for
  5698 a specific type involves instantiating the locale and possibly adding
  5699 declarations for computation on the coefficients.  The method is
  5700 already instantiated for natural numbers and for the axiomatic class
  5701 of idoms with numerals.  See also the paper by Chaieb and Wenzel at
  5702 CALCULEMUS 2007 for the general principles underlying this
  5703 architecture of context-aware proof-tools.
  5704 
  5705 * Method "ferrack" implements quantifier elimination over
  5706 special-purpose dense linear orders using locales (analogous to
  5707 "algebra"). The method is already installed for class
  5708 {ordered_field,recpower,number_ring} which subsumes real, hyperreal,
  5709 rat, etc.
  5710 
  5711 * Former constant "List.op @" now named "List.append".  Use ML
  5712 antiquotations @{const_name List.append} or @{term " ... @ ... "} to
  5713 circumvent possible incompatibilities when working on ML level.
  5714 
  5715 * primrec: missing cases mapped to "undefined" instead of "arbitrary".
  5716 
  5717 * New function listsum :: 'a list => 'a for arbitrary monoids.
  5718 Special syntax: "SUM x <- xs. f x" (and latex variants)
  5719 
  5720 * New syntax for Haskell-like list comprehension (input only), eg.
  5721 [(x,y). x <- xs, y <- ys, x ~= y], see also src/HOL/List.thy.
  5722 
  5723 * The special syntax for function "filter" has changed from [x :
  5724 xs. P] to [x <- xs. P] to avoid an ambiguity caused by list
  5725 comprehension syntax, and for uniformity.  INCOMPATIBILITY.
  5726 
  5727 * [a..b] is now defined for arbitrary linear orders.  It used to be
  5728 defined on nat only, as an abbreviation for [a..<Suc b]
  5729 INCOMPATIBILITY.
  5730 
  5731 * Renamed lemma "set_take_whileD"  to "set_takeWhileD".
  5732 
  5733 * New functions "sorted" and "sort" in src/HOL/List.thy.
  5734 
  5735 * New lemma collection field_simps (an extension of ring_simps) for
  5736 manipulating (in)equations involving division. Multiplies with all
  5737 denominators that can be proved to be non-zero (in equations) or
  5738 positive/negative (in inequations).
  5739 
  5740 * Lemma collections ring_eq_simps, group_eq_simps and ring_distrib
  5741 have been improved and renamed to ring_simps, group_simps and
  5742 ring_distribs.  Removed lemmas field_xyz in theory Ring_and_Field
  5743 because they were subsumed by lemmas xyz.  INCOMPATIBILITY.
  5744 
  5745 * Theory Library/Commutative_Ring: switched from recdef to function
  5746 package; constants add, mul, pow now curried.  Infix syntax for
  5747 algebraic operations.
  5748 
  5749 * Dropped redundant lemma def_imp_eq in favor of meta_eq_to_obj_eq.
  5750 INCOMPATIBILITY.
  5751 
  5752 * Dropped redundant lemma if_def2 in favor of if_bool_eq_conj.
  5753 INCOMPATIBILITY.
  5754 
  5755 * HOL/records: generalised field-update to take a function on the
  5756 field rather than the new value: r(|A := x|) is translated to A_update
  5757 (K x) r The K-combinator that is internally used is called K_record.
  5758 INCOMPATIBILITY: Usage of the plain update functions has to be
  5759 adapted.
  5760  
  5761 * Class "semiring_0" now contains annihilation axioms x * 0 = 0 and 0
  5762 * x = 0, which are required for a semiring.  Richer structures do not
  5763 inherit from semiring_0 anymore, because this property is a theorem
  5764 there, not an axiom.  INCOMPATIBILITY: In instances of semiring_0,
  5765 there is more to prove, but this is mostly trivial.
  5766 
  5767 * Class "recpower" is generalized to arbitrary monoids, not just
  5768 commutative semirings.  INCOMPATIBILITY: may need to incorporate
  5769 commutativity or semiring properties additionally.
  5770 
  5771 * Constant "List.list_all2" in List.thy now uses authentic syntax.
  5772 INCOMPATIBILITY: translations containing list_all2 may go wrong,
  5773 better use 'abbreviation'.
  5774 
  5775 * Renamed constant "List.op mem" to "List.member".  INCOMPATIBILITY.
  5776 
  5777 * Numeral syntax: type 'bin' which was a mere type copy of 'int' has
  5778 been abandoned in favour of plain 'int'.  INCOMPATIBILITY --
  5779 significant changes for setting up numeral syntax for types:
  5780   - New constants Numeral.pred and Numeral.succ instead
  5781       of former Numeral.bin_pred and Numeral.bin_succ.
  5782   - Use integer operations instead of bin_add, bin_mult and so on.
  5783   - Numeral simplification theorems named Numeral.numeral_simps instead of Bin_simps.
  5784   - ML structure Bin_Simprocs now named Int_Numeral_Base_Simprocs.
  5785 
  5786 See src/HOL/Integ/IntArith.thy for an example setup.
  5787 
  5788 * Command 'normal_form' computes the normal form of a term that may
  5789 contain free variables.  For example ``normal_form "rev [a, b, c]"''
  5790 produces ``[b, c, a]'' (without proof).  This command is suitable for
  5791 heavy-duty computations because the functions are compiled to ML
  5792 first.  Correspondingly, a method "normalization" is provided.  See
  5793 further src/HOL/ex/NormalForm.thy and src/Tools/nbe.ML.
  5794 
  5795 * Alternative iff syntax "A <-> B" for equality on bool (with priority
  5796 25 like -->); output depends on the "iff" print_mode, the default is
  5797 "A = B" (with priority 50).
  5798 
  5799 * Relations less (<) and less_eq (<=) are also available on type bool.
  5800 Modified syntax to disallow nesting without explicit parentheses,
  5801 e.g. "(x < y) < z" or "x < (y < z)", but NOT "x < y < z".  Potential
  5802 INCOMPATIBILITY.
  5803 
  5804 * "LEAST x:A. P" expands to "LEAST x. x:A & P" (input only).
  5805 
  5806 * Relation composition operator "op O" now has precedence 75 and binds
  5807 stronger than union and intersection. INCOMPATIBILITY.
  5808 
  5809 * The old set interval syntax "{m..n(}" (and relatives) has been
  5810 removed.  Use "{m..<n}" (and relatives) instead.
  5811 
  5812 * In the context of the assumption "~(s = t)" the Simplifier rewrites
  5813 "t = s" to False (by simproc "neq").  INCOMPATIBILITY, consider using
  5814 ``declare [[simproc del: neq]]''.
  5815 
  5816 * Simplifier: "m dvd n" where m and n are numbers is evaluated to
  5817 True/False.
  5818 
  5819 * Theorem Cons_eq_map_conv no longer declared as "simp".
  5820 
  5821 * Theorem setsum_mult renamed to setsum_right_distrib.
  5822 
  5823 * Prefer ex1I over ex_ex1I in single-step reasoning, e.g. by the
  5824 ``rule'' method.
  5825 
  5826 * Reimplemented methods "sat" and "satx", with several improvements:
  5827 goals no longer need to be stated as "<prems> ==> False", equivalences
  5828 (i.e. "=" on type bool) are handled, variable names of the form
  5829 "lit_<n>" are no longer reserved, significant speedup.
  5830 
  5831 * Methods "sat" and "satx" can now replay MiniSat proof traces.
  5832 zChaff is still supported as well.
  5833 
  5834 * 'inductive' and 'datatype': provide projections of mutual rules,
  5835 bundled as foo_bar.inducts;
  5836 
  5837 * Library: moved theories Parity, GCD, Binomial, Infinite_Set to
  5838 Library.
  5839 
  5840 * Library: moved theory Accessible_Part to main HOL.
  5841 
  5842 * Library: added theory Coinductive_List of potentially infinite lists
  5843 as greatest fixed-point.
  5844 
  5845 * Library: added theory AssocList which implements (finite) maps as
  5846 association lists.
  5847 
  5848 * Method "evaluation" solves goals (i.e. a boolean expression)
  5849 efficiently by compiling it to ML.  The goal is "proved" (via an
  5850 oracle) if it evaluates to True.
  5851 
  5852 * Linear arithmetic now splits certain operators (e.g. min, max, abs)
  5853 also when invoked by the simplifier.  This results in the Simplifier
  5854 being more powerful on arithmetic goals.  INCOMPATIBILITY.
  5855 Configuration option fast_arith_split_limit=0 recovers the old
  5856 behavior.
  5857 
  5858 * Support for hex (0x20) and binary (0b1001) numerals.
  5859 
  5860 * New method: reify eqs (t), where eqs are equations for an
  5861 interpretation I :: 'a list => 'b => 'c and t::'c is an optional
  5862 parameter, computes a term s::'b and a list xs::'a list and proves the
  5863 theorem I xs s = t. This is also known as reification or quoting. The
  5864 resulting theorem is applied to the subgoal to substitute t with I xs
  5865 s.  If t is omitted, the subgoal itself is reified.
  5866 
  5867 * New method: reflection corr_thm eqs (t). The parameters eqs and (t)
  5868 are as explained above. corr_thm is a theorem for I vs (f t) = I vs t,
  5869 where f is supposed to be a computable function (in the sense of code
  5870 generattion). The method uses reify to compute s and xs as above then
  5871 applies corr_thm and uses normalization by evaluation to "prove" f s =
  5872 r and finally gets the theorem t = r, which is again applied to the
  5873 subgoal. An Example is available in src/HOL/ex/ReflectionEx.thy.
  5874 
  5875 * Reflection: Automatic reification now handels binding, an example is
  5876 available in src/HOL/ex/ReflectionEx.thy
  5877 
  5878 * HOL-Statespace: ``State Spaces: The Locale Way'' introduces a
  5879 command 'statespace' that is similar to 'record', but introduces an
  5880 abstract specification based on the locale infrastructure instead of
  5881 HOL types.  This leads to extra flexibility in composing state spaces,
  5882 in particular multiple inheritance and renaming of components.
  5883 
  5884 
  5885 *** HOL-Complex ***
  5886 
  5887 * Hyperreal: Functions root and sqrt are now defined on negative real
  5888 inputs so that root n (- x) = - root n x and sqrt (- x) = - sqrt x.
  5889 Nonnegativity side conditions have been removed from many lemmas, so
  5890 that more subgoals may now be solved by simplification; potential
  5891 INCOMPATIBILITY.
  5892 
  5893 * Real: new type classes formalize real normed vector spaces and
  5894 algebras, using new overloaded constants scaleR :: real => 'a => 'a
  5895 and norm :: 'a => real.
  5896 
  5897 * Real: constant of_real :: real => 'a::real_algebra_1 injects from
  5898 reals into other types. The overloaded constant Reals :: 'a set is now
  5899 defined as range of_real; potential INCOMPATIBILITY.
  5900 
  5901 * Real: proper support for ML code generation, including 'quickcheck'.
  5902 Reals are implemented as arbitrary precision rationals.
  5903 
  5904 * Hyperreal: Several constants that previously worked only for the
  5905 reals have been generalized, so they now work over arbitrary vector
  5906 spaces. Type annotations may need to be added in some cases; potential
  5907 INCOMPATIBILITY.
  5908 
  5909   Infinitesimal  :: ('a::real_normed_vector) star set
  5910   HFinite        :: ('a::real_normed_vector) star set
  5911   HInfinite      :: ('a::real_normed_vector) star set
  5912   approx         :: ('a::real_normed_vector) star => 'a star => bool
  5913   monad          :: ('a::real_normed_vector) star => 'a star set
  5914   galaxy         :: ('a::real_normed_vector) star => 'a star set
  5915   (NS)LIMSEQ     :: [nat => 'a::real_normed_vector, 'a] => bool
  5916   (NS)convergent :: (nat => 'a::real_normed_vector) => bool
  5917   (NS)Bseq       :: (nat => 'a::real_normed_vector) => bool
  5918   (NS)Cauchy     :: (nat => 'a::real_normed_vector) => bool
  5919   (NS)LIM        :: ['a::real_normed_vector => 'b::real_normed_vector, 'a, 'b] => bool
  5920   is(NS)Cont     :: ['a::real_normed_vector => 'b::real_normed_vector, 'a] => bool
  5921   deriv          :: ['a::real_normed_field => 'a, 'a, 'a] => bool
  5922   sgn            :: 'a::real_normed_vector => 'a
  5923   exp            :: 'a::{recpower,real_normed_field,banach} => 'a
  5924 
  5925 * Complex: Some complex-specific constants are now abbreviations for
  5926 overloaded ones: complex_of_real = of_real, cmod = norm, hcmod =
  5927 hnorm.  Other constants have been entirely removed in favor of the
  5928 polymorphic versions (INCOMPATIBILITY):
  5929 
  5930   approx        <-- capprox
  5931   HFinite       <-- CFinite
  5932   HInfinite     <-- CInfinite
  5933   Infinitesimal <-- CInfinitesimal
  5934   monad         <-- cmonad
  5935   galaxy        <-- cgalaxy
  5936   (NS)LIM       <-- (NS)CLIM, (NS)CRLIM
  5937   is(NS)Cont    <-- is(NS)Contc, is(NS)contCR
  5938   (ns)deriv     <-- (ns)cderiv
  5939 
  5940 
  5941 *** HOL-Algebra ***
  5942 
  5943 * Formalisation of ideals and the quotient construction over rings.
  5944 
  5945 * Order and lattice theory no longer based on records.
  5946 INCOMPATIBILITY.
  5947 
  5948 * Renamed lemmas least_carrier -> least_closed and greatest_carrier ->
  5949 greatest_closed.  INCOMPATIBILITY.
  5950 
  5951 * Method algebra is now set up via an attribute.  For examples see
  5952 Ring.thy.  INCOMPATIBILITY: the method is now weaker on combinations
  5953 of algebraic structures.
  5954 
  5955 * Renamed theory CRing to Ring.
  5956 
  5957 
  5958 *** HOL-Nominal ***
  5959 
  5960 * Substantial, yet incomplete support for nominal datatypes (binding
  5961 structures) based on HOL-Nominal logic.  See src/HOL/Nominal and
  5962 src/HOL/Nominal/Examples.  Prospective users should consult
  5963 http://isabelle.in.tum.de/nominal/
  5964 
  5965 
  5966 *** ML ***
  5967 
  5968 * ML basics: just one true type int, which coincides with IntInf.int
  5969 (even on SML/NJ).
  5970 
  5971 * ML within Isar: antiquotations allow to embed statically-checked
  5972 formal entities in the source, referring to the context available at
  5973 compile-time.  For example:
  5974 
  5975 ML {* @{sort "{zero,one}"} *}
  5976 ML {* @{typ "'a => 'b"} *}
  5977 ML {* @{term "%x. x"} *}
  5978 ML {* @{prop "x == y"} *}
  5979 ML {* @{ctyp "'a => 'b"} *}
  5980 ML {* @{cterm "%x. x"} *}
  5981 ML {* @{cprop "x == y"} *}
  5982 ML {* @{thm asm_rl} *}
  5983 ML {* @{thms asm_rl} *}
  5984 ML {* @{type_name c} *}
  5985 ML {* @{type_syntax c} *}
  5986 ML {* @{const_name c} *}
  5987 ML {* @{const_syntax c} *}
  5988 ML {* @{context} *}
  5989 ML {* @{theory} *}
  5990 ML {* @{theory Pure} *}
  5991 ML {* @{theory_ref} *}
  5992 ML {* @{theory_ref Pure} *}
  5993 ML {* @{simpset} *}
  5994 ML {* @{claset} *}
  5995 ML {* @{clasimpset} *}
  5996 
  5997 The same works for sources being ``used'' within an Isar context.
  5998 
  5999 * ML in Isar: improved error reporting; extra verbosity with
  6000 ML_Context.trace enabled.
  6001 
  6002 * Pure/General/table.ML: the join operations now works via exceptions
  6003 DUP/SAME instead of type option. This is simpler in simple cases, and
  6004 admits slightly more efficient complex applications.
  6005 
  6006 * Pure: 'advanced' translation functions (parse_translation etc.) now
  6007 use Context.generic instead of just theory.
  6008 
  6009 * Pure: datatype Context.generic joins theory/Proof.context and
  6010 provides some facilities for code that works in either kind of
  6011 context, notably GenericDataFun for uniform theory and proof data.
  6012 
  6013 * Pure: simplified internal attribute type, which is now always
  6014 Context.generic * thm -> Context.generic * thm. Global (theory) vs.
  6015 local (Proof.context) attributes have been discontinued, while
  6016 minimizing code duplication. Thm.rule_attribute and
  6017 Thm.declaration_attribute build canonical attributes; see also structure
  6018 Context for further operations on Context.generic, notably
  6019 GenericDataFun. INCOMPATIBILITY, need to adapt attribute type
  6020 declarations and definitions.
  6021 
  6022 * Context data interfaces (Theory/Proof/GenericDataFun): removed
  6023 name/print, uninitialized data defaults to ad-hoc copy of empty value,
  6024 init only required for impure data. INCOMPATIBILITY: empty really need
  6025 to be empty (no dependencies on theory content!)
  6026 
  6027 * Pure/kernel: consts certification ignores sort constraints given in
  6028 signature declarations. (This information is not relevant to the
  6029 logic, but only for type inference.) SIGNIFICANT INTERNAL CHANGE,
  6030 potential INCOMPATIBILITY.
  6031 
  6032 * Pure: axiomatic type classes are now purely definitional, with
  6033 explicit proofs of class axioms and super class relations performed
  6034 internally. See Pure/axclass.ML for the main internal interfaces --
  6035 notably AxClass.define_class supercedes AxClass.add_axclass, and
  6036 AxClass.axiomatize_class/classrel/arity supersede
  6037 Sign.add_classes/classrel/arities.
  6038 
  6039 * Pure/Isar: Args/Attrib parsers operate on Context.generic --
  6040 global/local versions on theory vs. Proof.context have been
  6041 discontinued; Attrib.syntax and Method.syntax have been adapted
  6042 accordingly.  INCOMPATIBILITY, need to adapt parser expressions for
  6043 attributes, methods, etc.
  6044 
  6045 * Pure: several functions of signature "... -> theory -> theory * ..."
  6046 have been reoriented to "... -> theory -> ... * theory" in order to
  6047 allow natural usage in combination with the ||>, ||>>, |-> and
  6048 fold_map combinators.
  6049 
  6050 * Pure: official theorem names (closed derivations) and additional
  6051 comments (tags) are now strictly separate.  Name hints -- which are
  6052 maintained as tags -- may be attached any time without affecting the
  6053 derivation.
  6054 
  6055 * Pure: primitive rule lift_rule now takes goal cterm instead of an
  6056 actual goal state (thm).  Use Thm.lift_rule (Thm.cprem_of st i) to
  6057 achieve the old behaviour.
  6058 
  6059 * Pure: the "Goal" constant is now called "prop", supporting a
  6060 slightly more general idea of ``protecting'' meta-level rule
  6061 statements.
  6062 
  6063 * Pure: Logic.(un)varify only works in a global context, which is now
  6064 enforced instead of silently assumed.  INCOMPATIBILITY, may use
  6065 Logic.legacy_(un)varify as temporary workaround.
  6066 
  6067 * Pure: structure Name provides scalable operations for generating
  6068 internal variable names, notably Name.variants etc.  This replaces
  6069 some popular functions from term.ML:
  6070 
  6071   Term.variant		->  Name.variant
  6072   Term.variantlist	->  Name.variant_list
  6073   Term.invent_names	->  Name.invent_list
  6074 
  6075 Note that low-level renaming rarely occurs in new code -- operations
  6076 from structure Variable are used instead (see below).
  6077 
  6078 * Pure: structure Variable provides fundamental operations for proper
  6079 treatment of fixed/schematic variables in a context.  For example,
  6080 Variable.import introduces fixes for schematics of given facts and
  6081 Variable.export reverses the effect (up to renaming) -- this replaces
  6082 various freeze_thaw operations.
  6083 
  6084 * Pure: structure Goal provides simple interfaces for
  6085 init/conclude/finish and tactical prove operations (replacing former
  6086 Tactic.prove).  Goal.prove is the canonical way to prove results
  6087 within a given context; Goal.prove_global is a degraded version for
  6088 theory level goals, including a global Drule.standard.  Note that
  6089 OldGoals.prove_goalw_cterm has long been obsolete, since it is
  6090 ill-behaved in a local proof context (e.g. with local fixes/assumes or
  6091 in a locale context).
  6092 
  6093 * Pure/Syntax: generic interfaces for parsing (Syntax.parse_term etc.)
  6094 and type checking (Syntax.check_term etc.), with common combinations
  6095 (Syntax.read_term etc.). These supersede former Sign.read_term etc.
  6096 which are considered legacy and await removal.
  6097 
  6098 * Pure/Syntax: generic interfaces for type unchecking
  6099 (Syntax.uncheck_terms etc.) and unparsing (Syntax.unparse_term etc.),
  6100 with common combinations (Syntax.pretty_term, Syntax.string_of_term
  6101 etc.).  Former Sign.pretty_term, Sign.string_of_term etc. are still
  6102 available for convenience, but refer to the very same operations using
  6103 a mere theory instead of a full context.
  6104 
  6105 * Isar: simplified treatment of user-level errors, using exception
  6106 ERROR of string uniformly.  Function error now merely raises ERROR,
  6107 without any side effect on output channels.  The Isar toplevel takes
  6108 care of proper display of ERROR exceptions.  ML code may use plain
  6109 handle/can/try; cat_error may be used to concatenate errors like this:
  6110 
  6111   ... handle ERROR msg => cat_error msg "..."
  6112 
  6113 Toplevel ML code (run directly or through the Isar toplevel) may be
  6114 embedded into the Isar toplevel with exception display/debug like
  6115 this:
  6116 
  6117   Isar.toplevel (fn () => ...)
  6118 
  6119 INCOMPATIBILITY, removed special transform_error facilities, removed
  6120 obsolete variants of user-level exceptions (ERROR_MESSAGE,
  6121 Context.PROOF, ProofContext.CONTEXT, Proof.STATE, ProofHistory.FAIL)
  6122 -- use plain ERROR instead.
  6123 
  6124 * Isar: theory setup now has type (theory -> theory), instead of a
  6125 list.  INCOMPATIBILITY, may use #> to compose setup functions.
  6126 
  6127 * Isar: ML toplevel pretty printer for type Proof.context, subject to
  6128 ProofContext.debug/verbose flags.
  6129 
  6130 * Isar: Toplevel.theory_to_proof admits transactions that modify the
  6131 theory before entering a proof state.  Transactions now always see a
  6132 quasi-functional intermediate checkpoint, both in interactive and
  6133 batch mode.
  6134 
  6135 * Isar: simplified interfaces for outer syntax.  Renamed
  6136 OuterSyntax.add_keywords to OuterSyntax.keywords.  Removed
  6137 OuterSyntax.add_parsers -- this functionality is now included in
  6138 OuterSyntax.command etc.  INCOMPATIBILITY.
  6139 
  6140 * Simplifier: the simpset of a running simplification process now
  6141 contains a proof context (cf. Simplifier.the_context), which is the
  6142 very context that the initial simpset has been retrieved from (by
  6143 simpset_of/local_simpset_of).  Consequently, all plug-in components
  6144 (solver, looper etc.) may depend on arbitrary proof data.
  6145 
  6146 * Simplifier.inherit_context inherits the proof context (plus the
  6147 local bounds) of the current simplification process; any simproc
  6148 etc. that calls the Simplifier recursively should do this!  Removed
  6149 former Simplifier.inherit_bounds, which is already included here --
  6150 INCOMPATIBILITY.  Tools based on low-level rewriting may even have to
  6151 specify an explicit context using Simplifier.context/theory_context.
  6152 
  6153 * Simplifier/Classical Reasoner: more abstract interfaces
  6154 change_simpset/claset for modifying the simpset/claset reference of a
  6155 theory; raw versions simpset/claset_ref etc. have been discontinued --
  6156 INCOMPATIBILITY.
  6157 
  6158 * Provers: more generic wrt. syntax of object-logics, avoid hardwired
  6159 "Trueprop" etc.
  6160 
  6161 
  6162 *** System ***
  6163 
  6164 * settings: the default heap location within ISABELLE_HOME_USER now
  6165 includes ISABELLE_IDENTIFIER.  This simplifies use of multiple
  6166 Isabelle installations.
  6167 
  6168 * isabelle-process: option -S (secure mode) disables some critical
  6169 operations, notably runtime compilation and evaluation of ML source
  6170 code.
  6171 
  6172 * Basic Isabelle mode for jEdit, see Isabelle/lib/jedit/.
  6173 
  6174 * Support for parallel execution, using native multicore support of
  6175 Poly/ML 5.1.  The theory loader exploits parallelism when processing
  6176 independent theories, according to the given theory header
  6177 specifications. The maximum number of worker threads is specified via
  6178 usedir option -M or the "max-threads" setting in Proof General. A
  6179 speedup factor of 1.5--3.5 can be expected on a 4-core machine, and up
  6180 to 6 on a 8-core machine.  User-code needs to observe certain
  6181 guidelines for thread-safe programming, see appendix A in the Isar
  6182 Implementation manual.
  6183 
  6184 
  6185 
  6186 New in Isabelle2005 (October 2005)
  6187 ----------------------------------
  6188 
  6189 *** General ***
  6190 
  6191 * Theory headers: the new header syntax for Isar theories is
  6192 
  6193   theory <name>
  6194   imports <theory1> ... <theoryN>
  6195   uses <file1> ... <fileM>
  6196   begin
  6197 
  6198 where the 'uses' part is optional.  The previous syntax
  6199 
  6200   theory <name> = <theory1> + ... + <theoryN>:
  6201 
  6202 will disappear in the next release.  Use isatool fixheaders to convert
  6203 existing theory files.  Note that there is no change in ancient
  6204 non-Isar theories now, but these will disappear soon.
  6205 
  6206 * Theory loader: parent theories can now also be referred to via
  6207 relative and absolute paths.
  6208 
  6209 * Command 'find_theorems' searches for a list of criteria instead of a
  6210 list of constants. Known criteria are: intro, elim, dest, name:string,
  6211 simp:term, and any term. Criteria can be preceded by '-' to select
  6212 theorems that do not match. Intro, elim, dest select theorems that
  6213 match the current goal, name:s selects theorems whose fully qualified
  6214 name contain s, and simp:term selects all simplification rules whose
  6215 lhs match term.  Any other term is interpreted as pattern and selects
  6216 all theorems matching the pattern. Available in ProofGeneral under
  6217 'ProofGeneral -> Find Theorems' or C-c C-f.  Example:
  6218 
  6219   C-c C-f (100) "(_::nat) + _ + _" intro -name: "HOL."
  6220 
  6221 prints the last 100 theorems matching the pattern "(_::nat) + _ + _",
  6222 matching the current goal as introduction rule and not having "HOL."
  6223 in their name (i.e. not being defined in theory HOL).
  6224 
  6225 * Command 'thms_containing' has been discontinued in favour of
  6226 'find_theorems'; INCOMPATIBILITY.
  6227 
  6228 * Communication with Proof General is now 8bit clean, which means that
  6229 Unicode text in UTF-8 encoding may be used within theory texts (both
  6230 formal and informal parts).  Cf. option -U of the Isabelle Proof
  6231 General interface.  Here are some simple examples (cf. src/HOL/ex):
  6232 
  6233   http://isabelle.in.tum.de/library/HOL/ex/Hebrew.html
  6234   http://isabelle.in.tum.de/library/HOL/ex/Chinese.html
  6235 
  6236 * Improved efficiency of the Simplifier and, to a lesser degree, the
  6237 Classical Reasoner.  Typical big applications run around 2 times
  6238 faster.
  6239 
  6240