NEWS
author haftmann
Sun Jul 22 10:00:51 2012 +0200 (2012-07-22)
changeset 48428 ffa0618cc4d4
parent 48402 327ebf1c42a8
child 48431 6efff142bb54
permissions -rw-r--r--
NEWS
     1 Isabelle NEWS -- history user-relevant changes
     2 ==============================================
     3 
     4 New in this Isabelle version
     5 ----------------------------
     6 
     7 *** General ***
     8 
     9 * Discontinued obsolete method fastsimp / tactic fast_simp_tac, which
    10 is called fastforce / fast_force_tac already since Isabelle2011-1.
    11 
    12 * Updated and extended "isar-ref" manual, reduced remaining material
    13 in old "ref" manual.
    14 
    15 
    16 *** Pure ***
    17 
    18 * Command "export_code": relative file names are interpreted
    19 relatively to master directory of current theory rather than
    20 the rather arbitrary current working directory.
    21 INCOMPATIBILITY.
    22 
    23 * Discontinued obsolete attribute "COMP".  Potential INCOMPATIBILITY,
    24 use regular rule composition via "OF" / "THEN", or explicit proof
    25 structure instead.  Note that Isabelle/ML provides a variety of
    26 operators like COMP, INCR_COMP, COMP_INCR, which need to be applied
    27 with some care where this is really required.
    28 
    29 
    30 *** HOL ***
    31 
    32 * Library/Debug.thy and Library/Parallel.thy: debugging and parallel
    33 execution for code generated towards Isabelle/ML.
    34 
    35 * Simproc "finite_Collect" rewrites set comprehensions into pointfree
    36 expressions.
    37 
    38 * Quickcheck:
    39 
    40   - added an optimisation for equality premises.
    41     It is switched on by default, and can be switched off by setting
    42     the configuration quickcheck_optimise_equality to false.    
    43 
    44 * The SMT solver Z3 has now by default a restricted set of directly
    45 supported features. For the full set of features (div/mod, nonlinear
    46 arithmetic, datatypes/records) with potential proof reconstruction
    47 failures, enable the configuration option "z3_with_extensions".
    48 Minor INCOMPATIBILITY.
    49 
    50 * Sledgehammer:
    51 
    52   - Added MaSh relevance filter based on machine-learning; see the
    53     Sledgehammer manual for details.
    54   - Rationalized type encodings ("type_enc" option).
    55   - Renamed options:
    56       max_relevant ~> max_facts
    57       relevance_thresholds ~> fact_thresholds
    58 
    59 
    60 *** Document preparation ***
    61 
    62 * Default for \<euro> is now based on eurosym package, instead of
    63 slightly exotic babel/greek.
    64 
    65 
    66 *** System ***
    67 
    68 * Discontinued support for Poly/ML 5.2.1, which was the last version
    69 without exception positions and advanced ML compiler/toplevel
    70 configuration.
    71 
    72 
    73 
    74 New in Isabelle2012 (May 2012)
    75 ------------------------------
    76 
    77 *** General ***
    78 
    79 * Prover IDE (PIDE) improvements:
    80 
    81   - more robust Sledgehammer integration (as before the sledgehammer
    82     command-line needs to be typed into the source buffer)
    83   - markup for bound variables
    84   - markup for types of term variables (displayed as tooltips)
    85   - support for user-defined Isar commands within the running session
    86   - improved support for Unicode outside original 16bit range
    87     e.g. glyph for \<A> (thanks to jEdit 4.5.1)
    88 
    89 * Forward declaration of outer syntax keywords within the theory
    90 header -- minor INCOMPATIBILITY for user-defined commands.  Allow new
    91 commands to be used in the same theory where defined.
    92 
    93 * Auxiliary contexts indicate block structure for specifications with
    94 additional parameters and assumptions.  Such unnamed contexts may be
    95 nested within other targets, like 'theory', 'locale', 'class',
    96 'instantiation' etc.  Results from the local context are generalized
    97 accordingly and applied to the enclosing target context.  Example:
    98 
    99   context
   100     fixes x y z :: 'a
   101     assumes xy: "x = y" and yz: "y = z"
   102   begin
   103 
   104   lemma my_trans: "x = z" using xy yz by simp
   105 
   106   end
   107 
   108   thm my_trans
   109 
   110 The most basic application is to factor-out context elements of
   111 several fixes/assumes/shows theorem statements, e.g. see
   112 ~~/src/HOL/Isar_Examples/Group_Context.thy
   113 
   114 Any other local theory specification element works within the "context
   115 ... begin ... end" block as well.
   116 
   117 * Bundled declarations associate attributed fact expressions with a
   118 given name in the context.  These may be later included in other
   119 contexts.  This allows to manage context extensions casually, without
   120 the logical dependencies of locales and locale interpretation.  See
   121 commands 'bundle', 'include', 'including' etc. in the isar-ref manual.
   122 
   123 * Commands 'lemmas' and 'theorems' allow local variables using 'for'
   124 declaration, and results are standardized before being stored.  Thus
   125 old-style "standard" after instantiation or composition of facts
   126 becomes obsolete.  Minor INCOMPATIBILITY, due to potential change of
   127 indices of schematic variables.
   128 
   129 * Rule attributes in local theory declarations (e.g. locale or class)
   130 are now statically evaluated: the resulting theorem is stored instead
   131 of the original expression.  INCOMPATIBILITY in rare situations, where
   132 the historic accident of dynamic re-evaluation in interpretations
   133 etc. was exploited.
   134 
   135 * New tutorial "Programming and Proving in Isabelle/HOL"
   136 ("prog-prove").  It completely supersedes "A Tutorial Introduction to
   137 Structured Isar Proofs" ("isar-overview"), which has been removed.  It
   138 also supersedes "Isabelle/HOL, A Proof Assistant for Higher-Order
   139 Logic" as the recommended beginners tutorial, but does not cover all
   140 of the material of that old tutorial.
   141 
   142 * Updated and extended reference manuals: "isar-ref",
   143 "implementation", "system"; reduced remaining material in old "ref"
   144 manual.
   145 
   146 
   147 *** Pure ***
   148 
   149 * Command 'definition' no longer exports the foundational "raw_def"
   150 into the user context.  Minor INCOMPATIBILITY, may use the regular
   151 "def" result with attribute "abs_def" to imitate the old version.
   152 
   153 * Attribute "abs_def" turns an equation of the form "f x y == t" into
   154 "f == %x y. t", which ensures that "simp" or "unfold" steps always
   155 expand it.  This also works for object-logic equality.  (Formerly
   156 undocumented feature.)
   157 
   158 * Sort constraints are now propagated in simultaneous statements, just
   159 like type constraints.  INCOMPATIBILITY in rare situations, where
   160 distinct sorts used to be assigned accidentally.  For example:
   161 
   162   lemma "P (x::'a::foo)" and "Q (y::'a::bar)"  -- "now illegal"
   163 
   164   lemma "P (x::'a)" and "Q (y::'a::bar)"
   165     -- "now uniform 'a::bar instead of default sort for first occurrence (!)"
   166 
   167 * Rule composition via attribute "OF" (or ML functions OF/MRS) is more
   168 tolerant against multiple unifiers, as long as the final result is
   169 unique.  (As before, rules are composed in canonical right-to-left
   170 order to accommodate newly introduced premises.)
   171 
   172 * Renamed some inner syntax categories:
   173 
   174     num ~> num_token
   175     xnum ~> xnum_token
   176     xstr ~> str_token
   177 
   178 Minor INCOMPATIBILITY.  Note that in practice "num_const" or
   179 "num_position" etc. are mainly used instead (which also include
   180 position information via constraints).
   181 
   182 * Simplified configuration options for syntax ambiguity: see
   183 "syntax_ambiguity_warning" and "syntax_ambiguity_limit" in isar-ref
   184 manual.  Minor INCOMPATIBILITY.
   185 
   186 * Discontinued configuration option "syntax_positions": atomic terms
   187 in parse trees are always annotated by position constraints.
   188 
   189 * Old code generator for SML and its commands 'code_module',
   190 'code_library', 'consts_code', 'types_code' have been discontinued.
   191 Use commands of the generic code generator instead.  INCOMPATIBILITY.
   192 
   193 * Redundant attribute "code_inline" has been discontinued. Use
   194 "code_unfold" instead.  INCOMPATIBILITY.
   195 
   196 * Dropped attribute "code_unfold_post" in favor of the its dual
   197 "code_abbrev", which yields a common pattern in definitions like
   198 
   199   definition [code_abbrev]: "f = t"
   200 
   201 INCOMPATIBILITY.
   202 
   203 * Obsolete 'types' command has been discontinued.  Use 'type_synonym'
   204 instead.  INCOMPATIBILITY.
   205 
   206 * Discontinued old "prems" fact, which used to refer to the accidental
   207 collection of foundational premises in the context (already marked as
   208 legacy since Isabelle2011).
   209 
   210 
   211 *** HOL ***
   212 
   213 * Type 'a set is now a proper type constructor (just as before
   214 Isabelle2008).  Definitions mem_def and Collect_def have disappeared.
   215 Non-trivial INCOMPATIBILITY.  For developments keeping predicates and
   216 sets separate, it is often sufficient to rephrase some set S that has
   217 been accidentally used as predicates by "%x. x : S", and some
   218 predicate P that has been accidentally used as set by "{x. P x}".
   219 Corresponding proofs in a first step should be pruned from any
   220 tinkering with former theorems mem_def and Collect_def as far as
   221 possible.
   222 
   223 For developments which deliberately mix predicates and sets, a
   224 planning step is necessary to determine what should become a predicate
   225 and what a set.  It can be helpful to carry out that step in
   226 Isabelle2011-1 before jumping right into the current release.
   227 
   228 * Code generation by default implements sets as container type rather
   229 than predicates.  INCOMPATIBILITY.
   230 
   231 * New type synonym 'a rel = ('a * 'a) set
   232 
   233 * The representation of numerals has changed.  Datatype "num"
   234 represents strictly positive binary numerals, along with functions
   235 "numeral :: num => 'a" and "neg_numeral :: num => 'a" to represent
   236 positive and negated numeric literals, respectively.  See also
   237 definitions in ~~/src/HOL/Num.thy.  Potential INCOMPATIBILITY, some
   238 user theories may require adaptations as follows:
   239 
   240   - Theorems with number_ring or number_semiring constraints: These
   241     classes are gone; use comm_ring_1 or comm_semiring_1 instead.
   242 
   243   - Theories defining numeric types: Remove number, number_semiring,
   244     and number_ring instances. Defer all theorems about numerals until
   245     after classes one and semigroup_add have been instantiated.
   246 
   247   - Numeral-only simp rules: Replace each rule having a "number_of v"
   248     pattern with two copies, one for numeral and one for neg_numeral.
   249 
   250   - Theorems about subclasses of semiring_1 or ring_1: These classes
   251     automatically support numerals now, so more simp rules and
   252     simprocs may now apply within the proof.
   253 
   254   - Definitions and theorems using old constructors Pls/Min/Bit0/Bit1:
   255     Redefine using other integer operations.
   256 
   257 * Transfer: New package intended to generalize the existing
   258 "descending" method and related theorem attributes from the Quotient
   259 package.  (Not all functionality is implemented yet, but future
   260 development will focus on Transfer as an eventual replacement for the
   261 corresponding parts of the Quotient package.)
   262 
   263   - transfer_rule attribute: Maintains a collection of transfer rules,
   264     which relate constants at two different types. Transfer rules may
   265     relate different type instances of the same polymorphic constant,
   266     or they may relate an operation on a raw type to a corresponding
   267     operation on an abstract type (quotient or subtype). For example:
   268 
   269     ((A ===> B) ===> list_all2 A ===> list_all2 B) map map
   270     (cr_int ===> cr_int ===> cr_int) (%(x,y) (u,v). (x+u, y+v)) plus_int
   271 
   272   - transfer method: Replaces a subgoal on abstract types with an
   273     equivalent subgoal on the corresponding raw types. Constants are
   274     replaced with corresponding ones according to the transfer rules.
   275     Goals are generalized over all free variables by default; this is
   276     necessary for variables whose types change, but can be overridden
   277     for specific variables with e.g. "transfer fixing: x y z".  The
   278     variant transfer' method allows replacing a subgoal with one that
   279     is logically stronger (rather than equivalent).
   280 
   281   - relator_eq attribute: Collects identity laws for relators of
   282     various type constructors, e.g. "list_all2 (op =) = (op =)".  The
   283     transfer method uses these lemmas to infer transfer rules for
   284     non-polymorphic constants on the fly.
   285 
   286   - transfer_prover method: Assists with proving a transfer rule for a
   287     new constant, provided the constant is defined in terms of other
   288     constants that already have transfer rules. It should be applied
   289     after unfolding the constant definitions.
   290 
   291   - HOL/ex/Transfer_Int_Nat.thy: Example theory demonstrating transfer
   292     from type nat to type int.
   293 
   294 * Lifting: New package intended to generalize the quotient_definition
   295 facility of the Quotient package; designed to work with Transfer.
   296 
   297   - lift_definition command: Defines operations on an abstract type in
   298     terms of a corresponding operation on a representation
   299     type.  Example syntax:
   300 
   301     lift_definition dlist_insert :: "'a => 'a dlist => 'a dlist"
   302       is List.insert
   303 
   304     Users must discharge a respectfulness proof obligation when each
   305     constant is defined. (For a type copy, i.e. a typedef with UNIV,
   306     the proof is discharged automatically.) The obligation is
   307     presented in a user-friendly, readable form; a respectfulness
   308     theorem in the standard format and a transfer rule are generated
   309     by the package.
   310 
   311   - Integration with code_abstype: For typedefs (e.g. subtypes
   312     corresponding to a datatype invariant, such as dlist),
   313     lift_definition generates a code certificate theorem and sets up
   314     code generation for each constant.
   315 
   316   - setup_lifting command: Sets up the Lifting package to work with a
   317     user-defined type. The user must provide either a quotient theorem
   318     or a type_definition theorem.  The package configures transfer
   319     rules for equality and quantifiers on the type, and sets up the
   320     lift_definition command to work with the type.
   321 
   322   - Usage examples: See Quotient_Examples/Lift_DList.thy,
   323     Quotient_Examples/Lift_RBT.thy, Quotient_Examples/Lift_FSet.thy,
   324     Word/Word.thy and Library/Float.thy.
   325 
   326 * Quotient package:
   327 
   328   - The 'quotient_type' command now supports a 'morphisms' option with
   329     rep and abs functions, similar to typedef.
   330 
   331   - 'quotient_type' sets up new types to work with the Lifting and
   332     Transfer packages, as with 'setup_lifting'.
   333 
   334   - The 'quotient_definition' command now requires the user to prove a
   335     respectfulness property at the point where the constant is
   336     defined, similar to lift_definition; INCOMPATIBILITY.
   337 
   338   - Renamed predicate 'Quotient' to 'Quotient3', and renamed theorems
   339     accordingly, INCOMPATIBILITY.
   340 
   341 * New diagnostic command 'find_unused_assms' to find potentially
   342 superfluous assumptions in theorems using Quickcheck.
   343 
   344 * Quickcheck:
   345 
   346   - Quickcheck returns variable assignments as counterexamples, which
   347     allows to reveal the underspecification of functions under test.
   348     For example, refuting "hd xs = x", it presents the variable
   349     assignment xs = [] and x = a1 as a counterexample, assuming that
   350     any property is false whenever "hd []" occurs in it.
   351 
   352     These counterexample are marked as potentially spurious, as
   353     Quickcheck also returns "xs = []" as a counterexample to the
   354     obvious theorem "hd xs = hd xs".
   355 
   356     After finding a potentially spurious counterexample, Quickcheck
   357     continues searching for genuine ones.
   358 
   359     By default, Quickcheck shows potentially spurious and genuine
   360     counterexamples. The option "genuine_only" sets quickcheck to only
   361     show genuine counterexamples.
   362 
   363   - The command 'quickcheck_generator' creates random and exhaustive
   364     value generators for a given type and operations.
   365 
   366     It generates values by using the operations as if they were
   367     constructors of that type.
   368 
   369   - Support for multisets.
   370 
   371   - Added "use_subtype" options.
   372 
   373   - Added "quickcheck_locale" configuration to specify how to process
   374     conjectures in a locale context.
   375 
   376 * Nitpick: Fixed infinite loop caused by the 'peephole_optim' option
   377 and affecting 'rat' and 'real'.
   378 
   379 * Sledgehammer:
   380   - Integrated more tightly with SPASS, as described in the ITP 2012
   381     paper "More SPASS with Isabelle".
   382   - Made it try "smt" as a fallback if "metis" fails or times out.
   383   - Added support for the following provers: Alt-Ergo (via Why3 and
   384     TFF1), iProver, iProver-Eq.
   385   - Sped up the minimizer.
   386   - Added "lam_trans", "uncurry_aliases", and "minimize" options.
   387   - Renamed "slicing" ("no_slicing") option to "slice" ("dont_slice").
   388   - Renamed "sound" option to "strict".
   389 
   390 * Metis: Added possibility to specify lambda translations scheme as a
   391 parenthesized argument (e.g., "by (metis (lifting) ...)").
   392 
   393 * SMT: Renamed "smt_fixed" option to "smt_read_only_certificates".
   394 
   395 * Command 'try0': Renamed from 'try_methods'. INCOMPATIBILITY.
   396 
   397 * New "case_product" attribute to generate a case rule doing multiple
   398 case distinctions at the same time.  E.g.
   399 
   400   list.exhaust [case_product nat.exhaust]
   401 
   402 produces a rule which can be used to perform case distinction on both
   403 a list and a nat.
   404 
   405 * New "eventually_elim" method as a generalized variant of the
   406 eventually_elim* rules.  Supports structured proofs.
   407 
   408 * Typedef with implicit set definition is considered legacy.  Use
   409 "typedef (open)" form instead, which will eventually become the
   410 default.
   411 
   412 * Record: code generation can be switched off manually with
   413 
   414   declare [[record_coden = false]]  -- "default true"
   415 
   416 * Datatype: type parameters allow explicit sort constraints.
   417 
   418 * Concrete syntax for case expressions includes constraints for source
   419 positions, and thus produces Prover IDE markup for its bindings.
   420 INCOMPATIBILITY for old-style syntax translations that augment the
   421 pattern notation; e.g. see src/HOL/HOLCF/One.thy for translations of
   422 one_case.
   423 
   424 * Clarified attribute "mono_set": pure declaration without modifying
   425 the result of the fact expression.
   426 
   427 * More default pred/set conversions on a couple of relation operations
   428 and predicates.  Added powers of predicate relations.  Consolidation
   429 of some relation theorems:
   430 
   431   converse_def ~> converse_unfold
   432   rel_comp_def ~> relcomp_unfold
   433   symp_def ~> (modified, use symp_def and sym_def instead)
   434   transp_def ~> transp_trans
   435   Domain_def ~> Domain_unfold
   436   Range_def ~> Domain_converse [symmetric]
   437 
   438 Generalized theorems INF_INT_eq, INF_INT_eq2, SUP_UN_eq, SUP_UN_eq2.
   439 
   440 See theory "Relation" for examples for making use of pred/set
   441 conversions by means of attributes "to_set" and "to_pred".
   442 
   443 INCOMPATIBILITY.
   444 
   445 * Renamed facts about the power operation on relations, i.e., relpow
   446 to match the constant's name:
   447 
   448   rel_pow_1 ~> relpow_1
   449   rel_pow_0_I ~> relpow_0_I
   450   rel_pow_Suc_I ~> relpow_Suc_I
   451   rel_pow_Suc_I2 ~> relpow_Suc_I2
   452   rel_pow_0_E ~> relpow_0_E
   453   rel_pow_Suc_E ~> relpow_Suc_E
   454   rel_pow_E ~> relpow_E
   455   rel_pow_Suc_D2 ~> relpow_Suc_D2
   456   rel_pow_Suc_E2 ~> relpow_Suc_E2
   457   rel_pow_Suc_D2' ~> relpow_Suc_D2'
   458   rel_pow_E2 ~> relpow_E2
   459   rel_pow_add ~> relpow_add
   460   rel_pow_commute ~> relpow
   461   rel_pow_empty ~> relpow_empty:
   462   rtrancl_imp_UN_rel_pow ~> rtrancl_imp_UN_relpow
   463   rel_pow_imp_rtrancl ~> relpow_imp_rtrancl
   464   rtrancl_is_UN_rel_pow ~> rtrancl_is_UN_relpow
   465   rtrancl_imp_rel_pow ~> rtrancl_imp_relpow
   466   rel_pow_fun_conv ~> relpow_fun_conv
   467   rel_pow_finite_bounded1 ~> relpow_finite_bounded1
   468   rel_pow_finite_bounded ~> relpow_finite_bounded
   469   rtrancl_finite_eq_rel_pow ~> rtrancl_finite_eq_relpow
   470   trancl_finite_eq_rel_pow ~> trancl_finite_eq_relpow
   471   single_valued_rel_pow ~> single_valued_relpow
   472 
   473 INCOMPATIBILITY.
   474 
   475 * Theory Relation: Consolidated constant name for relation composition
   476 and corresponding theorem names:
   477 
   478   - Renamed constant rel_comp to relcomp.
   479 
   480   - Dropped abbreviation pred_comp. Use relcompp instead.
   481 
   482   - Renamed theorems:
   483 
   484     rel_compI ~> relcompI
   485     rel_compEpair ~> relcompEpair
   486     rel_compE ~> relcompE
   487     pred_comp_rel_comp_eq ~> relcompp_relcomp_eq
   488     rel_comp_empty1 ~> relcomp_empty1
   489     rel_comp_mono ~> relcomp_mono
   490     rel_comp_subset_Sigma ~> relcomp_subset_Sigma
   491     rel_comp_distrib ~> relcomp_distrib
   492     rel_comp_distrib2 ~> relcomp_distrib2
   493     rel_comp_UNION_distrib ~> relcomp_UNION_distrib
   494     rel_comp_UNION_distrib2 ~> relcomp_UNION_distrib2
   495     single_valued_rel_comp ~> single_valued_relcomp
   496     rel_comp_def ~> relcomp_unfold
   497     converse_rel_comp ~> converse_relcomp
   498     pred_compI ~> relcomppI
   499     pred_compE ~> relcomppE
   500     pred_comp_bot1 ~> relcompp_bot1
   501     pred_comp_bot2 ~> relcompp_bot2
   502     transp_pred_comp_less_eq ~> transp_relcompp_less_eq
   503     pred_comp_mono ~> relcompp_mono
   504     pred_comp_distrib ~> relcompp_distrib
   505     pred_comp_distrib2 ~> relcompp_distrib2
   506     converse_pred_comp ~> converse_relcompp
   507 
   508     finite_rel_comp ~> finite_relcomp
   509 
   510     set_rel_comp ~> set_relcomp
   511 
   512 INCOMPATIBILITY.
   513 
   514 * Theory Divides: Discontinued redundant theorems about div and mod.
   515 INCOMPATIBILITY, use the corresponding generic theorems instead.
   516 
   517   DIVISION_BY_ZERO ~> div_by_0, mod_by_0
   518   zdiv_self ~> div_self
   519   zmod_self ~> mod_self
   520   zdiv_zero ~> div_0
   521   zmod_zero ~> mod_0
   522   zdiv_zmod_equality ~> div_mod_equality2
   523   zdiv_zmod_equality2 ~> div_mod_equality
   524   zmod_zdiv_trivial ~> mod_div_trivial
   525   zdiv_zminus_zminus ~> div_minus_minus
   526   zmod_zminus_zminus ~> mod_minus_minus
   527   zdiv_zminus2 ~> div_minus_right
   528   zmod_zminus2 ~> mod_minus_right
   529   zdiv_minus1_right ~> div_minus1_right
   530   zmod_minus1_right ~> mod_minus1_right
   531   zdvd_mult_div_cancel ~> dvd_mult_div_cancel
   532   zmod_zmult1_eq ~> mod_mult_right_eq
   533   zpower_zmod ~> power_mod
   534   zdvd_zmod ~> dvd_mod
   535   zdvd_zmod_imp_zdvd ~> dvd_mod_imp_dvd
   536   mod_mult_distrib ~> mult_mod_left
   537   mod_mult_distrib2 ~> mult_mod_right
   538 
   539 * Removed redundant theorems nat_mult_2 and nat_mult_2_right; use
   540 generic mult_2 and mult_2_right instead. INCOMPATIBILITY.
   541 
   542 * Finite_Set.fold now qualified.  INCOMPATIBILITY.
   543 
   544 * Consolidated theorem names concerning fold combinators:
   545 
   546   inf_INFI_fold_inf ~> inf_INF_fold_inf
   547   sup_SUPR_fold_sup ~> sup_SUP_fold_sup
   548   INFI_fold_inf ~> INF_fold_inf
   549   SUPR_fold_sup ~> SUP_fold_sup
   550   union_set ~> union_set_fold
   551   minus_set ~> minus_set_fold
   552   INFI_set_fold ~> INF_set_fold
   553   SUPR_set_fold ~> SUP_set_fold
   554   INF_code ~> INF_set_foldr
   555   SUP_code ~> SUP_set_foldr
   556   foldr.simps ~> foldr.simps (in point-free formulation)
   557   foldr_fold_rev ~> foldr_conv_fold
   558   foldl_fold ~> foldl_conv_fold
   559   foldr_foldr ~> foldr_conv_foldl
   560   foldl_foldr ~> foldl_conv_foldr
   561   fold_set_remdups ~> fold_set_fold_remdups
   562   fold_set ~> fold_set_fold
   563   fold1_set ~> fold1_set_fold
   564 
   565 INCOMPATIBILITY.
   566 
   567 * Dropped rarely useful theorems concerning fold combinators:
   568 foldl_apply, foldl_fun_comm, foldl_rev, fold_weak_invariant,
   569 rev_foldl_cons, fold_set_remdups, fold_set, fold_set1,
   570 concat_conv_foldl, foldl_weak_invariant, foldl_invariant,
   571 foldr_invariant, foldl_absorb0, foldl_foldr1_lemma, foldl_foldr1,
   572 listsum_conv_fold, listsum_foldl, sort_foldl_insort, foldl_assoc,
   573 foldr_conv_foldl, start_le_sum, elem_le_sum, sum_eq_0_conv.
   574 INCOMPATIBILITY.  For the common phrases "%xs. List.foldr plus xs 0"
   575 and "List.foldl plus 0", prefer "List.listsum".  Otherwise it can be
   576 useful to boil down "List.foldr" and "List.foldl" to "List.fold" by
   577 unfolding "foldr_conv_fold" and "foldl_conv_fold".
   578 
   579 * Dropped lemmas minus_set_foldr, union_set_foldr, union_coset_foldr,
   580 inter_coset_foldr, Inf_fin_set_foldr, Sup_fin_set_foldr,
   581 Min_fin_set_foldr, Max_fin_set_foldr, Inf_set_foldr, Sup_set_foldr,
   582 INF_set_foldr, SUP_set_foldr.  INCOMPATIBILITY.  Prefer corresponding
   583 lemmas over fold rather than foldr, or make use of lemmas
   584 fold_conv_foldr and fold_rev.
   585 
   586 * Congruence rules Option.map_cong and Option.bind_cong for recursion
   587 through option types.
   588 
   589 * "Transitive_Closure.ntrancl": bounded transitive closure on
   590 relations.
   591 
   592 * Constant "Set.not_member" now qualified.  INCOMPATIBILITY.
   593 
   594 * Theory Int: Discontinued many legacy theorems specific to type int.
   595 INCOMPATIBILITY, use the corresponding generic theorems instead.
   596 
   597   zminus_zminus ~> minus_minus
   598   zminus_0 ~> minus_zero
   599   zminus_zadd_distrib ~> minus_add_distrib
   600   zadd_commute ~> add_commute
   601   zadd_assoc ~> add_assoc
   602   zadd_left_commute ~> add_left_commute
   603   zadd_ac ~> add_ac
   604   zmult_ac ~> mult_ac
   605   zadd_0 ~> add_0_left
   606   zadd_0_right ~> add_0_right
   607   zadd_zminus_inverse2 ~> left_minus
   608   zmult_zminus ~> mult_minus_left
   609   zmult_commute ~> mult_commute
   610   zmult_assoc ~> mult_assoc
   611   zadd_zmult_distrib ~> left_distrib
   612   zadd_zmult_distrib2 ~> right_distrib
   613   zdiff_zmult_distrib ~> left_diff_distrib
   614   zdiff_zmult_distrib2 ~> right_diff_distrib
   615   zmult_1 ~> mult_1_left
   616   zmult_1_right ~> mult_1_right
   617   zle_refl ~> order_refl
   618   zle_trans ~> order_trans
   619   zle_antisym ~> order_antisym
   620   zle_linear ~> linorder_linear
   621   zless_linear ~> linorder_less_linear
   622   zadd_left_mono ~> add_left_mono
   623   zadd_strict_right_mono ~> add_strict_right_mono
   624   zadd_zless_mono ~> add_less_le_mono
   625   int_0_less_1 ~> zero_less_one
   626   int_0_neq_1 ~> zero_neq_one
   627   zless_le ~> less_le
   628   zpower_zadd_distrib ~> power_add
   629   zero_less_zpower_abs_iff ~> zero_less_power_abs_iff
   630   zero_le_zpower_abs ~> zero_le_power_abs
   631 
   632 * Theory Deriv: Renamed
   633 
   634   DERIV_nonneg_imp_nonincreasing ~> DERIV_nonneg_imp_nondecreasing
   635 
   636 * Theory Library/Multiset: Improved code generation of multisets.
   637 
   638 * Theory HOL/Library/Set_Algebras: Addition and multiplication on sets
   639 are expressed via type classes again. The special syntax
   640 \<oplus>/\<otimes> has been replaced by plain +/*. Removed constant
   641 setsum_set, which is now subsumed by Big_Operators.setsum.
   642 INCOMPATIBILITY.
   643 
   644 * Theory HOL/Library/Diagonalize has been removed. INCOMPATIBILITY,
   645 use theory HOL/Library/Nat_Bijection instead.
   646 
   647 * Theory HOL/Library/RBT_Impl: Backing implementation of red-black
   648 trees is now inside a type class context.  Names of affected
   649 operations and lemmas have been prefixed by rbt_.  INCOMPATIBILITY for
   650 theories working directly with raw red-black trees, adapt the names as
   651 follows:
   652 
   653   Operations:
   654   bulkload -> rbt_bulkload
   655   del_from_left -> rbt_del_from_left
   656   del_from_right -> rbt_del_from_right
   657   del -> rbt_del
   658   delete -> rbt_delete
   659   ins -> rbt_ins
   660   insert -> rbt_insert
   661   insertw -> rbt_insert_with
   662   insert_with_key -> rbt_insert_with_key
   663   map_entry -> rbt_map_entry
   664   lookup -> rbt_lookup
   665   sorted -> rbt_sorted
   666   tree_greater -> rbt_greater
   667   tree_less -> rbt_less
   668   tree_less_symbol -> rbt_less_symbol
   669   union -> rbt_union
   670   union_with -> rbt_union_with
   671   union_with_key -> rbt_union_with_key
   672 
   673   Lemmas:
   674   balance_left_sorted -> balance_left_rbt_sorted
   675   balance_left_tree_greater -> balance_left_rbt_greater
   676   balance_left_tree_less -> balance_left_rbt_less
   677   balance_right_sorted -> balance_right_rbt_sorted
   678   balance_right_tree_greater -> balance_right_rbt_greater
   679   balance_right_tree_less -> balance_right_rbt_less
   680   balance_sorted -> balance_rbt_sorted
   681   balance_tree_greater -> balance_rbt_greater
   682   balance_tree_less -> balance_rbt_less
   683   bulkload_is_rbt -> rbt_bulkload_is_rbt
   684   combine_sorted -> combine_rbt_sorted
   685   combine_tree_greater -> combine_rbt_greater
   686   combine_tree_less -> combine_rbt_less
   687   delete_in_tree -> rbt_delete_in_tree
   688   delete_is_rbt -> rbt_delete_is_rbt
   689   del_from_left_tree_greater -> rbt_del_from_left_rbt_greater
   690   del_from_left_tree_less -> rbt_del_from_left_rbt_less
   691   del_from_right_tree_greater -> rbt_del_from_right_rbt_greater
   692   del_from_right_tree_less -> rbt_del_from_right_rbt_less
   693   del_in_tree -> rbt_del_in_tree
   694   del_inv1_inv2 -> rbt_del_inv1_inv2
   695   del_sorted -> rbt_del_rbt_sorted
   696   del_tree_greater -> rbt_del_rbt_greater
   697   del_tree_less -> rbt_del_rbt_less
   698   dom_lookup_Branch -> dom_rbt_lookup_Branch
   699   entries_lookup -> entries_rbt_lookup
   700   finite_dom_lookup -> finite_dom_rbt_lookup
   701   insert_sorted -> rbt_insert_rbt_sorted
   702   insertw_is_rbt -> rbt_insertw_is_rbt
   703   insertwk_is_rbt -> rbt_insertwk_is_rbt
   704   insertwk_sorted -> rbt_insertwk_rbt_sorted
   705   insertw_sorted -> rbt_insertw_rbt_sorted
   706   ins_sorted -> ins_rbt_sorted
   707   ins_tree_greater -> ins_rbt_greater
   708   ins_tree_less -> ins_rbt_less
   709   is_rbt_sorted -> is_rbt_rbt_sorted
   710   lookup_balance -> rbt_lookup_balance
   711   lookup_bulkload -> rbt_lookup_rbt_bulkload
   712   lookup_delete -> rbt_lookup_rbt_delete
   713   lookup_Empty -> rbt_lookup_Empty
   714   lookup_from_in_tree -> rbt_lookup_from_in_tree
   715   lookup_in_tree -> rbt_lookup_in_tree
   716   lookup_ins -> rbt_lookup_ins
   717   lookup_insert -> rbt_lookup_rbt_insert
   718   lookup_insertw -> rbt_lookup_rbt_insertw
   719   lookup_insertwk -> rbt_lookup_rbt_insertwk
   720   lookup_keys -> rbt_lookup_keys
   721   lookup_map -> rbt_lookup_map
   722   lookup_map_entry -> rbt_lookup_rbt_map_entry
   723   lookup_tree_greater -> rbt_lookup_rbt_greater
   724   lookup_tree_less -> rbt_lookup_rbt_less
   725   lookup_union -> rbt_lookup_rbt_union
   726   map_entry_color_of -> rbt_map_entry_color_of
   727   map_entry_inv1 -> rbt_map_entry_inv1
   728   map_entry_inv2 -> rbt_map_entry_inv2
   729   map_entry_is_rbt -> rbt_map_entry_is_rbt
   730   map_entry_sorted -> rbt_map_entry_rbt_sorted
   731   map_entry_tree_greater -> rbt_map_entry_rbt_greater
   732   map_entry_tree_less -> rbt_map_entry_rbt_less
   733   map_tree_greater -> map_rbt_greater
   734   map_tree_less -> map_rbt_less
   735   map_sorted -> map_rbt_sorted
   736   paint_sorted -> paint_rbt_sorted
   737   paint_lookup -> paint_rbt_lookup
   738   paint_tree_greater -> paint_rbt_greater
   739   paint_tree_less -> paint_rbt_less
   740   sorted_entries -> rbt_sorted_entries
   741   tree_greater_eq_trans -> rbt_greater_eq_trans
   742   tree_greater_nit -> rbt_greater_nit
   743   tree_greater_prop -> rbt_greater_prop
   744   tree_greater_simps -> rbt_greater_simps
   745   tree_greater_trans -> rbt_greater_trans
   746   tree_less_eq_trans -> rbt_less_eq_trans
   747   tree_less_nit -> rbt_less_nit
   748   tree_less_prop -> rbt_less_prop
   749   tree_less_simps -> rbt_less_simps
   750   tree_less_trans -> rbt_less_trans
   751   tree_ord_props -> rbt_ord_props
   752   union_Branch -> rbt_union_Branch
   753   union_is_rbt -> rbt_union_is_rbt
   754   unionw_is_rbt -> rbt_unionw_is_rbt
   755   unionwk_is_rbt -> rbt_unionwk_is_rbt
   756   unionwk_sorted -> rbt_unionwk_rbt_sorted
   757 
   758 * Theory HOL/Library/Float: Floating point numbers are now defined as
   759 a subset of the real numbers.  All operations are defined using the
   760 lifing-framework and proofs use the transfer method.  INCOMPATIBILITY.
   761 
   762   Changed Operations:
   763   float_abs -> abs
   764   float_nprt -> nprt
   765   float_pprt -> pprt
   766   pow2 -> use powr
   767   round_down -> float_round_down
   768   round_up -> float_round_up
   769   scale -> exponent
   770 
   771   Removed Operations:
   772   ceiling_fl, lb_mult, lb_mod, ub_mult, ub_mod
   773 
   774   Renamed Lemmas:
   775   abs_float_def -> Float.compute_float_abs
   776   bitlen_ge0 -> bitlen_nonneg
   777   bitlen.simps -> Float.compute_bitlen
   778   float_components -> Float_mantissa_exponent
   779   float_divl.simps -> Float.compute_float_divl
   780   float_divr.simps -> Float.compute_float_divr
   781   float_eq_odd -> mult_powr_eq_mult_powr_iff
   782   float_power -> real_of_float_power
   783   lapprox_posrat_def -> Float.compute_lapprox_posrat
   784   lapprox_rat.simps -> Float.compute_lapprox_rat
   785   le_float_def' -> Float.compute_float_le
   786   le_float_def -> less_eq_float.rep_eq
   787   less_float_def' -> Float.compute_float_less
   788   less_float_def -> less_float.rep_eq
   789   normfloat_def -> Float.compute_normfloat
   790   normfloat_imp_odd_or_zero -> mantissa_not_dvd and mantissa_noteq_0
   791   normfloat -> normfloat_def
   792   normfloat_unique -> use normfloat_def
   793   number_of_float_Float -> Float.compute_float_numeral, Float.compute_float_neg_numeral
   794   one_float_def -> Float.compute_float_one
   795   plus_float_def -> Float.compute_float_plus
   796   rapprox_posrat_def -> Float.compute_rapprox_posrat
   797   rapprox_rat.simps -> Float.compute_rapprox_rat
   798   real_of_float_0 -> zero_float.rep_eq
   799   real_of_float_1 -> one_float.rep_eq
   800   real_of_float_abs -> abs_float.rep_eq
   801   real_of_float_add -> plus_float.rep_eq
   802   real_of_float_minus -> uminus_float.rep_eq
   803   real_of_float_mult -> times_float.rep_eq
   804   real_of_float_simp -> Float.rep_eq
   805   real_of_float_sub -> minus_float.rep_eq
   806   round_down.simps -> Float.compute_float_round_down
   807   round_up.simps -> Float.compute_float_round_up
   808   times_float_def -> Float.compute_float_times
   809   uminus_float_def -> Float.compute_float_uminus
   810   zero_float_def -> Float.compute_float_zero
   811 
   812   Lemmas not necessary anymore, use the transfer method:
   813   bitlen_B0, bitlen_B1, bitlen_ge1, bitlen_Min, bitlen_Pls, float_divl,
   814   float_divr, float_le_simp, float_less1_mantissa_bound,
   815   float_less_simp, float_less_zero, float_le_zero,
   816   float_pos_less1_e_neg, float_pos_m_pos, float_split, float_split2,
   817   floor_pos_exp, lapprox_posrat, lapprox_posrat_bottom, lapprox_rat,
   818   lapprox_rat_bottom, normalized_float, rapprox_posrat,
   819   rapprox_posrat_le1, rapprox_rat, real_of_float_ge0_exp,
   820   real_of_float_neg_exp, real_of_float_nge0_exp, round_down floor_fl,
   821   round_up, zero_le_float, zero_less_float
   822 
   823 * New theory HOL/Library/DAList provides an abstract type for
   824 association lists with distinct keys.
   825 
   826 * Session HOL/IMP: Added new theory of abstract interpretation of
   827 annotated commands.
   828 
   829 * Session HOL-Import: Re-implementation from scratch is faster,
   830 simpler, and more scalable.  Requires a proof bundle, which is
   831 available as an external component.  Discontinued old (and mostly
   832 dead) Importer for HOL4 and HOL Light.  INCOMPATIBILITY.
   833 
   834 * Session HOL-Word: Discontinued many redundant theorems specific to
   835 type 'a word. INCOMPATIBILITY, use the corresponding generic theorems
   836 instead.
   837 
   838   word_sub_alt ~> word_sub_wi
   839   word_add_alt ~> word_add_def
   840   word_mult_alt ~> word_mult_def
   841   word_minus_alt ~> word_minus_def
   842   word_0_alt ~> word_0_wi
   843   word_1_alt ~> word_1_wi
   844   word_add_0 ~> add_0_left
   845   word_add_0_right ~> add_0_right
   846   word_mult_1 ~> mult_1_left
   847   word_mult_1_right ~> mult_1_right
   848   word_add_commute ~> add_commute
   849   word_add_assoc ~> add_assoc
   850   word_add_left_commute ~> add_left_commute
   851   word_mult_commute ~> mult_commute
   852   word_mult_assoc ~> mult_assoc
   853   word_mult_left_commute ~> mult_left_commute
   854   word_left_distrib ~> left_distrib
   855   word_right_distrib ~> right_distrib
   856   word_left_minus ~> left_minus
   857   word_diff_0_right ~> diff_0_right
   858   word_diff_self ~> diff_self
   859   word_sub_def ~> diff_minus
   860   word_diff_minus ~> diff_minus
   861   word_add_ac ~> add_ac
   862   word_mult_ac ~> mult_ac
   863   word_plus_ac0 ~> add_0_left add_0_right add_ac
   864   word_times_ac1 ~> mult_1_left mult_1_right mult_ac
   865   word_order_trans ~> order_trans
   866   word_order_refl ~> order_refl
   867   word_order_antisym ~> order_antisym
   868   word_order_linear ~> linorder_linear
   869   lenw1_zero_neq_one ~> zero_neq_one
   870   word_number_of_eq ~> number_of_eq
   871   word_of_int_add_hom ~> wi_hom_add
   872   word_of_int_sub_hom ~> wi_hom_sub
   873   word_of_int_mult_hom ~> wi_hom_mult
   874   word_of_int_minus_hom ~> wi_hom_neg
   875   word_of_int_succ_hom ~> wi_hom_succ
   876   word_of_int_pred_hom ~> wi_hom_pred
   877   word_of_int_0_hom ~> word_0_wi
   878   word_of_int_1_hom ~> word_1_wi
   879 
   880 * Session HOL-Word: New proof method "word_bitwise" for splitting
   881 machine word equalities and inequalities into logical circuits,
   882 defined in HOL/Word/WordBitwise.thy.  Supports addition, subtraction,
   883 multiplication, shifting by constants, bitwise operators and numeric
   884 constants.  Requires fixed-length word types, not 'a word.  Solves
   885 many standard word identities outright and converts more into first
   886 order problems amenable to blast or similar.  See also examples in
   887 HOL/Word/Examples/WordExamples.thy.
   888 
   889 * Session HOL-Probability: Introduced the type "'a measure" to
   890 represent measures, this replaces the records 'a algebra and 'a
   891 measure_space.  The locales based on subset_class now have two
   892 locale-parameters the space \<Omega> and the set of measurable sets M.
   893 The product of probability spaces uses now the same constant as the
   894 finite product of sigma-finite measure spaces "PiM :: ('i => 'a)
   895 measure".  Most constants are defined now outside of locales and gain
   896 an additional parameter, like null_sets, almost_eventually or \<mu>'.
   897 Measure space constructions for distributions and densities now got
   898 their own constants distr and density.  Instead of using locales to
   899 describe measure spaces with a finite space, the measure count_space
   900 and point_measure is introduced.  INCOMPATIBILITY.
   901 
   902   Renamed constants:
   903   measure -> emeasure
   904   finite_measure.\<mu>' -> measure
   905   product_algebra_generator -> prod_algebra
   906   product_prob_space.emb -> prod_emb
   907   product_prob_space.infprod_algebra -> PiM
   908 
   909   Removed locales:
   910   completeable_measure_space
   911   finite_measure_space
   912   finite_prob_space
   913   finite_product_finite_prob_space
   914   finite_product_sigma_algebra
   915   finite_sigma_algebra
   916   measure_space
   917   pair_finite_prob_space
   918   pair_finite_sigma_algebra
   919   pair_finite_space
   920   pair_sigma_algebra
   921   product_sigma_algebra
   922 
   923   Removed constants:
   924   conditional_space
   925   distribution -> use distr measure, or distributed predicate
   926   image_space
   927   joint_distribution -> use distr measure, or distributed predicate
   928   pair_measure_generator
   929   product_prob_space.infprod_algebra -> use PiM
   930   subvimage
   931 
   932   Replacement theorems:
   933   finite_additivity_sufficient -> ring_of_sets.countably_additiveI_finite
   934   finite_measure.empty_measure -> measure_empty
   935   finite_measure.finite_continuity_from_above -> finite_measure.finite_Lim_measure_decseq
   936   finite_measure.finite_continuity_from_below -> finite_measure.finite_Lim_measure_incseq
   937   finite_measure.finite_measure_countably_subadditive -> finite_measure.finite_measure_subadditive_countably
   938   finite_measure.finite_measure_eq -> finite_measure.emeasure_eq_measure
   939   finite_measure.finite_measure -> finite_measure.emeasure_finite
   940   finite_measure.finite_measure_finite_singleton -> finite_measure.finite_measure_eq_setsum_singleton
   941   finite_measure.positive_measure' -> measure_nonneg
   942   finite_measure.real_measure -> finite_measure.emeasure_real
   943   finite_product_prob_space.finite_measure_times -> finite_product_prob_space.finite_measure_PiM_emb
   944   finite_product_sigma_algebra.in_P -> sets_PiM_I_finite
   945   finite_product_sigma_algebra.P_empty -> space_PiM_empty, sets_PiM_empty
   946   information_space.conditional_entropy_eq -> information_space.conditional_entropy_simple_distributed
   947   information_space.conditional_entropy_positive -> information_space.conditional_entropy_nonneg_simple
   948   information_space.conditional_mutual_information_eq_mutual_information -> information_space.conditional_mutual_information_eq_mutual_information_simple
   949   information_space.conditional_mutual_information_generic_positive -> information_space.conditional_mutual_information_nonneg_simple
   950   information_space.conditional_mutual_information_positive -> information_space.conditional_mutual_information_nonneg_simple
   951   information_space.entropy_commute -> information_space.entropy_commute_simple
   952   information_space.entropy_eq -> information_space.entropy_simple_distributed
   953   information_space.entropy_generic_eq -> information_space.entropy_simple_distributed
   954   information_space.entropy_positive -> information_space.entropy_nonneg_simple
   955   information_space.entropy_uniform_max -> information_space.entropy_uniform
   956   information_space.KL_eq_0_imp -> information_space.KL_eq_0_iff_eq
   957   information_space.KL_eq_0 -> information_space.KL_same_eq_0
   958   information_space.KL_ge_0 -> information_space.KL_nonneg
   959   information_space.mutual_information_eq -> information_space.mutual_information_simple_distributed
   960   information_space.mutual_information_positive -> information_space.mutual_information_nonneg_simple
   961   Int_stable_cuboids -> Int_stable_atLeastAtMost
   962   Int_stable_product_algebra_generator -> positive_integral
   963   measure_preserving -> equality "distr M N f = N" "f : measurable M N"
   964   measure_space.additive -> emeasure_additive
   965   measure_space.AE_iff_null_set -> AE_iff_null
   966   measure_space.almost_everywhere_def -> eventually_ae_filter
   967   measure_space.almost_everywhere_vimage -> AE_distrD
   968   measure_space.continuity_from_above -> INF_emeasure_decseq
   969   measure_space.continuity_from_above_Lim -> Lim_emeasure_decseq
   970   measure_space.continuity_from_below_Lim -> Lim_emeasure_incseq
   971   measure_space.continuity_from_below -> SUP_emeasure_incseq
   972   measure_space_density -> emeasure_density
   973   measure_space.density_is_absolutely_continuous -> absolutely_continuousI_density
   974   measure_space.integrable_vimage -> integrable_distr
   975   measure_space.integral_translated_density -> integral_density
   976   measure_space.integral_vimage -> integral_distr
   977   measure_space.measure_additive -> plus_emeasure
   978   measure_space.measure_compl -> emeasure_compl
   979   measure_space.measure_countable_increasing -> emeasure_countable_increasing
   980   measure_space.measure_countably_subadditive -> emeasure_subadditive_countably
   981   measure_space.measure_decseq -> decseq_emeasure
   982   measure_space.measure_Diff -> emeasure_Diff
   983   measure_space.measure_Diff_null_set -> emeasure_Diff_null_set
   984   measure_space.measure_eq_0 -> emeasure_eq_0
   985   measure_space.measure_finitely_subadditive -> emeasure_subadditive_finite
   986   measure_space.measure_finite_singleton -> emeasure_eq_setsum_singleton
   987   measure_space.measure_incseq -> incseq_emeasure
   988   measure_space.measure_insert -> emeasure_insert
   989   measure_space.measure_mono -> emeasure_mono
   990   measure_space.measure_not_negative -> emeasure_not_MInf
   991   measure_space.measure_preserving_Int_stable -> measure_eqI_generator_eq
   992   measure_space.measure_setsum -> setsum_emeasure
   993   measure_space.measure_setsum_split -> setsum_emeasure_cover
   994   measure_space.measure_space_vimage -> emeasure_distr
   995   measure_space.measure_subadditive_finite -> emeasure_subadditive_finite
   996   measure_space.measure_subadditive -> subadditive
   997   measure_space.measure_top -> emeasure_space
   998   measure_space.measure_UN_eq_0 -> emeasure_UN_eq_0
   999   measure_space.measure_Un_null_set -> emeasure_Un_null_set
  1000   measure_space.positive_integral_translated_density -> positive_integral_density
  1001   measure_space.positive_integral_vimage -> positive_integral_distr
  1002   measure_space.real_continuity_from_above -> Lim_measure_decseq
  1003   measure_space.real_continuity_from_below -> Lim_measure_incseq
  1004   measure_space.real_measure_countably_subadditive -> measure_subadditive_countably
  1005   measure_space.real_measure_Diff -> measure_Diff
  1006   measure_space.real_measure_finite_Union -> measure_finite_Union
  1007   measure_space.real_measure_setsum_singleton -> measure_eq_setsum_singleton
  1008   measure_space.real_measure_subadditive -> measure_subadditive
  1009   measure_space.real_measure_Union -> measure_Union
  1010   measure_space.real_measure_UNION -> measure_UNION
  1011   measure_space.simple_function_vimage -> simple_function_comp
  1012   measure_space.simple_integral_vimage -> simple_integral_distr
  1013   measure_space.simple_integral_vimage -> simple_integral_distr
  1014   measure_unique_Int_stable -> measure_eqI_generator_eq
  1015   measure_unique_Int_stable_vimage -> measure_eqI_generator_eq
  1016   pair_sigma_algebra.measurable_cut_fst -> sets_Pair1
  1017   pair_sigma_algebra.measurable_cut_snd -> sets_Pair2
  1018   pair_sigma_algebra.measurable_pair_image_fst -> measurable_Pair1
  1019   pair_sigma_algebra.measurable_pair_image_snd -> measurable_Pair2
  1020   pair_sigma_algebra.measurable_product_swap -> measurable_pair_swap_iff
  1021   pair_sigma_algebra.pair_sigma_algebra_measurable -> measurable_pair_swap
  1022   pair_sigma_algebra.pair_sigma_algebra_swap_measurable -> measurable_pair_swap'
  1023   pair_sigma_algebra.sets_swap -> sets_pair_swap
  1024   pair_sigma_finite.measure_cut_measurable_fst -> pair_sigma_finite.measurable_emeasure_Pair1
  1025   pair_sigma_finite.measure_cut_measurable_snd -> pair_sigma_finite.measurable_emeasure_Pair2
  1026   pair_sigma_finite.measure_preserving_swap -> pair_sigma_finite.distr_pair_swap
  1027   pair_sigma_finite.pair_measure_alt2 -> pair_sigma_finite.emeasure_pair_measure_alt2
  1028   pair_sigma_finite.pair_measure_alt -> pair_sigma_finite.emeasure_pair_measure_alt
  1029   pair_sigma_finite.pair_measure_times -> pair_sigma_finite.emeasure_pair_measure_Times
  1030   prob_space.indep_distribution_eq_measure -> prob_space.indep_vars_iff_distr_eq_PiM
  1031   prob_space.indep_var_distributionD -> prob_space.indep_var_distribution_eq
  1032   prob_space.measure_space_1 -> prob_space.emeasure_space_1
  1033   prob_space.prob_space_vimage -> prob_space_distr
  1034   prob_space.random_variable_restrict -> measurable_restrict
  1035   prob_space_unique_Int_stable -> measure_eqI_prob_space
  1036   product_algebraE -> prod_algebraE_all
  1037   product_algebra_generator_der -> prod_algebra_eq_finite
  1038   product_algebra_generator_into_space -> prod_algebra_sets_into_space
  1039   product_algebraI -> sets_PiM_I_finite
  1040   product_measure_exists -> product_sigma_finite.sigma_finite
  1041   product_prob_space.finite_index_eq_finite_product -> product_prob_space.sets_PiM_generator
  1042   product_prob_space.finite_measure_infprod_emb_Pi -> product_prob_space.measure_PiM_emb
  1043   product_prob_space.infprod_spec -> product_prob_space.emeasure_PiM_emb_not_empty
  1044   product_prob_space.measurable_component -> measurable_component_singleton
  1045   product_prob_space.measurable_emb -> measurable_prod_emb
  1046   product_prob_space.measurable_into_infprod_algebra -> measurable_PiM_single
  1047   product_prob_space.measurable_singleton_infprod -> measurable_component_singleton
  1048   product_prob_space.measure_emb -> emeasure_prod_emb
  1049   product_prob_space.measure_preserving_restrict -> product_prob_space.distr_restrict
  1050   product_sigma_algebra.product_algebra_into_space -> space_closed
  1051   product_sigma_finite.measure_fold -> product_sigma_finite.distr_merge
  1052   product_sigma_finite.measure_preserving_component_singelton -> product_sigma_finite.distr_singleton
  1053   product_sigma_finite.measure_preserving_merge -> product_sigma_finite.distr_merge
  1054   sequence_space.measure_infprod -> sequence_space.measure_PiM_countable
  1055   sets_product_algebra -> sets_PiM
  1056   sigma_algebra.measurable_sigma -> measurable_measure_of
  1057   sigma_finite_measure.disjoint_sigma_finite -> sigma_finite_disjoint
  1058   sigma_finite_measure.RN_deriv_vimage -> sigma_finite_measure.RN_deriv_distr
  1059   sigma_product_algebra_sigma_eq -> sigma_prod_algebra_sigma_eq
  1060   space_product_algebra -> space_PiM
  1061 
  1062 * Session HOL-TPTP: support to parse and import TPTP problems (all
  1063 languages) into Isabelle/HOL.
  1064 
  1065 
  1066 *** FOL ***
  1067 
  1068 * New "case_product" attribute (see HOL).
  1069 
  1070 
  1071 *** ZF ***
  1072 
  1073 * Greater support for structured proofs involving induction or case
  1074 analysis.
  1075 
  1076 * Much greater use of mathematical symbols.
  1077 
  1078 * Removal of many ML theorem bindings.  INCOMPATIBILITY.
  1079 
  1080 
  1081 *** ML ***
  1082 
  1083 * Antiquotation @{keyword "name"} produces a parser for outer syntax
  1084 from a minor keyword introduced via theory header declaration.
  1085 
  1086 * Antiquotation @{command_spec "name"} produces the
  1087 Outer_Syntax.command_spec from a major keyword introduced via theory
  1088 header declaration; it can be passed to Outer_Syntax.command etc.
  1089 
  1090 * Local_Theory.define no longer hard-wires default theorem name
  1091 "foo_def", but retains the binding as given.  If that is Binding.empty
  1092 / Attrib.empty_binding, the result is not registered as user-level
  1093 fact.  The Local_Theory.define_internal variant allows to specify a
  1094 non-empty name (used for the foundation in the background theory),
  1095 while omitting the fact binding in the user-context.  Potential
  1096 INCOMPATIBILITY for derived definitional packages: need to specify
  1097 naming policy for primitive definitions more explicitly.
  1098 
  1099 * Renamed Thm.capply to Thm.apply, and Thm.cabs to Thm.lambda in
  1100 conformance with similar operations in structure Term and Logic.
  1101 
  1102 * Antiquotation @{attributes [...]} embeds attribute source
  1103 representation into the ML text, which is particularly useful with
  1104 declarations like Local_Theory.note.
  1105 
  1106 * Structure Proof_Context follows standard naming scheme.  Old
  1107 ProofContext has been discontinued.  INCOMPATIBILITY.
  1108 
  1109 * Refined Local_Theory.declaration {syntax, pervasive}, with subtle
  1110 change of semantics: update is applied to auxiliary local theory
  1111 context as well.
  1112 
  1113 * Modernized some old-style infix operations:
  1114 
  1115   addeqcongs    ~> Simplifier.add_eqcong
  1116   deleqcongs    ~> Simplifier.del_eqcong
  1117   addcongs      ~> Simplifier.add_cong
  1118   delcongs      ~> Simplifier.del_cong
  1119   setmksimps    ~> Simplifier.set_mksimps
  1120   setmkcong     ~> Simplifier.set_mkcong
  1121   setmksym      ~> Simplifier.set_mksym
  1122   setmkeqTrue   ~> Simplifier.set_mkeqTrue
  1123   settermless   ~> Simplifier.set_termless
  1124   setsubgoaler  ~> Simplifier.set_subgoaler
  1125   addsplits     ~> Splitter.add_split
  1126   delsplits     ~> Splitter.del_split
  1127 
  1128 
  1129 *** System ***
  1130 
  1131 * USER_HOME settings variable points to cross-platform user home
  1132 directory, which coincides with HOME on POSIX systems only.  Likewise,
  1133 the Isabelle path specification "~" now expands to $USER_HOME, instead
  1134 of former $HOME.  A different default for USER_HOME may be set
  1135 explicitly in shell environment, before Isabelle settings are
  1136 evaluated.  Minor INCOMPATIBILITY: need to adapt Isabelle path where
  1137 the generic user home was intended.
  1138 
  1139 * ISABELLE_HOME_WINDOWS refers to ISABELLE_HOME in windows file name
  1140 notation, which is useful for the jEdit file browser, for example.
  1141 
  1142 * ISABELLE_JDK_HOME settings variable points to JDK with javac and jar
  1143 (not just JRE).
  1144 
  1145 
  1146 
  1147 New in Isabelle2011-1 (October 2011)
  1148 ------------------------------------
  1149 
  1150 *** General ***
  1151 
  1152 * Improved Isabelle/jEdit Prover IDE (PIDE), which can be invoked as
  1153 "isabelle jedit" or "ISABELLE_HOME/Isabelle" on the command line.
  1154 
  1155   - Management of multiple theory files directly from the editor
  1156     buffer store -- bypassing the file-system (no requirement to save
  1157     files for checking).
  1158 
  1159   - Markup of formal entities within the text buffer, with semantic
  1160     highlighting, tooltips and hyperlinks to jump to defining source
  1161     positions.
  1162 
  1163   - Improved text rendering, with sub/superscripts in the source
  1164     buffer (including support for copy/paste wrt. output panel, HTML
  1165     theory output and other non-Isabelle text boxes).
  1166 
  1167   - Refined scheduling of proof checking and printing of results,
  1168     based on interactive editor view.  (Note: jEdit folding and
  1169     narrowing allows to restrict buffer perspectives explicitly.)
  1170 
  1171   - Reduced CPU performance requirements, usable on machines with few
  1172     cores.
  1173 
  1174   - Reduced memory requirements due to pruning of unused document
  1175     versions (garbage collection).
  1176 
  1177 See also ~~/src/Tools/jEdit/README.html for further information,
  1178 including some remaining limitations.
  1179 
  1180 * Theory loader: source files are exclusively located via the master
  1181 directory of each theory node (where the .thy file itself resides).
  1182 The global load path (such as src/HOL/Library) has been discontinued.
  1183 Note that the path element ~~ may be used to reference theories in the
  1184 Isabelle home folder -- for instance, "~~/src/HOL/Library/FuncSet".
  1185 INCOMPATIBILITY.
  1186 
  1187 * Theory loader: source files are identified by content via SHA1
  1188 digests.  Discontinued former path/modtime identification and optional
  1189 ISABELLE_FILE_IDENT plugin scripts.
  1190 
  1191 * Parallelization of nested Isar proofs is subject to
  1192 Goal.parallel_proofs_threshold (default 100).  See also isabelle
  1193 usedir option -Q.
  1194 
  1195 * Name space: former unsynchronized references are now proper
  1196 configuration options, with more conventional names:
  1197 
  1198   long_names   ~> names_long
  1199   short_names  ~> names_short
  1200   unique_names ~> names_unique
  1201 
  1202 Minor INCOMPATIBILITY, need to declare options in context like this:
  1203 
  1204   declare [[names_unique = false]]
  1205 
  1206 * Literal facts `prop` may contain dummy patterns, e.g. `_ = _`.  Note
  1207 that the result needs to be unique, which means fact specifications
  1208 may have to be refined after enriching a proof context.
  1209 
  1210 * Attribute "case_names" has been refined: the assumptions in each case
  1211 can be named now by following the case name with [name1 name2 ...].
  1212 
  1213 * Isabelle/Isar reference manual has been updated and extended:
  1214   - "Synopsis" provides a catalog of main Isar language concepts.
  1215   - Formal references in syntax diagrams, via @{rail} antiquotation.
  1216   - Updated material from classic "ref" manual, notably about
  1217     "Classical Reasoner".
  1218 
  1219 
  1220 *** HOL ***
  1221 
  1222 * Class bot and top require underlying partial order rather than
  1223 preorder: uniqueness of bot and top is guaranteed.  INCOMPATIBILITY.
  1224 
  1225 * Class complete_lattice: generalized a couple of lemmas from sets;
  1226 generalized theorems INF_cong and SUP_cong.  New type classes for
  1227 complete boolean algebras and complete linear orders.  Lemmas
  1228 Inf_less_iff, less_Sup_iff, INF_less_iff, less_SUP_iff now reside in
  1229 class complete_linorder.
  1230 
  1231 Changed proposition of lemmas Inf_bool_def, Sup_bool_def, Inf_fun_def,
  1232 Sup_fun_def, Inf_apply, Sup_apply.
  1233 
  1234 Removed redundant lemmas (the right hand side gives hints how to
  1235 replace them for (metis ...), or (simp only: ...) proofs):
  1236 
  1237   Inf_singleton ~> Inf_insert [where A="{}", unfolded Inf_empty inf_top_right]
  1238   Sup_singleton ~> Sup_insert [where A="{}", unfolded Sup_empty sup_bot_right]
  1239   Inf_binary ~> Inf_insert, Inf_empty, and inf_top_right
  1240   Sup_binary ~> Sup_insert, Sup_empty, and sup_bot_right
  1241   Int_eq_Inter ~> Inf_insert, Inf_empty, and inf_top_right
  1242   Un_eq_Union ~> Sup_insert, Sup_empty, and sup_bot_right
  1243   Inter_def ~> INF_def, image_def
  1244   Union_def ~> SUP_def, image_def
  1245   INT_eq ~> INF_def, and image_def
  1246   UN_eq ~> SUP_def, and image_def
  1247   INF_subset ~> INF_superset_mono [OF _ order_refl]
  1248 
  1249 More consistent and comprehensive names:
  1250 
  1251   INTER_eq_Inter_image ~> INF_def
  1252   UNION_eq_Union_image ~> SUP_def
  1253   INFI_def ~> INF_def
  1254   SUPR_def ~> SUP_def
  1255   INF_leI ~> INF_lower
  1256   INF_leI2 ~> INF_lower2
  1257   le_INFI ~> INF_greatest
  1258   le_SUPI ~> SUP_upper
  1259   le_SUPI2 ~> SUP_upper2
  1260   SUP_leI ~> SUP_least
  1261   INFI_bool_eq ~> INF_bool_eq
  1262   SUPR_bool_eq ~> SUP_bool_eq
  1263   INFI_apply ~> INF_apply
  1264   SUPR_apply ~> SUP_apply
  1265   INTER_def ~> INTER_eq
  1266   UNION_def ~> UNION_eq
  1267 
  1268 INCOMPATIBILITY.
  1269 
  1270 * Renamed theory Complete_Lattice to Complete_Lattices.
  1271 INCOMPATIBILITY.
  1272 
  1273 * Theory Complete_Lattices: lemmas Inf_eq_top_iff, INF_eq_top_iff,
  1274 INF_image, Inf_insert, INF_top, Inf_top_conv, INF_top_conv, SUP_bot,
  1275 Sup_bot_conv, SUP_bot_conv, Sup_eq_top_iff, SUP_eq_top_iff, SUP_image,
  1276 Sup_insert are now declared as [simp].  INCOMPATIBILITY.
  1277 
  1278 * Theory Lattice: lemmas compl_inf_bot, compl_le_comp_iff,
  1279 compl_sup_top, inf_idem, inf_left_idem, inf_sup_absorb, sup_idem,
  1280 sup_inf_absob, sup_left_idem are now declared as [simp].  Minor
  1281 INCOMPATIBILITY.
  1282 
  1283 * Added syntactic classes "inf" and "sup" for the respective
  1284 constants.  INCOMPATIBILITY: Changes in the argument order of the
  1285 (mostly internal) locale predicates for some derived classes.
  1286 
  1287 * Theorem collections ball_simps and bex_simps do not contain theorems
  1288 referring to UNION any longer; these have been moved to collection
  1289 UN_ball_bex_simps.  INCOMPATIBILITY.
  1290 
  1291 * Theory Archimedean_Field: floor now is defined as parameter of a
  1292 separate type class floor_ceiling.
  1293 
  1294 * Theory Finite_Set: more coherent development of fold_set locales:
  1295 
  1296     locale fun_left_comm ~> locale comp_fun_commute
  1297     locale fun_left_comm_idem ~> locale comp_fun_idem
  1298 
  1299 Both use point-free characterization; interpretation proofs may need
  1300 adjustment.  INCOMPATIBILITY.
  1301 
  1302 * Theory Limits: Type "'a net" has been renamed to "'a filter", in
  1303 accordance with standard mathematical terminology. INCOMPATIBILITY.
  1304 
  1305 * Theory Complex_Main: The locale interpretations for the
  1306 bounded_linear and bounded_bilinear locales have been removed, in
  1307 order to reduce the number of duplicate lemmas. Users must use the
  1308 original names for distributivity theorems, potential INCOMPATIBILITY.
  1309 
  1310   divide.add ~> add_divide_distrib
  1311   divide.diff ~> diff_divide_distrib
  1312   divide.setsum ~> setsum_divide_distrib
  1313   mult.add_right ~> right_distrib
  1314   mult.diff_right ~> right_diff_distrib
  1315   mult_right.setsum ~> setsum_right_distrib
  1316   mult_left.diff ~> left_diff_distrib
  1317 
  1318 * Theory Complex_Main: Several redundant theorems have been removed or
  1319 replaced by more general versions. INCOMPATIBILITY.
  1320 
  1321   real_diff_def ~> minus_real_def
  1322   real_divide_def ~> divide_real_def
  1323   real_less_def ~> less_le
  1324   real_abs_def ~> abs_real_def
  1325   real_sgn_def ~> sgn_real_def
  1326   real_mult_commute ~> mult_commute
  1327   real_mult_assoc ~> mult_assoc
  1328   real_mult_1 ~> mult_1_left
  1329   real_add_mult_distrib ~> left_distrib
  1330   real_zero_not_eq_one ~> zero_neq_one
  1331   real_mult_inverse_left ~> left_inverse
  1332   INVERSE_ZERO ~> inverse_zero
  1333   real_le_refl ~> order_refl
  1334   real_le_antisym ~> order_antisym
  1335   real_le_trans ~> order_trans
  1336   real_le_linear ~> linear
  1337   real_le_eq_diff ~> le_iff_diff_le_0
  1338   real_add_left_mono ~> add_left_mono
  1339   real_mult_order ~> mult_pos_pos
  1340   real_mult_less_mono2 ~> mult_strict_left_mono
  1341   real_of_int_real_of_nat ~> real_of_int_of_nat_eq
  1342   real_0_le_divide_iff ~> zero_le_divide_iff
  1343   realpow_two_disj ~> power2_eq_iff
  1344   real_squared_diff_one_factored ~> square_diff_one_factored
  1345   realpow_two_diff ~> square_diff_square_factored
  1346   reals_complete2 ~> complete_real
  1347   real_sum_squared_expand ~> power2_sum
  1348   exp_ln_eq ~> ln_unique
  1349   expi_add ~> exp_add
  1350   expi_zero ~> exp_zero
  1351   lemma_DERIV_subst ~> DERIV_cong
  1352   LIMSEQ_Zfun_iff ~> tendsto_Zfun_iff
  1353   LIMSEQ_const ~> tendsto_const
  1354   LIMSEQ_norm ~> tendsto_norm
  1355   LIMSEQ_add ~> tendsto_add
  1356   LIMSEQ_minus ~> tendsto_minus
  1357   LIMSEQ_minus_cancel ~> tendsto_minus_cancel
  1358   LIMSEQ_diff ~> tendsto_diff
  1359   bounded_linear.LIMSEQ ~> bounded_linear.tendsto
  1360   bounded_bilinear.LIMSEQ ~> bounded_bilinear.tendsto
  1361   LIMSEQ_mult ~> tendsto_mult
  1362   LIMSEQ_inverse ~> tendsto_inverse
  1363   LIMSEQ_divide ~> tendsto_divide
  1364   LIMSEQ_pow ~> tendsto_power
  1365   LIMSEQ_setsum ~> tendsto_setsum
  1366   LIMSEQ_setprod ~> tendsto_setprod
  1367   LIMSEQ_norm_zero ~> tendsto_norm_zero_iff
  1368   LIMSEQ_rabs_zero ~> tendsto_rabs_zero_iff
  1369   LIMSEQ_imp_rabs ~> tendsto_rabs
  1370   LIMSEQ_add_minus ~> tendsto_add [OF _ tendsto_minus]
  1371   LIMSEQ_add_const ~> tendsto_add [OF _ tendsto_const]
  1372   LIMSEQ_diff_const ~> tendsto_diff [OF _ tendsto_const]
  1373   LIMSEQ_Complex ~> tendsto_Complex
  1374   LIM_ident ~> tendsto_ident_at
  1375   LIM_const ~> tendsto_const
  1376   LIM_add ~> tendsto_add
  1377   LIM_add_zero ~> tendsto_add_zero
  1378   LIM_minus ~> tendsto_minus
  1379   LIM_diff ~> tendsto_diff
  1380   LIM_norm ~> tendsto_norm
  1381   LIM_norm_zero ~> tendsto_norm_zero
  1382   LIM_norm_zero_cancel ~> tendsto_norm_zero_cancel
  1383   LIM_norm_zero_iff ~> tendsto_norm_zero_iff
  1384   LIM_rabs ~> tendsto_rabs
  1385   LIM_rabs_zero ~> tendsto_rabs_zero
  1386   LIM_rabs_zero_cancel ~> tendsto_rabs_zero_cancel
  1387   LIM_rabs_zero_iff ~> tendsto_rabs_zero_iff
  1388   LIM_compose ~> tendsto_compose
  1389   LIM_mult ~> tendsto_mult
  1390   LIM_scaleR ~> tendsto_scaleR
  1391   LIM_of_real ~> tendsto_of_real
  1392   LIM_power ~> tendsto_power
  1393   LIM_inverse ~> tendsto_inverse
  1394   LIM_sgn ~> tendsto_sgn
  1395   isCont_LIM_compose ~> isCont_tendsto_compose
  1396   bounded_linear.LIM ~> bounded_linear.tendsto
  1397   bounded_linear.LIM_zero ~> bounded_linear.tendsto_zero
  1398   bounded_bilinear.LIM ~> bounded_bilinear.tendsto
  1399   bounded_bilinear.LIM_prod_zero ~> bounded_bilinear.tendsto_zero
  1400   bounded_bilinear.LIM_left_zero ~> bounded_bilinear.tendsto_left_zero
  1401   bounded_bilinear.LIM_right_zero ~> bounded_bilinear.tendsto_right_zero
  1402   LIM_inverse_fun ~> tendsto_inverse [OF tendsto_ident_at]
  1403 
  1404 * Theory Complex_Main: The definition of infinite series was
  1405 generalized.  Now it is defined on the type class {topological_space,
  1406 comm_monoid_add}.  Hence it is useable also for extended real numbers.
  1407 
  1408 * Theory Complex_Main: The complex exponential function "expi" is now
  1409 a type-constrained abbreviation for "exp :: complex => complex"; thus
  1410 several polymorphic lemmas about "exp" are now applicable to "expi".
  1411 
  1412 * Code generation:
  1413 
  1414   - Theory Library/Code_Char_ord provides native ordering of
  1415     characters in the target language.
  1416 
  1417   - Commands code_module and code_library are legacy, use export_code
  1418     instead.
  1419 
  1420   - Method "evaluation" is legacy, use method "eval" instead.
  1421 
  1422   - Legacy evaluator "SML" is deactivated by default.  May be
  1423     reactivated by the following theory command:
  1424 
  1425       setup {* Value.add_evaluator ("SML", Codegen.eval_term) *}
  1426 
  1427 * Declare ext [intro] by default.  Rare INCOMPATIBILITY.
  1428 
  1429 * New proof method "induction" that gives induction hypotheses the
  1430 name "IH", thus distinguishing them from further hypotheses that come
  1431 from rule induction.  The latter are still called "hyps".  Method
  1432 "induction" is a thin wrapper around "induct" and follows the same
  1433 syntax.
  1434 
  1435 * Method "fastsimp" has been renamed to "fastforce", but "fastsimp" is
  1436 still available as a legacy feature for some time.
  1437 
  1438 * Nitpick:
  1439   - Added "need" and "total_consts" options.
  1440   - Reintroduced "show_skolems" option by popular demand.
  1441   - Renamed attribute: nitpick_def ~> nitpick_unfold.
  1442     INCOMPATIBILITY.
  1443 
  1444 * Sledgehammer:
  1445   - Use quasi-sound (and efficient) translations by default.
  1446   - Added support for the following provers: E-ToFoF, LEO-II,
  1447     Satallax, SNARK, Waldmeister, and Z3 with TPTP syntax.
  1448   - Automatically preplay and minimize proofs before showing them if
  1449     this can be done within reasonable time.
  1450   - sledgehammer available_provers ~> sledgehammer supported_provers.
  1451     INCOMPATIBILITY.
  1452   - Added "preplay_timeout", "slicing", "type_enc", "sound",
  1453     "max_mono_iters", and "max_new_mono_instances" options.
  1454   - Removed "explicit_apply" and "full_types" options as well as "Full
  1455     Types" Proof General menu item. INCOMPATIBILITY.
  1456 
  1457 * Metis:
  1458   - Removed "metisF" -- use "metis" instead. INCOMPATIBILITY.
  1459   - Obsoleted "metisFT" -- use "metis (full_types)" instead.
  1460     INCOMPATIBILITY.
  1461 
  1462 * Command 'try':
  1463   - Renamed 'try_methods' and added "simp:", "intro:", "dest:", and
  1464     "elim:" options. INCOMPATIBILITY.
  1465   - Introduced 'try' that not only runs 'try_methods' but also
  1466     'solve_direct', 'sledgehammer', 'quickcheck', and 'nitpick'.
  1467 
  1468 * Quickcheck:
  1469   - Added "eval" option to evaluate terms for the found counterexample
  1470     (currently only supported by the default (exhaustive) tester).
  1471   - Added post-processing of terms to obtain readable counterexamples
  1472     (currently only supported by the default (exhaustive) tester).
  1473   - New counterexample generator quickcheck[narrowing] enables
  1474     narrowing-based testing.  Requires the Glasgow Haskell compiler
  1475     with its installation location defined in the Isabelle settings
  1476     environment as ISABELLE_GHC.
  1477   - Removed quickcheck tester "SML" based on the SML code generator
  1478     (formly in HOL/Library).
  1479 
  1480 * Function package: discontinued option "tailrec".  INCOMPATIBILITY,
  1481 use 'partial_function' instead.
  1482 
  1483 * Theory Library/Extended_Reals replaces now the positive extended
  1484 reals found in probability theory. This file is extended by
  1485 Multivariate_Analysis/Extended_Real_Limits.
  1486 
  1487 * Theory Library/Old_Recdef: old 'recdef' package has been moved here,
  1488 from where it must be imported explicitly if it is really required.
  1489 INCOMPATIBILITY.
  1490 
  1491 * Theory Library/Wfrec: well-founded recursion combinator "wfrec" has
  1492 been moved here.  INCOMPATIBILITY.
  1493 
  1494 * Theory Library/Saturated provides type of numbers with saturated
  1495 arithmetic.
  1496 
  1497 * Theory Library/Product_Lattice defines a pointwise ordering for the
  1498 product type 'a * 'b, and provides instance proofs for various order
  1499 and lattice type classes.
  1500 
  1501 * Theory Library/Countable now provides the "countable_datatype" proof
  1502 method for proving "countable" class instances for datatypes.
  1503 
  1504 * Theory Library/Cset_Monad allows do notation for computable sets
  1505 (cset) via the generic monad ad-hoc overloading facility.
  1506 
  1507 * Library: Theories of common data structures are split into theories
  1508 for implementation, an invariant-ensuring type, and connection to an
  1509 abstract type. INCOMPATIBILITY.
  1510 
  1511   - RBT is split into RBT and RBT_Mapping.
  1512   - AssocList is split and renamed into AList and AList_Mapping.
  1513   - DList is split into DList_Impl, DList, and DList_Cset.
  1514   - Cset is split into Cset and List_Cset.
  1515 
  1516 * Theory Library/Nat_Infinity has been renamed to
  1517 Library/Extended_Nat, with name changes of the following types and
  1518 constants:
  1519 
  1520   type inat   ~> type enat
  1521   Fin         ~> enat
  1522   Infty       ~> infinity (overloaded)
  1523   iSuc        ~> eSuc
  1524   the_Fin     ~> the_enat
  1525 
  1526 Every theorem name containing "inat", "Fin", "Infty", or "iSuc" has
  1527 been renamed accordingly. INCOMPATIBILITY.
  1528 
  1529 * Session Multivariate_Analysis: The euclidean_space type class now
  1530 fixes a constant "Basis :: 'a set" consisting of the standard
  1531 orthonormal basis for the type. Users now have the option of
  1532 quantifying over this set instead of using the "basis" function, e.g.
  1533 "ALL x:Basis. P x" vs "ALL i<DIM('a). P (basis i)".
  1534 
  1535 * Session Multivariate_Analysis: Type "('a, 'b) cart" has been renamed
  1536 to "('a, 'b) vec" (the syntax "'a ^ 'b" remains unaffected). Constants
  1537 "Cart_nth" and "Cart_lambda" have been respectively renamed to
  1538 "vec_nth" and "vec_lambda"; theorems mentioning those names have
  1539 changed to match. Definition theorems for overloaded constants now use
  1540 the standard "foo_vec_def" naming scheme. A few other theorems have
  1541 been renamed as follows (INCOMPATIBILITY):
  1542 
  1543   Cart_eq          ~> vec_eq_iff
  1544   dist_nth_le_cart ~> dist_vec_nth_le
  1545   tendsto_vector   ~> vec_tendstoI
  1546   Cauchy_vector    ~> vec_CauchyI
  1547 
  1548 * Session Multivariate_Analysis: Several duplicate theorems have been
  1549 removed, and other theorems have been renamed or replaced with more
  1550 general versions. INCOMPATIBILITY.
  1551 
  1552   finite_choice ~> finite_set_choice
  1553   eventually_conjI ~> eventually_conj
  1554   eventually_and ~> eventually_conj_iff
  1555   eventually_false ~> eventually_False
  1556   setsum_norm ~> norm_setsum
  1557   Lim_sequentially ~> LIMSEQ_def
  1558   Lim_ident_at ~> LIM_ident
  1559   Lim_const ~> tendsto_const
  1560   Lim_cmul ~> tendsto_scaleR [OF tendsto_const]
  1561   Lim_neg ~> tendsto_minus
  1562   Lim_add ~> tendsto_add
  1563   Lim_sub ~> tendsto_diff
  1564   Lim_mul ~> tendsto_scaleR
  1565   Lim_vmul ~> tendsto_scaleR [OF _ tendsto_const]
  1566   Lim_null_norm ~> tendsto_norm_zero_iff [symmetric]
  1567   Lim_linear ~> bounded_linear.tendsto
  1568   Lim_component ~> tendsto_euclidean_component
  1569   Lim_component_cart ~> tendsto_vec_nth
  1570   Lim_inner ~> tendsto_inner [OF tendsto_const]
  1571   dot_lsum ~> inner_setsum_left
  1572   dot_rsum ~> inner_setsum_right
  1573   continuous_cmul ~> continuous_scaleR [OF continuous_const]
  1574   continuous_neg ~> continuous_minus
  1575   continuous_sub ~> continuous_diff
  1576   continuous_vmul ~> continuous_scaleR [OF _ continuous_const]
  1577   continuous_mul ~> continuous_scaleR
  1578   continuous_inv ~> continuous_inverse
  1579   continuous_at_within_inv ~> continuous_at_within_inverse
  1580   continuous_at_inv ~> continuous_at_inverse
  1581   continuous_at_norm ~> continuous_norm [OF continuous_at_id]
  1582   continuous_at_infnorm ~> continuous_infnorm [OF continuous_at_id]
  1583   continuous_at_component ~> continuous_component [OF continuous_at_id]
  1584   continuous_on_neg ~> continuous_on_minus
  1585   continuous_on_sub ~> continuous_on_diff
  1586   continuous_on_cmul ~> continuous_on_scaleR [OF continuous_on_const]
  1587   continuous_on_vmul ~> continuous_on_scaleR [OF _ continuous_on_const]
  1588   continuous_on_mul ~> continuous_on_scaleR
  1589   continuous_on_mul_real ~> continuous_on_mult
  1590   continuous_on_inner ~> continuous_on_inner [OF continuous_on_const]
  1591   continuous_on_norm ~> continuous_on_norm [OF continuous_on_id]
  1592   continuous_on_inverse ~> continuous_on_inv
  1593   uniformly_continuous_on_neg ~> uniformly_continuous_on_minus
  1594   uniformly_continuous_on_sub ~> uniformly_continuous_on_diff
  1595   subset_interior ~> interior_mono
  1596   subset_closure ~> closure_mono
  1597   closure_univ ~> closure_UNIV
  1598   real_arch_lt ~> reals_Archimedean2
  1599   real_arch ~> reals_Archimedean3
  1600   real_abs_norm ~> abs_norm_cancel
  1601   real_abs_sub_norm ~> norm_triangle_ineq3
  1602   norm_cauchy_schwarz_abs ~> Cauchy_Schwarz_ineq2
  1603 
  1604 * Session HOL-Probability:
  1605   - Caratheodory's extension lemma is now proved for ring_of_sets.
  1606   - Infinite products of probability measures are now available.
  1607   - Sigma closure is independent, if the generator is independent
  1608   - Use extended reals instead of positive extended
  1609     reals. INCOMPATIBILITY.
  1610 
  1611 * Session HOLCF: Discontinued legacy theorem names, INCOMPATIBILITY.
  1612 
  1613   expand_fun_below ~> fun_below_iff
  1614   below_fun_ext ~> fun_belowI
  1615   expand_cfun_eq ~> cfun_eq_iff
  1616   ext_cfun ~> cfun_eqI
  1617   expand_cfun_below ~> cfun_below_iff
  1618   below_cfun_ext ~> cfun_belowI
  1619   monofun_fun_fun ~> fun_belowD
  1620   monofun_fun_arg ~> monofunE
  1621   monofun_lub_fun ~> adm_monofun [THEN admD]
  1622   cont_lub_fun ~> adm_cont [THEN admD]
  1623   cont2cont_Rep_CFun ~> cont2cont_APP
  1624   cont_Rep_CFun_app ~> cont_APP_app
  1625   cont_Rep_CFun_app_app ~> cont_APP_app_app
  1626   cont_cfun_fun ~> cont_Rep_cfun1 [THEN contE]
  1627   cont_cfun_arg ~> cont_Rep_cfun2 [THEN contE]
  1628   contlub_cfun ~> lub_APP [symmetric]
  1629   contlub_LAM ~> lub_LAM [symmetric]
  1630   thelubI ~> lub_eqI
  1631   UU_I ~> bottomI
  1632   lift_distinct1 ~> lift.distinct(1)
  1633   lift_distinct2 ~> lift.distinct(2)
  1634   Def_not_UU ~> lift.distinct(2)
  1635   Def_inject ~> lift.inject
  1636   below_UU_iff ~> below_bottom_iff
  1637   eq_UU_iff ~> eq_bottom_iff
  1638 
  1639 
  1640 *** Document preparation ***
  1641 
  1642 * Antiquotation @{rail} layouts railroad syntax diagrams, see also
  1643 isar-ref manual, both for description and actual application of the
  1644 same.
  1645 
  1646 * Antiquotation @{value} evaluates the given term and presents its
  1647 result.
  1648 
  1649 * Antiquotations: term style "isub" provides ad-hoc conversion of
  1650 variables x1, y23 into subscripted form x\<^isub>1,
  1651 y\<^isub>2\<^isub>3.
  1652 
  1653 * Predefined LaTeX macros for Isabelle symbols \<bind> and \<then>
  1654 (e.g. see ~~/src/HOL/Library/Monad_Syntax.thy).
  1655 
  1656 * Localized \isabellestyle switch can be used within blocks or groups
  1657 like this:
  1658 
  1659   \isabellestyle{it}  %preferred default
  1660   {\isabellestylett @{text "typewriter stuff"}}
  1661 
  1662 * Discontinued special treatment of hard tabulators.  Implicit
  1663 tab-width is now defined as 1.  Potential INCOMPATIBILITY for visual
  1664 layouts.
  1665 
  1666 
  1667 *** ML ***
  1668 
  1669 * The inner syntax of sort/type/term/prop supports inlined YXML
  1670 representations within quoted string tokens.  By encoding logical
  1671 entities via Term_XML (in ML or Scala) concrete syntax can be
  1672 bypassed, which is particularly useful for producing bits of text
  1673 under external program control.
  1674 
  1675 * Antiquotations for ML and document preparation are managed as theory
  1676 data, which requires explicit setup.
  1677 
  1678 * Isabelle_Process.is_active allows tools to check if the official
  1679 process wrapper is running (Isabelle/Scala/jEdit) or the old TTY loop
  1680 (better known as Proof General).
  1681 
  1682 * Structure Proof_Context follows standard naming scheme.  Old
  1683 ProofContext is still available for some time as legacy alias.
  1684 
  1685 * Structure Timing provides various operations for timing; supersedes
  1686 former start_timing/end_timing etc.
  1687 
  1688 * Path.print is the official way to show file-system paths to users
  1689 (including quotes etc.).
  1690 
  1691 * Inner syntax: identifiers in parse trees of generic categories
  1692 "logic", "aprop", "idt" etc. carry position information (disguised as
  1693 type constraints).  Occasional INCOMPATIBILITY with non-compliant
  1694 translations that choke on unexpected type constraints.  Positions can
  1695 be stripped in ML translations via Syntax.strip_positions /
  1696 Syntax.strip_positions_ast, or via the syntax constant
  1697 "_strip_positions" within parse trees.  As last resort, positions can
  1698 be disabled via the configuration option Syntax.positions, which is
  1699 called "syntax_positions" in Isar attribute syntax.
  1700 
  1701 * Discontinued special status of various ML structures that contribute
  1702 to structure Syntax (Ast, Lexicon, Mixfix, Parser, Printer etc.): less
  1703 pervasive content, no inclusion in structure Syntax.  INCOMPATIBILITY,
  1704 refer directly to Ast.Constant, Lexicon.is_identifier,
  1705 Syntax_Trans.mk_binder_tr etc.
  1706 
  1707 * Typed print translation: discontinued show_sorts argument, which is
  1708 already available via context of "advanced" translation.
  1709 
  1710 * Refined PARALLEL_GOALS tactical: degrades gracefully for schematic
  1711 goal states; body tactic needs to address all subgoals uniformly.
  1712 
  1713 * Slightly more special eq_list/eq_set, with shortcut involving
  1714 pointer equality (assumes that eq relation is reflexive).
  1715 
  1716 * Classical tactics use proper Proof.context instead of historic types
  1717 claset/clasimpset.  Old-style declarations like addIs, addEs, addDs
  1718 operate directly on Proof.context.  Raw type claset retains its use as
  1719 snapshot of the classical context, which can be recovered via
  1720 (put_claset HOL_cs) etc.  Type clasimpset has been discontinued.
  1721 INCOMPATIBILITY, classical tactics and derived proof methods require
  1722 proper Proof.context.
  1723 
  1724 
  1725 *** System ***
  1726 
  1727 * Discontinued support for Poly/ML 5.2, which was the last version
  1728 without proper multithreading and TimeLimit implementation.
  1729 
  1730 * Discontinued old lib/scripts/polyml-platform, which has been
  1731 obsolete since Isabelle2009-2.
  1732 
  1733 * Various optional external tools are referenced more robustly and
  1734 uniformly by explicit Isabelle settings as follows:
  1735 
  1736   ISABELLE_CSDP   (formerly CSDP_EXE)
  1737   ISABELLE_GHC    (formerly EXEC_GHC or GHC_PATH)
  1738   ISABELLE_OCAML  (formerly EXEC_OCAML)
  1739   ISABELLE_SWIPL  (formerly EXEC_SWIPL)
  1740   ISABELLE_YAP    (formerly EXEC_YAP)
  1741 
  1742 Note that automated detection from the file-system or search path has
  1743 been discontinued.  INCOMPATIBILITY.
  1744 
  1745 * Scala layer provides JVM method invocation service for static
  1746 methods of type (String)String, see Invoke_Scala.method in ML.  For
  1747 example:
  1748 
  1749   Invoke_Scala.method "java.lang.System.getProperty" "java.home"
  1750 
  1751 Together with YXML.string_of_body/parse_body and XML.Encode/Decode
  1752 this allows to pass structured values between ML and Scala.
  1753 
  1754 * The IsabelleText fonts includes some further glyphs to support the
  1755 Prover IDE.  Potential INCOMPATIBILITY: users who happen to have
  1756 installed a local copy (which is normally *not* required) need to
  1757 delete or update it from ~~/lib/fonts/.
  1758 
  1759 
  1760 
  1761 New in Isabelle2011 (January 2011)
  1762 ----------------------------------
  1763 
  1764 *** General ***
  1765 
  1766 * Experimental Prover IDE based on Isabelle/Scala and jEdit (see
  1767 src/Tools/jEdit).  This also serves as IDE for Isabelle/ML, with
  1768 useful tooltips and hyperlinks produced from its static analysis.  The
  1769 bundled component provides an executable Isabelle tool that can be run
  1770 like this:
  1771 
  1772   Isabelle2011/bin/isabelle jedit
  1773 
  1774 * Significantly improved Isabelle/Isar implementation manual.
  1775 
  1776 * System settings: ISABELLE_HOME_USER now includes ISABELLE_IDENTIFIER
  1777 (and thus refers to something like $HOME/.isabelle/Isabelle2011),
  1778 while the default heap location within that directory lacks that extra
  1779 suffix.  This isolates multiple Isabelle installations from each
  1780 other, avoiding problems with old settings in new versions.
  1781 INCOMPATIBILITY, need to copy/upgrade old user settings manually.
  1782 
  1783 * Source files are always encoded as UTF-8, instead of old-fashioned
  1784 ISO-Latin-1.  INCOMPATIBILITY.  Isabelle LaTeX documents might require
  1785 the following package declarations:
  1786 
  1787   \usepackage[utf8]{inputenc}
  1788   \usepackage{textcomp}
  1789 
  1790 * Explicit treatment of UTF-8 sequences as Isabelle symbols, such that
  1791 a Unicode character is treated as a single symbol, not a sequence of
  1792 non-ASCII bytes as before.  Since Isabelle/ML string literals may
  1793 contain symbols without further backslash escapes, Unicode can now be
  1794 used here as well.  Recall that Symbol.explode in ML provides a
  1795 consistent view on symbols, while raw explode (or String.explode)
  1796 merely give a byte-oriented representation.
  1797 
  1798 * Theory loader: source files are primarily located via the master
  1799 directory of each theory node (where the .thy file itself resides).
  1800 The global load path is still partially available as legacy feature.
  1801 Minor INCOMPATIBILITY due to subtle change in file lookup: use
  1802 explicit paths, relatively to the theory.
  1803 
  1804 * Special treatment of ML file names has been discontinued.
  1805 Historically, optional extensions .ML or .sml were added on demand --
  1806 at the cost of clarity of file dependencies.  Recall that Isabelle/ML
  1807 files exclusively use the .ML extension.  Minor INCOMPATIBILTY.
  1808 
  1809 * Various options that affect pretty printing etc. are now properly
  1810 handled within the context via configuration options, instead of
  1811 unsynchronized references or print modes.  There are both ML Config.T
  1812 entities and Isar declaration attributes to access these.
  1813 
  1814   ML (Config.T)                 Isar (attribute)
  1815 
  1816   eta_contract                  eta_contract
  1817   show_brackets                 show_brackets
  1818   show_sorts                    show_sorts
  1819   show_types                    show_types
  1820   show_question_marks           show_question_marks
  1821   show_consts                   show_consts
  1822   show_abbrevs                  show_abbrevs
  1823 
  1824   Syntax.ast_trace              syntax_ast_trace
  1825   Syntax.ast_stat               syntax_ast_stat
  1826   Syntax.ambiguity_level        syntax_ambiguity_level
  1827 
  1828   Goal_Display.goals_limit      goals_limit
  1829   Goal_Display.show_main_goal   show_main_goal
  1830 
  1831   Method.rule_trace             rule_trace
  1832 
  1833   Thy_Output.display            thy_output_display
  1834   Thy_Output.quotes             thy_output_quotes
  1835   Thy_Output.indent             thy_output_indent
  1836   Thy_Output.source             thy_output_source
  1837   Thy_Output.break              thy_output_break
  1838 
  1839 Note that corresponding "..._default" references in ML may only be
  1840 changed globally at the ROOT session setup, but *not* within a theory.
  1841 The option "show_abbrevs" supersedes the former print mode
  1842 "no_abbrevs" with inverted meaning.
  1843 
  1844 * More systematic naming of some configuration options.
  1845 INCOMPATIBILITY.
  1846 
  1847   trace_simp  ~>  simp_trace
  1848   debug_simp  ~>  simp_debug
  1849 
  1850 * Support for real valued configuration options, using simplistic
  1851 floating-point notation that coincides with the inner syntax for
  1852 float_token.
  1853 
  1854 * Support for real valued preferences (with approximative PGIP type):
  1855 front-ends need to accept "pgint" values in float notation.
  1856 INCOMPATIBILITY.
  1857 
  1858 * The IsabelleText font now includes Cyrillic, Hebrew, Arabic from
  1859 DejaVu Sans.
  1860 
  1861 * Discontinued support for Poly/ML 5.0 and 5.1 versions.
  1862 
  1863 
  1864 *** Pure ***
  1865 
  1866 * Command 'type_synonym' (with single argument) replaces somewhat
  1867 outdated 'types', which is still available as legacy feature for some
  1868 time.
  1869 
  1870 * Command 'nonterminal' (with 'and' separated list of arguments)
  1871 replaces somewhat outdated 'nonterminals'.  INCOMPATIBILITY.
  1872 
  1873 * Command 'notepad' replaces former 'example_proof' for
  1874 experimentation in Isar without any result.  INCOMPATIBILITY.
  1875 
  1876 * Locale interpretation commands 'interpret' and 'sublocale' accept
  1877 lists of equations to map definitions in a locale to appropriate
  1878 entities in the context of the interpretation.  The 'interpretation'
  1879 command already provided this functionality.
  1880 
  1881 * Diagnostic command 'print_dependencies' prints the locale instances
  1882 that would be activated if the specified expression was interpreted in
  1883 the current context.  Variant "print_dependencies!" assumes a context
  1884 without interpretations.
  1885 
  1886 * Diagnostic command 'print_interps' prints interpretations in proofs
  1887 in addition to interpretations in theories.
  1888 
  1889 * Discontinued obsolete 'global' and 'local' commands to manipulate
  1890 the theory name space.  Rare INCOMPATIBILITY.  The ML functions
  1891 Sign.root_path and Sign.local_path may be applied directly where this
  1892 feature is still required for historical reasons.
  1893 
  1894 * Discontinued obsolete 'constdefs' command.  INCOMPATIBILITY, use
  1895 'definition' instead.
  1896 
  1897 * The "prems" fact, which refers to the accidental collection of
  1898 foundational premises in the context, is now explicitly marked as
  1899 legacy feature and will be discontinued soon.  Consider using "assms"
  1900 of the head statement or reference facts by explicit names.
  1901 
  1902 * Document antiquotations @{class} and @{type} print classes and type
  1903 constructors.
  1904 
  1905 * Document antiquotation @{file} checks file/directory entries within
  1906 the local file system.
  1907 
  1908 
  1909 *** HOL ***
  1910 
  1911 * Coercive subtyping: functions can be declared as coercions and type
  1912 inference will add them as necessary upon input of a term.  Theory
  1913 Complex_Main declares real :: nat => real and real :: int => real as
  1914 coercions. A coercion function f is declared like this:
  1915 
  1916   declare [[coercion f]]
  1917 
  1918 To lift coercions through type constructors (e.g. from nat => real to
  1919 nat list => real list), map functions can be declared, e.g.
  1920 
  1921   declare [[coercion_map map]]
  1922 
  1923 Currently coercion inference is activated only in theories including
  1924 real numbers, i.e. descendants of Complex_Main.  This is controlled by
  1925 the configuration option "coercion_enabled", e.g. it can be enabled in
  1926 other theories like this:
  1927 
  1928   declare [[coercion_enabled]]
  1929 
  1930 * Command 'partial_function' provides basic support for recursive
  1931 function definitions over complete partial orders.  Concrete instances
  1932 are provided for i) the option type, ii) tail recursion on arbitrary
  1933 types, and iii) the heap monad of Imperative_HOL.  See
  1934 src/HOL/ex/Fundefs.thy and src/HOL/Imperative_HOL/ex/Linked_Lists.thy
  1935 for examples.
  1936 
  1937 * Function package: f.psimps rules are no longer implicitly declared
  1938 as [simp].  INCOMPATIBILITY.
  1939 
  1940 * Datatype package: theorems generated for executable equality (class
  1941 "eq") carry proper names and are treated as default code equations.
  1942 
  1943 * Inductive package: now offers command 'inductive_simps' to
  1944 automatically derive instantiated and simplified equations for
  1945 inductive predicates, similar to 'inductive_cases'.
  1946 
  1947 * Command 'enriched_type' allows to register properties of the
  1948 functorial structure of types.
  1949 
  1950 * Improved infrastructure for term evaluation using code generator
  1951 techniques, in particular static evaluation conversions.
  1952 
  1953 * Code generator: Scala (2.8 or higher) has been added to the target
  1954 languages.
  1955 
  1956 * Code generator: globbing constant expressions "*" and "Theory.*"
  1957 have been replaced by the more idiomatic "_" and "Theory._".
  1958 INCOMPATIBILITY.
  1959 
  1960 * Code generator: export_code without explicit file declaration prints
  1961 to standard output.  INCOMPATIBILITY.
  1962 
  1963 * Code generator: do not print function definitions for case
  1964 combinators any longer.
  1965 
  1966 * Code generator: simplification with rules determined with
  1967 src/Tools/Code/code_simp.ML and method "code_simp".
  1968 
  1969 * Code generator for records: more idiomatic representation of record
  1970 types.  Warning: records are not covered by ancient SML code
  1971 generation any longer.  INCOMPATIBILITY.  In cases of need, a suitable
  1972 rep_datatype declaration helps to succeed then:
  1973 
  1974   record 'a foo = ...
  1975   ...
  1976   rep_datatype foo_ext ...
  1977 
  1978 * Records: logical foundation type for records does not carry a
  1979 '_type' suffix any longer (obsolete due to authentic syntax).
  1980 INCOMPATIBILITY.
  1981 
  1982 * Quickcheck now by default uses exhaustive testing instead of random
  1983 testing.  Random testing can be invoked by "quickcheck [random]",
  1984 exhaustive testing by "quickcheck [exhaustive]".
  1985 
  1986 * Quickcheck instantiates polymorphic types with small finite
  1987 datatypes by default. This enables a simple execution mechanism to
  1988 handle quantifiers and function equality over the finite datatypes.
  1989 
  1990 * Quickcheck random generator has been renamed from "code" to
  1991 "random".  INCOMPATIBILITY.
  1992 
  1993 * Quickcheck now has a configurable time limit which is set to 30
  1994 seconds by default. This can be changed by adding [timeout = n] to the
  1995 quickcheck command. The time limit for Auto Quickcheck is still set
  1996 independently.
  1997 
  1998 * Quickcheck in locales considers interpretations of that locale for
  1999 counter example search.
  2000 
  2001 * Sledgehammer:
  2002   - Added "smt" and "remote_smt" provers based on the "smt" proof
  2003     method. See the Sledgehammer manual for details ("isabelle doc
  2004     sledgehammer").
  2005   - Renamed commands:
  2006     sledgehammer atp_info ~> sledgehammer running_provers
  2007     sledgehammer atp_kill ~> sledgehammer kill_provers
  2008     sledgehammer available_atps ~> sledgehammer available_provers
  2009     INCOMPATIBILITY.
  2010   - Renamed options:
  2011     sledgehammer [atps = ...] ~> sledgehammer [provers = ...]
  2012     sledgehammer [atp = ...] ~> sledgehammer [prover = ...]
  2013     sledgehammer [timeout = 77 s] ~> sledgehammer [timeout = 77]
  2014     (and "ms" and "min" are no longer supported)
  2015     INCOMPATIBILITY.
  2016 
  2017 * Nitpick:
  2018   - Renamed options:
  2019     nitpick [timeout = 77 s] ~> nitpick [timeout = 77]
  2020     nitpick [tac_timeout = 777 ms] ~> nitpick [tac_timeout = 0.777]
  2021     INCOMPATIBILITY.
  2022   - Added support for partial quotient types.
  2023   - Added local versions of the "Nitpick.register_xxx" functions.
  2024   - Added "whack" option.
  2025   - Allow registration of quotient types as codatatypes.
  2026   - Improved "merge_type_vars" option to merge more types.
  2027   - Removed unsound "fast_descrs" option.
  2028   - Added custom symmetry breaking for datatypes, making it possible to reach
  2029     higher cardinalities.
  2030   - Prevent the expansion of too large definitions.
  2031 
  2032 * Proof methods "metis" and "meson" now have configuration options
  2033 "meson_trace", "metis_trace", and "metis_verbose" that can be enabled
  2034 to diagnose these tools. E.g.
  2035 
  2036     using [[metis_trace = true]]
  2037 
  2038 * Auto Solve: Renamed "Auto Solve Direct".  The tool is now available
  2039 manually as command 'solve_direct'.
  2040 
  2041 * The default SMT solver Z3 must be enabled explicitly (due to
  2042 licensing issues) by setting the environment variable
  2043 Z3_NON_COMMERCIAL in etc/settings of the component, for example.  For
  2044 commercial applications, the SMT solver CVC3 is provided as fall-back;
  2045 changing the SMT solver is done via the configuration option
  2046 "smt_solver".
  2047 
  2048 * Remote SMT solvers need to be referred to by the "remote_" prefix,
  2049 i.e. "remote_cvc3" and "remote_z3".
  2050 
  2051 * Added basic SMT support for datatypes, records, and typedefs using
  2052 the oracle mode (no proofs).  Direct support of pairs has been dropped
  2053 in exchange (pass theorems fst_conv snd_conv pair_collapse to the SMT
  2054 support for a similar behavior).  Minor INCOMPATIBILITY.
  2055 
  2056 * Changed SMT configuration options:
  2057   - Renamed:
  2058     z3_proofs ~> smt_oracle (with inverted meaning)
  2059     z3_trace_assms ~> smt_trace_used_facts
  2060     INCOMPATIBILITY.
  2061   - Added:
  2062     smt_verbose
  2063     smt_random_seed
  2064     smt_datatypes
  2065     smt_infer_triggers
  2066     smt_monomorph_limit
  2067     cvc3_options
  2068     remote_cvc3_options
  2069     remote_z3_options
  2070     yices_options
  2071 
  2072 * Boogie output files (.b2i files) need to be declared in the theory
  2073 header.
  2074 
  2075 * Simplification procedure "list_to_set_comprehension" rewrites list
  2076 comprehensions applied to List.set to set comprehensions.  Occasional
  2077 INCOMPATIBILITY, may be deactivated like this:
  2078 
  2079   declare [[simproc del: list_to_set_comprehension]]
  2080 
  2081 * Removed old version of primrec package.  INCOMPATIBILITY.
  2082 
  2083 * Removed simplifier congruence rule of "prod_case", as has for long
  2084 been the case with "split".  INCOMPATIBILITY.
  2085 
  2086 * String.literal is a type, but not a datatype.  INCOMPATIBILITY.
  2087 
  2088 * Removed [split_format ... and ... and ...] version of
  2089 [split_format].  Potential INCOMPATIBILITY.
  2090 
  2091 * Predicate "sorted" now defined inductively, with nice induction
  2092 rules.  INCOMPATIBILITY: former sorted.simps now named sorted_simps.
  2093 
  2094 * Constant "contents" renamed to "the_elem", to free the generic name
  2095 contents for other uses.  INCOMPATIBILITY.
  2096 
  2097 * Renamed class eq and constant eq (for code generation) to class
  2098 equal and constant equal, plus renaming of related facts and various
  2099 tuning.  INCOMPATIBILITY.
  2100 
  2101 * Dropped type classes mult_mono and mult_mono1.  INCOMPATIBILITY.
  2102 
  2103 * Removed output syntax "'a ~=> 'b" for "'a => 'b option".
  2104 INCOMPATIBILITY.
  2105 
  2106 * Renamed theory Fset to Cset, type Fset.fset to Cset.set, in order to
  2107 avoid confusion with finite sets.  INCOMPATIBILITY.
  2108 
  2109 * Abandoned locales equiv, congruent and congruent2 for equivalence
  2110 relations.  INCOMPATIBILITY: use equivI rather than equiv_intro (same
  2111 for congruent(2)).
  2112 
  2113 * Some previously unqualified names have been qualified:
  2114 
  2115   types
  2116     bool ~> HOL.bool
  2117     nat ~> Nat.nat
  2118 
  2119   constants
  2120     Trueprop ~> HOL.Trueprop
  2121     True ~> HOL.True
  2122     False ~> HOL.False
  2123     op & ~> HOL.conj
  2124     op | ~> HOL.disj
  2125     op --> ~> HOL.implies
  2126     op = ~> HOL.eq
  2127     Not ~> HOL.Not
  2128     The ~> HOL.The
  2129     All ~> HOL.All
  2130     Ex ~> HOL.Ex
  2131     Ex1 ~> HOL.Ex1
  2132     Let ~> HOL.Let
  2133     If ~> HOL.If
  2134     Ball ~> Set.Ball
  2135     Bex ~> Set.Bex
  2136     Suc ~> Nat.Suc
  2137     Pair ~> Product_Type.Pair
  2138     fst ~> Product_Type.fst
  2139     snd ~> Product_Type.snd
  2140     curry ~> Product_Type.curry
  2141     op : ~> Set.member
  2142     Collect ~> Set.Collect
  2143 
  2144 INCOMPATIBILITY.
  2145 
  2146 * More canonical naming convention for some fundamental definitions:
  2147 
  2148     bot_bool_eq ~> bot_bool_def
  2149     top_bool_eq ~> top_bool_def
  2150     inf_bool_eq ~> inf_bool_def
  2151     sup_bool_eq ~> sup_bool_def
  2152     bot_fun_eq  ~> bot_fun_def
  2153     top_fun_eq  ~> top_fun_def
  2154     inf_fun_eq  ~> inf_fun_def
  2155     sup_fun_eq  ~> sup_fun_def
  2156 
  2157 INCOMPATIBILITY.
  2158 
  2159 * More stylized fact names:
  2160 
  2161   expand_fun_eq ~> fun_eq_iff
  2162   expand_set_eq ~> set_eq_iff
  2163   set_ext       ~> set_eqI
  2164   nat_number    ~> eval_nat_numeral
  2165 
  2166 INCOMPATIBILITY.
  2167 
  2168 * Refactoring of code-generation specific operations in theory List:
  2169 
  2170   constants
  2171     null ~> List.null
  2172 
  2173   facts
  2174     mem_iff ~> member_def
  2175     null_empty ~> null_def
  2176 
  2177 INCOMPATIBILITY.  Note that these were not supposed to be used
  2178 regularly unless for striking reasons; their main purpose was code
  2179 generation.
  2180 
  2181 Various operations from the Haskell prelude are used for generating
  2182 Haskell code.
  2183 
  2184 * Term "bij f" is now an abbreviation of "bij_betw f UNIV UNIV".  Term
  2185 "surj f" is now an abbreviation of "range f = UNIV".  The theorems
  2186 bij_def and surj_def are unchanged.  INCOMPATIBILITY.
  2187 
  2188 * Abolished some non-alphabetic type names: "prod" and "sum" replace
  2189 "*" and "+" respectively.  INCOMPATIBILITY.
  2190 
  2191 * Name "Plus" of disjoint sum operator "<+>" is now hidden.  Write
  2192 "Sum_Type.Plus" instead.
  2193 
  2194 * Constant "split" has been merged with constant "prod_case"; names of
  2195 ML functions, facts etc. involving split have been retained so far,
  2196 though.  INCOMPATIBILITY.
  2197 
  2198 * Dropped old infix syntax "_ mem _" for List.member; use "_ : set _"
  2199 instead.  INCOMPATIBILITY.
  2200 
  2201 * Removed lemma "Option.is_none_none" which duplicates "is_none_def".
  2202 INCOMPATIBILITY.
  2203 
  2204 * Former theory Library/Enum is now part of the HOL-Main image.
  2205 INCOMPATIBILITY: all constants of the Enum theory now have to be
  2206 referred to by its qualified name.
  2207 
  2208   enum    ~>  Enum.enum
  2209   nlists  ~>  Enum.nlists
  2210   product ~>  Enum.product
  2211 
  2212 * Theory Library/Monad_Syntax provides do-syntax for monad types.
  2213 Syntax in Library/State_Monad has been changed to avoid ambiguities.
  2214 INCOMPATIBILITY.
  2215 
  2216 * Theory Library/SetsAndFunctions has been split into
  2217 Library/Function_Algebras and Library/Set_Algebras; canonical names
  2218 for instance definitions for functions; various improvements.
  2219 INCOMPATIBILITY.
  2220 
  2221 * Theory Library/Multiset provides stable quicksort implementation of
  2222 sort_key.
  2223 
  2224 * Theory Library/Multiset: renamed empty_idemp ~> empty_neutral.
  2225 INCOMPATIBILITY.
  2226 
  2227 * Session Multivariate_Analysis: introduced a type class for euclidean
  2228 space.  Most theorems are now stated in terms of euclidean spaces
  2229 instead of finite cartesian products.
  2230 
  2231   types
  2232     real ^ 'n ~>  'a::real_vector
  2233               ~>  'a::euclidean_space
  2234               ~>  'a::ordered_euclidean_space
  2235         (depends on your needs)
  2236 
  2237   constants
  2238      _ $ _        ~> _ $$ _
  2239      \<chi> x. _  ~> \<chi>\<chi> x. _
  2240      CARD('n)     ~> DIM('a)
  2241 
  2242 Also note that the indices are now natural numbers and not from some
  2243 finite type. Finite cartesian products of euclidean spaces, products
  2244 of euclidean spaces the real and complex numbers are instantiated to
  2245 be euclidean_spaces.  INCOMPATIBILITY.
  2246 
  2247 * Session Probability: introduced pextreal as positive extended real
  2248 numbers.  Use pextreal as value for measures.  Introduce the
  2249 Radon-Nikodym derivative, product spaces and Fubini's theorem for
  2250 arbitrary sigma finite measures.  Introduces Lebesgue measure based on
  2251 the integral in Multivariate Analysis.  INCOMPATIBILITY.
  2252 
  2253 * Session Imperative_HOL: revamped, corrected dozens of inadequacies.
  2254 INCOMPATIBILITY.
  2255 
  2256 * Session SPARK (with image HOL-SPARK) provides commands to load and
  2257 prove verification conditions generated by the SPARK Ada program
  2258 verifier.  See also src/HOL/SPARK and src/HOL/SPARK/Examples.
  2259 
  2260 
  2261 *** HOL-Algebra ***
  2262 
  2263 * Theorems for additive ring operations (locale abelian_monoid and
  2264 descendants) are generated by interpretation from their multiplicative
  2265 counterparts.  Names (in particular theorem names) have the mandatory
  2266 qualifier 'add'.  Previous theorem names are redeclared for
  2267 compatibility.
  2268 
  2269 * Structure "int_ring" is now an abbreviation (previously a
  2270 definition).  This fits more natural with advanced interpretations.
  2271 
  2272 
  2273 *** HOLCF ***
  2274 
  2275 * The domain package now runs in definitional mode by default: The
  2276 former command 'new_domain' is now called 'domain'.  To use the domain
  2277 package in its original axiomatic mode, use 'domain (unsafe)'.
  2278 INCOMPATIBILITY.
  2279 
  2280 * The new class "domain" is now the default sort.  Class "predomain"
  2281 is an unpointed version of "domain". Theories can be updated by
  2282 replacing sort annotations as shown below.  INCOMPATIBILITY.
  2283 
  2284   'a::type ~> 'a::countable
  2285   'a::cpo  ~> 'a::predomain
  2286   'a::pcpo ~> 'a::domain
  2287 
  2288 * The old type class "rep" has been superseded by class "domain".
  2289 Accordingly, users of the definitional package must remove any
  2290 "default_sort rep" declarations.  INCOMPATIBILITY.
  2291 
  2292 * The domain package (definitional mode) now supports unpointed
  2293 predomain argument types, as long as they are marked 'lazy'. (Strict
  2294 arguments must be in class "domain".) For example, the following
  2295 domain definition now works:
  2296 
  2297   domain natlist = nil | cons (lazy "nat discr") (lazy "natlist")
  2298 
  2299 * Theory HOLCF/Library/HOL_Cpo provides cpo and predomain class
  2300 instances for types from main HOL: bool, nat, int, char, 'a + 'b,
  2301 'a option, and 'a list.  Additionally, it configures fixrec and the
  2302 domain package to work with these types.  For example:
  2303 
  2304   fixrec isInl :: "('a + 'b) u -> tr"
  2305     where "isInl$(up$(Inl x)) = TT" | "isInl$(up$(Inr y)) = FF"
  2306 
  2307   domain V = VFun (lazy "V -> V") | VCon (lazy "nat") (lazy "V list")
  2308 
  2309 * The "(permissive)" option of fixrec has been replaced with a
  2310 per-equation "(unchecked)" option. See
  2311 src/HOL/HOLCF/Tutorial/Fixrec_ex.thy for examples. INCOMPATIBILITY.
  2312 
  2313 * The "bifinite" class no longer fixes a constant "approx"; the class
  2314 now just asserts that such a function exists.  INCOMPATIBILITY.
  2315 
  2316 * Former type "alg_defl" has been renamed to "defl".  HOLCF no longer
  2317 defines an embedding of type 'a defl into udom by default; instances
  2318 of "bifinite" and "domain" classes are available in
  2319 src/HOL/HOLCF/Library/Defl_Bifinite.thy.
  2320 
  2321 * The syntax "REP('a)" has been replaced with "DEFL('a)".
  2322 
  2323 * The predicate "directed" has been removed.  INCOMPATIBILITY.
  2324 
  2325 * The type class "finite_po" has been removed.  INCOMPATIBILITY.
  2326 
  2327 * The function "cprod_map" has been renamed to "prod_map".
  2328 INCOMPATIBILITY.
  2329 
  2330 * The monadic bind operator on each powerdomain has new binder syntax
  2331 similar to sets, e.g. "\<Union>\<sharp>x\<in>xs. t" represents
  2332 "upper_bind\<cdot>xs\<cdot>(\<Lambda> x. t)".
  2333 
  2334 * The infix syntax for binary union on each powerdomain has changed
  2335 from e.g. "+\<sharp>" to "\<union>\<sharp>", for consistency with set
  2336 syntax.  INCOMPATIBILITY.
  2337 
  2338 * The constant "UU" has been renamed to "bottom".  The syntax "UU" is
  2339 still supported as an input translation.
  2340 
  2341 * Renamed some theorems (the original names are also still available).
  2342 
  2343   expand_fun_below   ~> fun_below_iff
  2344   below_fun_ext      ~> fun_belowI
  2345   expand_cfun_eq     ~> cfun_eq_iff
  2346   ext_cfun           ~> cfun_eqI
  2347   expand_cfun_below  ~> cfun_below_iff
  2348   below_cfun_ext     ~> cfun_belowI
  2349   cont2cont_Rep_CFun ~> cont2cont_APP
  2350 
  2351 * The Abs and Rep functions for various types have changed names.
  2352 Related theorem names have also changed to match. INCOMPATIBILITY.
  2353 
  2354   Rep_CFun  ~> Rep_cfun
  2355   Abs_CFun  ~> Abs_cfun
  2356   Rep_Sprod ~> Rep_sprod
  2357   Abs_Sprod ~> Abs_sprod
  2358   Rep_Ssum  ~> Rep_ssum
  2359   Abs_Ssum  ~> Abs_ssum
  2360 
  2361 * Lemmas with names of the form *_defined_iff or *_strict_iff have
  2362 been renamed to *_bottom_iff.  INCOMPATIBILITY.
  2363 
  2364 * Various changes to bisimulation/coinduction with domain package:
  2365 
  2366   - Definitions of "bisim" constants no longer mention definedness.
  2367   - With mutual recursion, "bisim" predicate is now curried.
  2368   - With mutual recursion, each type gets a separate coind theorem.
  2369   - Variable names in bisim_def and coinduct rules have changed.
  2370 
  2371 INCOMPATIBILITY.
  2372 
  2373 * Case combinators generated by the domain package for type "foo" are
  2374 now named "foo_case" instead of "foo_when".  INCOMPATIBILITY.
  2375 
  2376 * Several theorems have been renamed to more accurately reflect the
  2377 names of constants and types involved.  INCOMPATIBILITY.
  2378 
  2379   thelub_const    ~> lub_const
  2380   lub_const       ~> is_lub_const
  2381   thelubI         ~> lub_eqI
  2382   is_lub_lub      ~> is_lubD2
  2383   lubI            ~> is_lub_lub
  2384   unique_lub      ~> is_lub_unique
  2385   is_ub_lub       ~> is_lub_rangeD1
  2386   lub_bin_chain   ~> is_lub_bin_chain
  2387   lub_fun         ~> is_lub_fun
  2388   thelub_fun      ~> lub_fun
  2389   thelub_cfun     ~> lub_cfun
  2390   thelub_Pair     ~> lub_Pair
  2391   lub_cprod       ~> is_lub_prod
  2392   thelub_cprod    ~> lub_prod
  2393   minimal_cprod   ~> minimal_prod
  2394   inst_cprod_pcpo ~> inst_prod_pcpo
  2395   UU_I            ~> bottomI
  2396   compact_UU      ~> compact_bottom
  2397   deflation_UU    ~> deflation_bottom
  2398   finite_deflation_UU ~> finite_deflation_bottom
  2399 
  2400 * Many legacy theorem names have been discontinued.  INCOMPATIBILITY.
  2401 
  2402   sq_ord_less_eq_trans ~> below_eq_trans
  2403   sq_ord_eq_less_trans ~> eq_below_trans
  2404   refl_less            ~> below_refl
  2405   trans_less           ~> below_trans
  2406   antisym_less         ~> below_antisym
  2407   antisym_less_inverse ~> po_eq_conv [THEN iffD1]
  2408   box_less             ~> box_below
  2409   rev_trans_less       ~> rev_below_trans
  2410   not_less2not_eq      ~> not_below2not_eq
  2411   less_UU_iff          ~> below_UU_iff
  2412   flat_less_iff        ~> flat_below_iff
  2413   adm_less             ~> adm_below
  2414   adm_not_less         ~> adm_not_below
  2415   adm_compact_not_less ~> adm_compact_not_below
  2416   less_fun_def         ~> below_fun_def
  2417   expand_fun_less      ~> fun_below_iff
  2418   less_fun_ext         ~> fun_belowI
  2419   less_discr_def       ~> below_discr_def
  2420   discr_less_eq        ~> discr_below_eq
  2421   less_unit_def        ~> below_unit_def
  2422   less_cprod_def       ~> below_prod_def
  2423   prod_lessI           ~> prod_belowI
  2424   Pair_less_iff        ~> Pair_below_iff
  2425   fst_less_iff         ~> fst_below_iff
  2426   snd_less_iff         ~> snd_below_iff
  2427   expand_cfun_less     ~> cfun_below_iff
  2428   less_cfun_ext        ~> cfun_belowI
  2429   injection_less       ~> injection_below
  2430   less_up_def          ~> below_up_def
  2431   not_Iup_less         ~> not_Iup_below
  2432   Iup_less             ~> Iup_below
  2433   up_less              ~> up_below
  2434   Def_inject_less_eq   ~> Def_below_Def
  2435   Def_less_is_eq       ~> Def_below_iff
  2436   spair_less_iff       ~> spair_below_iff
  2437   less_sprod           ~> below_sprod
  2438   spair_less           ~> spair_below
  2439   sfst_less_iff        ~> sfst_below_iff
  2440   ssnd_less_iff        ~> ssnd_below_iff
  2441   fix_least_less       ~> fix_least_below
  2442   dist_less_one        ~> dist_below_one
  2443   less_ONE             ~> below_ONE
  2444   ONE_less_iff         ~> ONE_below_iff
  2445   less_sinlD           ~> below_sinlD
  2446   less_sinrD           ~> below_sinrD
  2447 
  2448 
  2449 *** FOL and ZF ***
  2450 
  2451 * All constant names are now qualified internally and use proper
  2452 identifiers, e.g. "IFOL.eq" instead of "op =".  INCOMPATIBILITY.
  2453 
  2454 
  2455 *** ML ***
  2456 
  2457 * Antiquotation @{assert} inlines a function bool -> unit that raises
  2458 Fail if the argument is false.  Due to inlining the source position of
  2459 failed assertions is included in the error output.
  2460 
  2461 * Discontinued antiquotation @{theory_ref}, which is obsolete since ML
  2462 text is in practice always evaluated with a stable theory checkpoint.
  2463 Minor INCOMPATIBILITY, use (Theory.check_thy @{theory}) instead.
  2464 
  2465 * Antiquotation @{theory A} refers to theory A from the ancestry of
  2466 the current context, not any accidental theory loader state as before.
  2467 Potential INCOMPATIBILITY, subtle change in semantics.
  2468 
  2469 * Syntax.pretty_priority (default 0) configures the required priority
  2470 of pretty-printed output and thus affects insertion of parentheses.
  2471 
  2472 * Syntax.default_root (default "any") configures the inner syntax
  2473 category (nonterminal symbol) for parsing of terms.
  2474 
  2475 * Former exception Library.UnequalLengths now coincides with
  2476 ListPair.UnequalLengths.
  2477 
  2478 * Renamed structure MetaSimplifier to Raw_Simplifier.  Note that the
  2479 main functionality is provided by structure Simplifier.
  2480 
  2481 * Renamed raw "explode" function to "raw_explode" to emphasize its
  2482 meaning.  Note that internally to Isabelle, Symbol.explode is used in
  2483 almost all situations.
  2484 
  2485 * Discontinued obsolete function sys_error and exception SYS_ERROR.
  2486 See implementation manual for further details on exceptions in
  2487 Isabelle/ML.
  2488 
  2489 * Renamed setmp_noncritical to Unsynchronized.setmp to emphasize its
  2490 meaning.
  2491 
  2492 * Renamed structure PureThy to Pure_Thy and moved most of its
  2493 operations to structure Global_Theory, to emphasize that this is
  2494 rarely-used global-only stuff.
  2495 
  2496 * Discontinued Output.debug.  Minor INCOMPATIBILITY, use plain writeln
  2497 instead (or tracing for high-volume output).
  2498 
  2499 * Configuration option show_question_marks only affects regular pretty
  2500 printing of types and terms, not raw Term.string_of_vname.
  2501 
  2502 * ML_Context.thm and ML_Context.thms are no longer pervasive.  Rare
  2503 INCOMPATIBILITY, superseded by static antiquotations @{thm} and
  2504 @{thms} for most purposes.
  2505 
  2506 * ML structure Unsynchronized is never opened, not even in Isar
  2507 interaction mode as before.  Old Unsynchronized.set etc. have been
  2508 discontinued -- use plain := instead.  This should be *rare* anyway,
  2509 since modern tools always work via official context data, notably
  2510 configuration options.
  2511 
  2512 * Parallel and asynchronous execution requires special care concerning
  2513 interrupts.  Structure Exn provides some convenience functions that
  2514 avoid working directly with raw Interrupt.  User code must not absorb
  2515 interrupts -- intermediate handling (for cleanup etc.) needs to be
  2516 followed by re-raising of the original exception.  Another common
  2517 source of mistakes are "handle _" patterns, which make the meaning of
  2518 the program subject to physical effects of the environment.
  2519 
  2520 
  2521 
  2522 New in Isabelle2009-2 (June 2010)
  2523 ---------------------------------
  2524 
  2525 *** General ***
  2526 
  2527 * Authentic syntax for *all* logical entities (type classes, type
  2528 constructors, term constants): provides simple and robust
  2529 correspondence between formal entities and concrete syntax.  Within
  2530 the parse tree / AST representations, "constants" are decorated by
  2531 their category (class, type, const) and spelled out explicitly with
  2532 their full internal name.
  2533 
  2534 Substantial INCOMPATIBILITY concerning low-level syntax declarations
  2535 and translations (translation rules and translation functions in ML).
  2536 Some hints on upgrading:
  2537 
  2538   - Many existing uses of 'syntax' and 'translations' can be replaced
  2539     by more modern 'type_notation', 'notation' and 'abbreviation',
  2540     which are independent of this issue.
  2541 
  2542   - 'translations' require markup within the AST; the term syntax
  2543     provides the following special forms:
  2544 
  2545       CONST c   -- produces syntax version of constant c from context
  2546       XCONST c  -- literally c, checked as constant from context
  2547       c         -- literally c, if declared by 'syntax'
  2548 
  2549     Plain identifiers are treated as AST variables -- occasionally the
  2550     system indicates accidental variables via the error "rhs contains
  2551     extra variables".
  2552 
  2553     Type classes and type constructors are marked according to their
  2554     concrete syntax.  Some old translations rules need to be written
  2555     for the "type" category, using type constructor application
  2556     instead of pseudo-term application of the default category
  2557     "logic".
  2558 
  2559   - 'parse_translation' etc. in ML may use the following
  2560     antiquotations:
  2561 
  2562       @{class_syntax c}   -- type class c within parse tree / AST
  2563       @{term_syntax c}    -- type constructor c within parse tree / AST
  2564       @{const_syntax c}   -- ML version of "CONST c" above
  2565       @{syntax_const c}   -- literally c (checked wrt. 'syntax' declarations)
  2566 
  2567   - Literal types within 'typed_print_translations', i.e. those *not*
  2568     represented as pseudo-terms are represented verbatim.  Use @{class
  2569     c} or @{type_name c} here instead of the above syntax
  2570     antiquotations.
  2571 
  2572 Note that old non-authentic syntax was based on unqualified base
  2573 names, so all of the above "constant" names would coincide.  Recall
  2574 that 'print_syntax' and ML_command "set Syntax.trace_ast" help to
  2575 diagnose syntax problems.
  2576 
  2577 * Type constructors admit general mixfix syntax, not just infix.
  2578 
  2579 * Concrete syntax may be attached to local entities without a proof
  2580 body, too.  This works via regular mixfix annotations for 'fix',
  2581 'def', 'obtain' etc. or via the explicit 'write' command, which is
  2582 similar to the 'notation' command in theory specifications.
  2583 
  2584 * Discontinued unnamed infix syntax (legacy feature for many years) --
  2585 need to specify constant name and syntax separately.  Internal ML
  2586 datatype constructors have been renamed from InfixName to Infix etc.
  2587 Minor INCOMPATIBILITY.
  2588 
  2589 * Schematic theorem statements need to be explicitly markup as such,
  2590 via commands 'schematic_lemma', 'schematic_theorem',
  2591 'schematic_corollary'.  Thus the relevance of the proof is made
  2592 syntactically clear, which impacts performance in a parallel or
  2593 asynchronous interactive environment.  Minor INCOMPATIBILITY.
  2594 
  2595 * Use of cumulative prems via "!" in some proof methods has been
  2596 discontinued (old legacy feature).
  2597 
  2598 * References 'trace_simp' and 'debug_simp' have been replaced by
  2599 configuration options stored in the context. Enabling tracing (the
  2600 case of debugging is similar) in proofs works via
  2601 
  2602   using [[trace_simp = true]]
  2603 
  2604 Tracing is then active for all invocations of the simplifier in
  2605 subsequent goal refinement steps. Tracing may also still be enabled or
  2606 disabled via the ProofGeneral settings menu.
  2607 
  2608 * Separate commands 'hide_class', 'hide_type', 'hide_const',
  2609 'hide_fact' replace the former 'hide' KIND command.  Minor
  2610 INCOMPATIBILITY.
  2611 
  2612 * Improved parallelism of proof term normalization: usedir -p2 -q0 is
  2613 more efficient than combinations with -q1 or -q2.
  2614 
  2615 
  2616 *** Pure ***
  2617 
  2618 * Proofterms record type-class reasoning explicitly, using the
  2619 "unconstrain" operation internally.  This eliminates all sort
  2620 constraints from a theorem and proof, introducing explicit
  2621 OFCLASS-premises.  On the proof term level, this operation is
  2622 automatically applied at theorem boundaries, such that closed proofs
  2623 are always free of sort constraints.  INCOMPATIBILITY for tools that
  2624 inspect proof terms.
  2625 
  2626 * Local theory specifications may depend on extra type variables that
  2627 are not present in the result type -- arguments TYPE('a) :: 'a itself
  2628 are added internally.  For example:
  2629 
  2630   definition unitary :: bool where "unitary = (ALL (x::'a) y. x = y)"
  2631 
  2632 * Predicates of locales introduced by classes carry a mandatory
  2633 "class" prefix.  INCOMPATIBILITY.
  2634 
  2635 * Vacuous class specifications observe default sort.  INCOMPATIBILITY.
  2636 
  2637 * Old 'axclass' command has been discontinued.  INCOMPATIBILITY, use
  2638 'class' instead.
  2639 
  2640 * Command 'code_reflect' allows to incorporate generated ML code into
  2641 runtime environment; replaces immature code_datatype antiquotation.
  2642 INCOMPATIBILITY.
  2643 
  2644 * Code generator: simple concept for abstract datatypes obeying
  2645 invariants.
  2646 
  2647 * Code generator: details of internal data cache have no impact on the
  2648 user space functionality any longer.
  2649 
  2650 * Methods "unfold_locales" and "intro_locales" ignore non-locale
  2651 subgoals.  This is more appropriate for interpretations with 'where'.
  2652 INCOMPATIBILITY.
  2653 
  2654 * Command 'example_proof' opens an empty proof body.  This allows to
  2655 experiment with Isar, without producing any persistent result.
  2656 
  2657 * Commands 'type_notation' and 'no_type_notation' declare type syntax
  2658 within a local theory context, with explicit checking of the
  2659 constructors involved (in contrast to the raw 'syntax' versions).
  2660 
  2661 * Commands 'types' and 'typedecl' now work within a local theory
  2662 context -- without introducing dependencies on parameters or
  2663 assumptions, which is not possible in Isabelle/Pure.
  2664 
  2665 * Command 'defaultsort' has been renamed to 'default_sort', it works
  2666 within a local theory context.  Minor INCOMPATIBILITY.
  2667 
  2668 
  2669 *** HOL ***
  2670 
  2671 * Command 'typedef' now works within a local theory context -- without
  2672 introducing dependencies on parameters or assumptions, which is not
  2673 possible in Isabelle/Pure/HOL.  Note that the logical environment may
  2674 contain multiple interpretations of local typedefs (with different
  2675 non-emptiness proofs), even in a global theory context.
  2676 
  2677 * New package for quotient types.  Commands 'quotient_type' and
  2678 'quotient_definition' may be used for defining types and constants by
  2679 quotient constructions.  An example is the type of integers created by
  2680 quotienting pairs of natural numbers:
  2681 
  2682   fun
  2683     intrel :: "(nat * nat) => (nat * nat) => bool"
  2684   where
  2685     "intrel (x, y) (u, v) = (x + v = u + y)"
  2686 
  2687   quotient_type int = "nat * nat" / intrel
  2688     by (auto simp add: equivp_def expand_fun_eq)
  2689 
  2690   quotient_definition
  2691     "0::int" is "(0::nat, 0::nat)"
  2692 
  2693 The method "lifting" can be used to lift of theorems from the
  2694 underlying "raw" type to the quotient type.  The example
  2695 src/HOL/Quotient_Examples/FSet.thy includes such a quotient
  2696 construction and provides a reasoning infrastructure for finite sets.
  2697 
  2698 * Renamed Library/Quotient.thy to Library/Quotient_Type.thy to avoid
  2699 clash with new theory Quotient in Main HOL.
  2700 
  2701 * Moved the SMT binding into the main HOL session, eliminating
  2702 separate HOL-SMT session.
  2703 
  2704 * List membership infix mem operation is only an input abbreviation.
  2705 INCOMPATIBILITY.
  2706 
  2707 * Theory Library/Word.thy has been removed.  Use library Word/Word.thy
  2708 for future developements; former Library/Word.thy is still present in
  2709 the AFP entry RSAPPS.
  2710 
  2711 * Theorem Int.int_induct renamed to Int.int_of_nat_induct and is no
  2712 longer shadowed.  INCOMPATIBILITY.
  2713 
  2714 * Dropped theorem duplicate comp_arith; use semiring_norm instead.
  2715 INCOMPATIBILITY.
  2716 
  2717 * Dropped theorem RealPow.real_sq_order; use power2_le_imp_le instead.
  2718 INCOMPATIBILITY.
  2719 
  2720 * Dropped normalizing_semiring etc; use the facts in semiring classes
  2721 instead.  INCOMPATIBILITY.
  2722 
  2723 * Dropped several real-specific versions of lemmas about floor and
  2724 ceiling; use the generic lemmas from theory "Archimedean_Field"
  2725 instead.  INCOMPATIBILITY.
  2726 
  2727   floor_number_of_eq         ~> floor_number_of
  2728   le_floor_eq_number_of      ~> number_of_le_floor
  2729   le_floor_eq_zero           ~> zero_le_floor
  2730   le_floor_eq_one            ~> one_le_floor
  2731   floor_less_eq_number_of    ~> floor_less_number_of
  2732   floor_less_eq_zero         ~> floor_less_zero
  2733   floor_less_eq_one          ~> floor_less_one
  2734   less_floor_eq_number_of    ~> number_of_less_floor
  2735   less_floor_eq_zero         ~> zero_less_floor
  2736   less_floor_eq_one          ~> one_less_floor
  2737   floor_le_eq_number_of      ~> floor_le_number_of
  2738   floor_le_eq_zero           ~> floor_le_zero
  2739   floor_le_eq_one            ~> floor_le_one
  2740   floor_subtract_number_of   ~> floor_diff_number_of
  2741   floor_subtract_one         ~> floor_diff_one
  2742   ceiling_number_of_eq       ~> ceiling_number_of
  2743   ceiling_le_eq_number_of    ~> ceiling_le_number_of
  2744   ceiling_le_zero_eq         ~> ceiling_le_zero
  2745   ceiling_le_eq_one          ~> ceiling_le_one
  2746   less_ceiling_eq_number_of  ~> number_of_less_ceiling
  2747   less_ceiling_eq_zero       ~> zero_less_ceiling
  2748   less_ceiling_eq_one        ~> one_less_ceiling
  2749   ceiling_less_eq_number_of  ~> ceiling_less_number_of
  2750   ceiling_less_eq_zero       ~> ceiling_less_zero
  2751   ceiling_less_eq_one        ~> ceiling_less_one
  2752   le_ceiling_eq_number_of    ~> number_of_le_ceiling
  2753   le_ceiling_eq_zero         ~> zero_le_ceiling
  2754   le_ceiling_eq_one          ~> one_le_ceiling
  2755   ceiling_subtract_number_of ~> ceiling_diff_number_of
  2756   ceiling_subtract_one       ~> ceiling_diff_one
  2757 
  2758 * Theory "Finite_Set": various folding_XXX locales facilitate the
  2759 application of the various fold combinators on finite sets.
  2760 
  2761 * Library theory "RBT" renamed to "RBT_Impl"; new library theory "RBT"
  2762 provides abstract red-black tree type which is backed by "RBT_Impl" as
  2763 implementation.  INCOMPATIBILTY.
  2764 
  2765 * Theory Library/Coinductive_List has been removed -- superseded by
  2766 AFP/thys/Coinductive.
  2767 
  2768 * Theory PReal, including the type "preal" and related operations, has
  2769 been removed.  INCOMPATIBILITY.
  2770 
  2771 * Real: new development using Cauchy Sequences.
  2772 
  2773 * Split off theory "Big_Operators" containing setsum, setprod,
  2774 Inf_fin, Sup_fin, Min, Max from theory Finite_Set.  INCOMPATIBILITY.
  2775 
  2776 * Theory "Rational" renamed to "Rat", for consistency with "Nat",
  2777 "Int" etc.  INCOMPATIBILITY.
  2778 
  2779 * Constant Rat.normalize needs to be qualified.  INCOMPATIBILITY.
  2780 
  2781 * New set of rules "ac_simps" provides combined assoc / commute
  2782 rewrites for all interpretations of the appropriate generic locales.
  2783 
  2784 * Renamed theory "OrderedGroup" to "Groups" and split theory
  2785 "Ring_and_Field" into theories "Rings" and "Fields"; for more
  2786 appropriate and more consistent names suitable for name prefixes
  2787 within the HOL theories.  INCOMPATIBILITY.
  2788 
  2789 * Some generic constants have been put to appropriate theories:
  2790   - less_eq, less: Orderings
  2791   - zero, one, plus, minus, uminus, times, abs, sgn: Groups
  2792   - inverse, divide: Rings
  2793 INCOMPATIBILITY.
  2794 
  2795 * More consistent naming of type classes involving orderings (and
  2796 lattices):
  2797 
  2798     lower_semilattice                   ~> semilattice_inf
  2799     upper_semilattice                   ~> semilattice_sup
  2800 
  2801     dense_linear_order                  ~> dense_linorder
  2802 
  2803     pordered_ab_group_add               ~> ordered_ab_group_add
  2804     pordered_ab_group_add_abs           ~> ordered_ab_group_add_abs
  2805     pordered_ab_semigroup_add           ~> ordered_ab_semigroup_add
  2806     pordered_ab_semigroup_add_imp_le    ~> ordered_ab_semigroup_add_imp_le
  2807     pordered_cancel_ab_semigroup_add    ~> ordered_cancel_ab_semigroup_add
  2808     pordered_cancel_comm_semiring       ~> ordered_cancel_comm_semiring
  2809     pordered_cancel_semiring            ~> ordered_cancel_semiring
  2810     pordered_comm_monoid_add            ~> ordered_comm_monoid_add
  2811     pordered_comm_ring                  ~> ordered_comm_ring
  2812     pordered_comm_semiring              ~> ordered_comm_semiring
  2813     pordered_ring                       ~> ordered_ring
  2814     pordered_ring_abs                   ~> ordered_ring_abs
  2815     pordered_semiring                   ~> ordered_semiring
  2816 
  2817     ordered_ab_group_add                ~> linordered_ab_group_add
  2818     ordered_ab_semigroup_add            ~> linordered_ab_semigroup_add
  2819     ordered_cancel_ab_semigroup_add     ~> linordered_cancel_ab_semigroup_add
  2820     ordered_comm_semiring_strict        ~> linordered_comm_semiring_strict
  2821     ordered_field                       ~> linordered_field
  2822     ordered_field_no_lb                 ~> linordered_field_no_lb
  2823     ordered_field_no_ub                 ~> linordered_field_no_ub
  2824     ordered_field_dense_linear_order    ~> dense_linordered_field
  2825     ordered_idom                        ~> linordered_idom
  2826     ordered_ring                        ~> linordered_ring
  2827     ordered_ring_le_cancel_factor       ~> linordered_ring_le_cancel_factor
  2828     ordered_ring_less_cancel_factor     ~> linordered_ring_less_cancel_factor
  2829     ordered_ring_strict                 ~> linordered_ring_strict
  2830     ordered_semidom                     ~> linordered_semidom
  2831     ordered_semiring                    ~> linordered_semiring
  2832     ordered_semiring_1                  ~> linordered_semiring_1
  2833     ordered_semiring_1_strict           ~> linordered_semiring_1_strict
  2834     ordered_semiring_strict             ~> linordered_semiring_strict
  2835 
  2836   The following slightly odd type classes have been moved to a
  2837   separate theory Library/Lattice_Algebras:
  2838 
  2839     lordered_ab_group_add               ~> lattice_ab_group_add
  2840     lordered_ab_group_add_abs           ~> lattice_ab_group_add_abs
  2841     lordered_ab_group_add_meet          ~> semilattice_inf_ab_group_add
  2842     lordered_ab_group_add_join          ~> semilattice_sup_ab_group_add
  2843     lordered_ring                       ~> lattice_ring
  2844 
  2845 INCOMPATIBILITY.
  2846 
  2847 * Refined field classes:
  2848   - classes division_ring_inverse_zero, field_inverse_zero,
  2849     linordered_field_inverse_zero include rule inverse 0 = 0 --
  2850     subsumes former division_by_zero class;
  2851   - numerous lemmas have been ported from field to division_ring.
  2852 INCOMPATIBILITY.
  2853 
  2854 * Refined algebra theorem collections:
  2855   - dropped theorem group group_simps, use algebra_simps instead;
  2856   - dropped theorem group ring_simps, use field_simps instead;
  2857   - proper theorem collection field_simps subsumes former theorem
  2858     groups field_eq_simps and field_simps;
  2859   - dropped lemma eq_minus_self_iff which is a duplicate for
  2860     equal_neg_zero.
  2861 INCOMPATIBILITY.
  2862 
  2863 * Theory Finite_Set and List: some lemmas have been generalized from
  2864 sets to lattices:
  2865 
  2866   fun_left_comm_idem_inter      ~> fun_left_comm_idem_inf
  2867   fun_left_comm_idem_union      ~> fun_left_comm_idem_sup
  2868   inter_Inter_fold_inter        ~> inf_Inf_fold_inf
  2869   union_Union_fold_union        ~> sup_Sup_fold_sup
  2870   Inter_fold_inter              ~> Inf_fold_inf
  2871   Union_fold_union              ~> Sup_fold_sup
  2872   inter_INTER_fold_inter        ~> inf_INFI_fold_inf
  2873   union_UNION_fold_union        ~> sup_SUPR_fold_sup
  2874   INTER_fold_inter              ~> INFI_fold_inf
  2875   UNION_fold_union              ~> SUPR_fold_sup
  2876 
  2877 * Theory "Complete_Lattice": lemmas top_def and bot_def have been
  2878 replaced by the more convenient lemmas Inf_empty and Sup_empty.
  2879 Dropped lemmas Inf_insert_simp and Sup_insert_simp, which are subsumed
  2880 by Inf_insert and Sup_insert.  Lemmas Inf_UNIV and Sup_UNIV replace
  2881 former Inf_Univ and Sup_Univ.  Lemmas inf_top_right and sup_bot_right
  2882 subsume inf_top and sup_bot respectively.  INCOMPATIBILITY.
  2883 
  2884 * Reorganized theory Multiset: swapped notation of pointwise and
  2885 multiset order:
  2886 
  2887   - pointwise ordering is instance of class order with standard syntax
  2888     <= and <;
  2889   - multiset ordering has syntax <=# and <#; partial order properties
  2890     are provided by means of interpretation with prefix
  2891     multiset_order;
  2892   - less duplication, less historical organization of sections,
  2893     conversion from associations lists to multisets, rudimentary code
  2894     generation;
  2895   - use insert_DiffM2 [symmetric] instead of elem_imp_eq_diff_union,
  2896     if needed.
  2897 
  2898 Renamed:
  2899 
  2900   multiset_eq_conv_count_eq  ~>  multiset_ext_iff
  2901   multi_count_ext  ~>  multiset_ext
  2902   diff_union_inverse2  ~>  diff_union_cancelR
  2903 
  2904 INCOMPATIBILITY.
  2905 
  2906 * Theory Permutation: replaced local "remove" by List.remove1.
  2907 
  2908 * Code generation: ML and OCaml code is decorated with signatures.
  2909 
  2910 * Theory List: added transpose.
  2911 
  2912 * Library/Nat_Bijection.thy is a collection of bijective functions
  2913 between nat and other types, which supersedes the older libraries
  2914 Library/Nat_Int_Bij.thy and HOLCF/NatIso.thy.  INCOMPATIBILITY.
  2915 
  2916   Constants:
  2917   Nat_Int_Bij.nat2_to_nat         ~> prod_encode
  2918   Nat_Int_Bij.nat_to_nat2         ~> prod_decode
  2919   Nat_Int_Bij.int_to_nat_bij      ~> int_encode
  2920   Nat_Int_Bij.nat_to_int_bij      ~> int_decode
  2921   Countable.pair_encode           ~> prod_encode
  2922   NatIso.prod2nat                 ~> prod_encode
  2923   NatIso.nat2prod                 ~> prod_decode
  2924   NatIso.sum2nat                  ~> sum_encode
  2925   NatIso.nat2sum                  ~> sum_decode
  2926   NatIso.list2nat                 ~> list_encode
  2927   NatIso.nat2list                 ~> list_decode
  2928   NatIso.set2nat                  ~> set_encode
  2929   NatIso.nat2set                  ~> set_decode
  2930 
  2931   Lemmas:
  2932   Nat_Int_Bij.bij_nat_to_int_bij  ~> bij_int_decode
  2933   Nat_Int_Bij.nat2_to_nat_inj     ~> inj_prod_encode
  2934   Nat_Int_Bij.nat2_to_nat_surj    ~> surj_prod_encode
  2935   Nat_Int_Bij.nat_to_nat2_inj     ~> inj_prod_decode
  2936   Nat_Int_Bij.nat_to_nat2_surj    ~> surj_prod_decode
  2937   Nat_Int_Bij.i2n_n2i_id          ~> int_encode_inverse
  2938   Nat_Int_Bij.n2i_i2n_id          ~> int_decode_inverse
  2939   Nat_Int_Bij.surj_nat_to_int_bij ~> surj_int_encode
  2940   Nat_Int_Bij.surj_int_to_nat_bij ~> surj_int_decode
  2941   Nat_Int_Bij.inj_nat_to_int_bij  ~> inj_int_encode
  2942   Nat_Int_Bij.inj_int_to_nat_bij  ~> inj_int_decode
  2943   Nat_Int_Bij.bij_nat_to_int_bij  ~> bij_int_encode
  2944   Nat_Int_Bij.bij_int_to_nat_bij  ~> bij_int_decode
  2945 
  2946 * Sledgehammer:
  2947   - Renamed ATP commands:
  2948     atp_info     ~> sledgehammer running_atps
  2949     atp_kill     ~> sledgehammer kill_atps
  2950     atp_messages ~> sledgehammer messages
  2951     atp_minimize ~> sledgehammer minimize
  2952     print_atps   ~> sledgehammer available_atps
  2953     INCOMPATIBILITY.
  2954   - Added user's manual ("isabelle doc sledgehammer").
  2955   - Added option syntax and "sledgehammer_params" to customize
  2956     Sledgehammer's behavior.  See the manual for details.
  2957   - Modified the Isar proof reconstruction code so that it produces
  2958     direct proofs rather than proofs by contradiction.  (This feature
  2959     is still experimental.)
  2960   - Made Isar proof reconstruction work for SPASS, remote ATPs, and in
  2961     full-typed mode.
  2962   - Added support for TPTP syntax for SPASS via the "spass_tptp" ATP.
  2963 
  2964 * Nitpick:
  2965   - Added and implemented "binary_ints" and "bits" options.
  2966   - Added "std" option and implemented support for nonstandard models.
  2967   - Added and implemented "finitize" option to improve the precision
  2968     of infinite datatypes based on a monotonicity analysis.
  2969   - Added support for quotient types.
  2970   - Added support for "specification" and "ax_specification"
  2971     constructs.
  2972   - Added support for local definitions (for "function" and
  2973     "termination" proofs).
  2974   - Added support for term postprocessors.
  2975   - Optimized "Multiset.multiset" and "FinFun.finfun".
  2976   - Improved efficiency of "destroy_constrs" optimization.
  2977   - Fixed soundness bugs related to "destroy_constrs" optimization and
  2978     record getters.
  2979   - Fixed soundness bug related to higher-order constructors.
  2980   - Fixed soundness bug when "full_descrs" is enabled.
  2981   - Improved precision of set constructs.
  2982   - Added "atoms" option.
  2983   - Added cache to speed up repeated Kodkod invocations on the same
  2984     problems.
  2985   - Renamed "MiniSatJNI", "zChaffJNI", "BerkMinAlloy", and
  2986     "SAT4JLight" to "MiniSat_JNI", "zChaff_JNI", "BerkMin_Alloy", and
  2987     "SAT4J_Light".  INCOMPATIBILITY.
  2988   - Removed "skolemize", "uncurry", "sym_break", "flatten_prop",
  2989     "sharing_depth", and "show_skolems" options.  INCOMPATIBILITY.
  2990   - Removed "nitpick_intro" attribute.  INCOMPATIBILITY.
  2991 
  2992 * Method "induct" now takes instantiations of the form t, where t is not
  2993   a variable, as a shorthand for "x == t", where x is a fresh variable.
  2994   If this is not intended, t has to be enclosed in parentheses.
  2995   By default, the equalities generated by definitional instantiations
  2996   are pre-simplified, which may cause parameters of inductive cases
  2997   to disappear, or may even delete some of the inductive cases.
  2998   Use "induct (no_simp)" instead of "induct" to restore the old
  2999   behaviour. The (no_simp) option is also understood by the "cases"
  3000   and "nominal_induct" methods, which now perform pre-simplification, too.
  3001   INCOMPATIBILITY.
  3002 
  3003 
  3004 *** HOLCF ***
  3005 
  3006 * Variable names in lemmas generated by the domain package have
  3007 changed; the naming scheme is now consistent with the HOL datatype
  3008 package.  Some proof scripts may be affected, INCOMPATIBILITY.
  3009 
  3010 * The domain package no longer defines the function "foo_copy" for
  3011 recursive domain "foo".  The reach lemma is now stated directly in
  3012 terms of "foo_take".  Lemmas and proofs that mention "foo_copy" must
  3013 be reformulated in terms of "foo_take", INCOMPATIBILITY.
  3014 
  3015 * Most definedness lemmas generated by the domain package (previously
  3016 of the form "x ~= UU ==> foo$x ~= UU") now have an if-and-only-if form
  3017 like "foo$x = UU <-> x = UU", which works better as a simp rule.
  3018 Proofs that used definedness lemmas as intro rules may break,
  3019 potential INCOMPATIBILITY.
  3020 
  3021 * Induction and casedist rules generated by the domain package now
  3022 declare proper case_names (one called "bottom", and one named for each
  3023 constructor).  INCOMPATIBILITY.
  3024 
  3025 * For mutually-recursive domains, separate "reach" and "take_lemma"
  3026 rules are generated for each domain, INCOMPATIBILITY.
  3027 
  3028   foo_bar.reach       ~> foo.reach  bar.reach
  3029   foo_bar.take_lemmas ~> foo.take_lemma  bar.take_lemma
  3030 
  3031 * Some lemmas generated by the domain package have been renamed for
  3032 consistency with the datatype package, INCOMPATIBILITY.
  3033 
  3034   foo.ind        ~> foo.induct
  3035   foo.finite_ind ~> foo.finite_induct
  3036   foo.coind      ~> foo.coinduct
  3037   foo.casedist   ~> foo.exhaust
  3038   foo.exhaust    ~> foo.nchotomy
  3039 
  3040 * For consistency with other definition packages, the fixrec package
  3041 now generates qualified theorem names, INCOMPATIBILITY.
  3042 
  3043   foo_simps  ~> foo.simps
  3044   foo_unfold ~> foo.unfold
  3045   foo_induct ~> foo.induct
  3046 
  3047 * The "fixrec_simp" attribute has been removed.  The "fixrec_simp"
  3048 method and internal fixrec proofs now use the default simpset instead.
  3049 INCOMPATIBILITY.
  3050 
  3051 * The "contlub" predicate has been removed.  Proof scripts should use
  3052 lemma contI2 in place of monocontlub2cont, INCOMPATIBILITY.
  3053 
  3054 * The "admw" predicate has been removed, INCOMPATIBILITY.
  3055 
  3056 * The constants cpair, cfst, and csnd have been removed in favor of
  3057 Pair, fst, and snd from Isabelle/HOL, INCOMPATIBILITY.
  3058 
  3059 
  3060 *** ML ***
  3061 
  3062 * Antiquotations for basic formal entities:
  3063 
  3064     @{class NAME}         -- type class
  3065     @{class_syntax NAME}  -- syntax representation of the above
  3066 
  3067     @{type_name NAME}     -- logical type
  3068     @{type_abbrev NAME}   -- type abbreviation
  3069     @{nonterminal NAME}   -- type of concrete syntactic category
  3070     @{type_syntax NAME}   -- syntax representation of any of the above
  3071 
  3072     @{const_name NAME}    -- logical constant (INCOMPATIBILITY)
  3073     @{const_abbrev NAME}  -- abbreviated constant
  3074     @{const_syntax NAME}  -- syntax representation of any of the above
  3075 
  3076 * Antiquotation @{syntax_const NAME} ensures that NAME refers to a raw
  3077 syntax constant (cf. 'syntax' command).
  3078 
  3079 * Antiquotation @{make_string} inlines a function to print arbitrary
  3080 values similar to the ML toplevel.  The result is compiler dependent
  3081 and may fall back on "?" in certain situations.
  3082 
  3083 * Diagnostic commands 'ML_val' and 'ML_command' may refer to
  3084 antiquotations @{Isar.state} and @{Isar.goal}.  This replaces impure
  3085 Isar.state() and Isar.goal(), which belong to the old TTY loop and do
  3086 not work with the asynchronous Isar document model.
  3087 
  3088 * Configuration options now admit dynamic default values, depending on
  3089 the context or even global references.
  3090 
  3091 * SHA1.digest digests strings according to SHA-1 (see RFC 3174).  It
  3092 uses an efficient external library if available (for Poly/ML).
  3093 
  3094 * Renamed some important ML structures, while keeping the old names
  3095 for some time as aliases within the structure Legacy:
  3096 
  3097   OuterKeyword  ~>  Keyword
  3098   OuterLex      ~>  Token
  3099   OuterParse    ~>  Parse
  3100   OuterSyntax   ~>  Outer_Syntax
  3101   PrintMode     ~>  Print_Mode
  3102   SpecParse     ~>  Parse_Spec
  3103   ThyInfo       ~>  Thy_Info
  3104   ThyLoad       ~>  Thy_Load
  3105   ThyOutput     ~>  Thy_Output
  3106   TypeInfer     ~>  Type_Infer
  3107 
  3108 Note that "open Legacy" simplifies porting of sources, but forgetting
  3109 to remove it again will complicate porting again in the future.
  3110 
  3111 * Most operations that refer to a global context are named
  3112 accordingly, e.g. Simplifier.global_context or
  3113 ProofContext.init_global.  There are some situations where a global
  3114 context actually works, but under normal circumstances one needs to
  3115 pass the proper local context through the code!
  3116 
  3117 * Discontinued old TheoryDataFun with its copy/init operation -- data
  3118 needs to be pure.  Functor Theory_Data_PP retains the traditional
  3119 Pretty.pp argument to merge, which is absent in the standard
  3120 Theory_Data version.
  3121 
  3122 * Sorts.certify_sort and derived "cert" operations for types and terms
  3123 no longer minimize sorts.  Thus certification at the boundary of the
  3124 inference kernel becomes invariant under addition of class relations,
  3125 which is an important monotonicity principle.  Sorts are now minimized
  3126 in the syntax layer only, at the boundary between the end-user and the
  3127 system.  Subtle INCOMPATIBILITY, may have to use Sign.minimize_sort
  3128 explicitly in rare situations.
  3129 
  3130 * Renamed old-style Drule.standard to Drule.export_without_context, to
  3131 emphasize that this is in no way a standard operation.
  3132 INCOMPATIBILITY.
  3133 
  3134 * Subgoal.FOCUS (and variants): resulting goal state is normalized as
  3135 usual for resolution.  Rare INCOMPATIBILITY.
  3136 
  3137 * Renamed varify/unvarify operations to varify_global/unvarify_global
  3138 to emphasize that these only work in a global situation (which is
  3139 quite rare).
  3140 
  3141 * Curried take and drop in library.ML; negative length is interpreted
  3142 as infinity (as in chop).  Subtle INCOMPATIBILITY.
  3143 
  3144 * Proof terms: type substitutions on proof constants now use canonical
  3145 order of type variables.  INCOMPATIBILITY for tools working with proof
  3146 terms.
  3147 
  3148 * Raw axioms/defs may no longer carry sort constraints, and raw defs
  3149 may no longer carry premises.  User-level specifications are
  3150 transformed accordingly by Thm.add_axiom/add_def.
  3151 
  3152 
  3153 *** System ***
  3154 
  3155 * Discontinued special HOL_USEDIR_OPTIONS for the main HOL image;
  3156 ISABELLE_USEDIR_OPTIONS applies uniformly to all sessions.  Note that
  3157 proof terms are enabled unconditionally in the new HOL-Proofs image.
  3158 
  3159 * Discontinued old ISABELLE and ISATOOL environment settings (legacy
  3160 feature since Isabelle2009).  Use ISABELLE_PROCESS and ISABELLE_TOOL,
  3161 respectively.
  3162 
  3163 * Old lib/scripts/polyml-platform is superseded by the
  3164 ISABELLE_PLATFORM setting variable, which defaults to the 32 bit
  3165 variant, even on a 64 bit machine.  The following example setting
  3166 prefers 64 bit if available:
  3167 
  3168   ML_PLATFORM="${ISABELLE_PLATFORM64:-$ISABELLE_PLATFORM}"
  3169 
  3170 * The preliminary Isabelle/jEdit application demonstrates the emerging
  3171 Isabelle/Scala layer for advanced prover interaction and integration.
  3172 See src/Tools/jEdit or "isabelle jedit" provided by the properly built
  3173 component.
  3174 
  3175 * "IsabelleText" is a Unicode font derived from Bitstream Vera Mono
  3176 and Bluesky TeX fonts.  It provides the usual Isabelle symbols,
  3177 similar to the default assignment of the document preparation system
  3178 (cf. isabellesym.sty).  The Isabelle/Scala class Isabelle_System
  3179 provides some operations for direct access to the font without asking
  3180 the user for manual installation.
  3181 
  3182 
  3183 
  3184 New in Isabelle2009-1 (December 2009)
  3185 -------------------------------------
  3186 
  3187 *** General ***
  3188 
  3189 * Discontinued old form of "escaped symbols" such as \\<forall>.  Only
  3190 one backslash should be used, even in ML sources.
  3191 
  3192 
  3193 *** Pure ***
  3194 
  3195 * Locale interpretation propagates mixins along the locale hierarchy.
  3196 The currently only available mixins are the equations used to map
  3197 local definitions to terms of the target domain of an interpretation.
  3198 
  3199 * Reactivated diagnostic command 'print_interps'.  Use "print_interps
  3200 loc" to print all interpretations of locale "loc" in the theory.
  3201 Interpretations in proofs are not shown.
  3202 
  3203 * Thoroughly revised locales tutorial.  New section on conditional
  3204 interpretation.
  3205 
  3206 * On instantiation of classes, remaining undefined class parameters
  3207 are formally declared.  INCOMPATIBILITY.
  3208 
  3209 
  3210 *** Document preparation ***
  3211 
  3212 * New generalized style concept for printing terms: @{foo (style) ...}
  3213 instead of @{foo_style style ...}  (old form is still retained for
  3214 backward compatibility).  Styles can be also applied for
  3215 antiquotations prop, term_type and typeof.
  3216 
  3217 
  3218 *** HOL ***
  3219 
  3220 * New proof method "smt" for a combination of first-order logic with
  3221 equality, linear and nonlinear (natural/integer/real) arithmetic, and
  3222 fixed-size bitvectors; there is also basic support for higher-order
  3223 features (esp. lambda abstractions).  It is an incomplete decision
  3224 procedure based on external SMT solvers using the oracle mechanism;
  3225 for the SMT solver Z3, this method is proof-producing.  Certificates
  3226 are provided to avoid calling the external solvers solely for
  3227 re-checking proofs.  Due to a remote SMT service there is no need for
  3228 installing SMT solvers locally.  See src/HOL/SMT.
  3229 
  3230 * New commands to load and prove verification conditions generated by
  3231 the Boogie program verifier or derived systems (e.g. the Verifying C
  3232 Compiler (VCC) or Spec#).  See src/HOL/Boogie.
  3233 
  3234 * New counterexample generator tool 'nitpick' based on the Kodkod
  3235 relational model finder.  See src/HOL/Tools/Nitpick and
  3236 src/HOL/Nitpick_Examples.
  3237 
  3238 * New commands 'code_pred' and 'values' to invoke the predicate
  3239 compiler and to enumerate values of inductive predicates.
  3240 
  3241 * A tabled implementation of the reflexive transitive closure.
  3242 
  3243 * New implementation of quickcheck uses generic code generator;
  3244 default generators are provided for all suitable HOL types, records
  3245 and datatypes.  Old quickcheck can be re-activated importing theory
  3246 Library/SML_Quickcheck.
  3247 
  3248 * New testing tool Mirabelle for automated proof tools.  Applies
  3249 several tools and tactics like sledgehammer, metis, or quickcheck, to
  3250 every proof step in a theory.  To be used in batch mode via the
  3251 "mirabelle" utility.
  3252 
  3253 * New proof method "sos" (sum of squares) for nonlinear real
  3254 arithmetic (originally due to John Harison). It requires theory
  3255 Library/Sum_Of_Squares.  It is not a complete decision procedure but
  3256 works well in practice on quantifier-free real arithmetic with +, -,
  3257 *, ^, =, <= and <, i.e. boolean combinations of equalities and
  3258 inequalities between polynomials.  It makes use of external
  3259 semidefinite programming solvers.  Method "sos" generates a
  3260 certificate that can be pasted into the proof thus avoiding the need
  3261 to call an external tool every time the proof is checked.  See
  3262 src/HOL/Library/Sum_Of_Squares.
  3263 
  3264 * New method "linarith" invokes existing linear arithmetic decision
  3265 procedure only.
  3266 
  3267 * New command 'atp_minimal' reduces result produced by Sledgehammer.
  3268 
  3269 * New Sledgehammer option "Full Types" in Proof General settings menu.
  3270 Causes full type information to be output to the ATPs.  This slows
  3271 ATPs down considerably but eliminates a source of unsound "proofs"
  3272 that fail later.
  3273 
  3274 * New method "metisFT": A version of metis that uses full type
  3275 information in order to avoid failures of proof reconstruction.
  3276 
  3277 * New evaluator "approximate" approximates an real valued term using
  3278 the same method as the approximation method.
  3279 
  3280 * Method "approximate" now supports arithmetic expressions as
  3281 boundaries of intervals and implements interval splitting and Taylor
  3282 series expansion.
  3283 
  3284 * ML antiquotation @{code_datatype} inserts definition of a datatype
  3285 generated by the code generator; e.g. see src/HOL/Predicate.thy.
  3286 
  3287 * New theory SupInf of the supremum and infimum operators for sets of
  3288 reals.
  3289 
  3290 * New theory Probability, which contains a development of measure
  3291 theory, eventually leading to Lebesgue integration and probability.
  3292 
  3293 * Extended Multivariate Analysis to include derivation and Brouwer's
  3294 fixpoint theorem.
  3295 
  3296 * Reorganization of number theory, INCOMPATIBILITY:
  3297   - new number theory development for nat and int, in theories Divides
  3298     and GCD as well as in new session Number_Theory
  3299   - some constants and facts now suffixed with _nat and _int
  3300     accordingly
  3301   - former session NumberTheory now named Old_Number_Theory, including
  3302     theories Legacy_GCD and Primes (prefer Number_Theory if possible)
  3303   - moved theory Pocklington from src/HOL/Library to
  3304     src/HOL/Old_Number_Theory
  3305 
  3306 * Theory GCD includes functions Gcd/GCD and Lcm/LCM for the gcd and
  3307 lcm of finite and infinite sets. It is shown that they form a complete
  3308 lattice.
  3309 
  3310 * Class semiring_div requires superclass no_zero_divisors and proof of
  3311 div_mult_mult1; theorems div_mult_mult1, div_mult_mult2,
  3312 div_mult_mult1_if, div_mult_mult1 and div_mult_mult2 have been
  3313 generalized to class semiring_div, subsuming former theorems
  3314 zdiv_zmult_zmult1, zdiv_zmult_zmult1_if, zdiv_zmult_zmult1 and
  3315 zdiv_zmult_zmult2.  div_mult_mult1 is now [simp] by default.
  3316 INCOMPATIBILITY.
  3317 
  3318 * Refinements to lattice classes and sets:
  3319   - less default intro/elim rules in locale variant, more default
  3320     intro/elim rules in class variant: more uniformity
  3321   - lemma ge_sup_conv renamed to le_sup_iff, in accordance with
  3322     le_inf_iff
  3323   - dropped lemma alias inf_ACI for inf_aci (same for sup_ACI and
  3324     sup_aci)
  3325   - renamed ACI to inf_sup_aci
  3326   - new class "boolean_algebra"
  3327   - class "complete_lattice" moved to separate theory
  3328     "Complete_Lattice"; corresponding constants (and abbreviations)
  3329     renamed and with authentic syntax:
  3330     Set.Inf ~>    Complete_Lattice.Inf
  3331     Set.Sup ~>    Complete_Lattice.Sup
  3332     Set.INFI ~>   Complete_Lattice.INFI
  3333     Set.SUPR ~>   Complete_Lattice.SUPR
  3334     Set.Inter ~>  Complete_Lattice.Inter
  3335     Set.Union ~>  Complete_Lattice.Union
  3336     Set.INTER ~>  Complete_Lattice.INTER
  3337     Set.UNION ~>  Complete_Lattice.UNION
  3338   - authentic syntax for
  3339     Set.Pow
  3340     Set.image
  3341   - mere abbreviations:
  3342     Set.empty               (for bot)
  3343     Set.UNIV                (for top)
  3344     Set.inter               (for inf, formerly Set.Int)
  3345     Set.union               (for sup, formerly Set.Un)
  3346     Complete_Lattice.Inter  (for Inf)
  3347     Complete_Lattice.Union  (for Sup)
  3348     Complete_Lattice.INTER  (for INFI)
  3349     Complete_Lattice.UNION  (for SUPR)
  3350   - object-logic definitions as far as appropriate
  3351 
  3352 INCOMPATIBILITY.  Care is required when theorems Int_subset_iff or
  3353 Un_subset_iff are explicitly deleted as default simp rules; then also
  3354 their lattice counterparts le_inf_iff and le_sup_iff have to be
  3355 deleted to achieve the desired effect.
  3356 
  3357 * Rules inf_absorb1, inf_absorb2, sup_absorb1, sup_absorb2 are no simp
  3358 rules by default any longer; the same applies to min_max.inf_absorb1
  3359 etc.  INCOMPATIBILITY.
  3360 
  3361 * Rules sup_Int_eq and sup_Un_eq are no longer declared as
  3362 pred_set_conv by default.  INCOMPATIBILITY.
  3363 
  3364 * Power operations on relations and functions are now one dedicated
  3365 constant "compow" with infix syntax "^^".  Power operation on
  3366 multiplicative monoids retains syntax "^" and is now defined generic
  3367 in class power.  INCOMPATIBILITY.
  3368 
  3369 * Relation composition "R O S" now has a more standard argument order:
  3370 "R O S = {(x, z). EX y. (x, y) : R & (y, z) : S}".  INCOMPATIBILITY,
  3371 rewrite propositions with "S O R" --> "R O S". Proofs may occasionally
  3372 break, since the O_assoc rule was not rewritten like this.  Fix using
  3373 O_assoc[symmetric].  The same applies to the curried version "R OO S".
  3374 
  3375 * Function "Inv" is renamed to "inv_into" and function "inv" is now an
  3376 abbreviation for "inv_into UNIV".  Lemmas are renamed accordingly.
  3377 INCOMPATIBILITY.
  3378 
  3379 * Most rules produced by inductive and datatype package have mandatory
  3380 prefixes.  INCOMPATIBILITY.
  3381 
  3382 * Changed "DERIV_intros" to a dynamic fact, which can be augmented by
  3383 the attribute of the same name.  Each of the theorems in the list
  3384 DERIV_intros assumes composition with an additional function and
  3385 matches a variable to the derivative, which has to be solved by the
  3386 Simplifier.  Hence (auto intro!: DERIV_intros) computes the derivative
  3387 of most elementary terms.  Former Maclauren.DERIV_tac and
  3388 Maclauren.deriv_tac should be replaced by (auto intro!: DERIV_intros).
  3389 INCOMPATIBILITY.
  3390 
  3391 * Code generator attributes follow the usual underscore convention:
  3392     code_unfold     replaces    code unfold
  3393     code_post       replaces    code post
  3394     etc.
  3395   INCOMPATIBILITY.
  3396 
  3397 * Renamed methods:
  3398     sizechange -> size_change
  3399     induct_scheme -> induction_schema
  3400   INCOMPATIBILITY.
  3401 
  3402 * Discontinued abbreviation "arbitrary" of constant "undefined".
  3403 INCOMPATIBILITY, use "undefined" directly.
  3404 
  3405 * Renamed theorems:
  3406     Suc_eq_add_numeral_1 -> Suc_eq_plus1
  3407     Suc_eq_add_numeral_1_left -> Suc_eq_plus1_left
  3408     Suc_plus1 -> Suc_eq_plus1
  3409     *anti_sym -> *antisym*
  3410     vector_less_eq_def -> vector_le_def
  3411   INCOMPATIBILITY.
  3412 
  3413 * Added theorem List.map_map as [simp].  Removed List.map_compose.
  3414 INCOMPATIBILITY.
  3415 
  3416 * Removed predicate "M hassize n" (<--> card M = n & finite M).
  3417 INCOMPATIBILITY.
  3418 
  3419 
  3420 *** HOLCF ***
  3421 
  3422 * Theory Representable defines a class "rep" of domains that are
  3423 representable (via an ep-pair) in the universal domain type "udom".
  3424 Instances are provided for all type constructors defined in HOLCF.
  3425 
  3426 * The 'new_domain' command is a purely definitional version of the
  3427 domain package, for representable domains.  Syntax is identical to the
  3428 old domain package.  The 'new_domain' package also supports indirect
  3429 recursion using previously-defined type constructors.  See
  3430 src/HOLCF/ex/New_Domain.thy for examples.
  3431 
  3432 * Method "fixrec_simp" unfolds one step of a fixrec-defined constant
  3433 on the left-hand side of an equation, and then performs
  3434 simplification.  Rewriting is done using rules declared with the
  3435 "fixrec_simp" attribute.  The "fixrec_simp" method is intended as a
  3436 replacement for "fixpat"; see src/HOLCF/ex/Fixrec_ex.thy for examples.
  3437 
  3438 * The pattern-match compiler in 'fixrec' can now handle constructors
  3439 with HOL function types.  Pattern-match combinators for the Pair
  3440 constructor are pre-configured.
  3441 
  3442 * The 'fixrec' package now produces better fixed-point induction rules
  3443 for mutually-recursive definitions:  Induction rules have conclusions
  3444 of the form "P foo bar" instead of "P <foo, bar>".
  3445 
  3446 * The constant "sq_le" (with infix syntax "<<" or "\<sqsubseteq>") has
  3447 been renamed to "below".  The name "below" now replaces "less" in many
  3448 theorem names.  (Legacy theorem names using "less" are still supported
  3449 as well.)
  3450 
  3451 * The 'fixrec' package now supports "bottom patterns".  Bottom
  3452 patterns can be used to generate strictness rules, or to make
  3453 functions more strict (much like the bang-patterns supported by the
  3454 Glasgow Haskell Compiler).  See src/HOLCF/ex/Fixrec_ex.thy for
  3455 examples.
  3456 
  3457 
  3458 *** ML ***
  3459 
  3460 * Support for Poly/ML 5.3.0, with improved reporting of compiler
  3461 errors and run-time exceptions, including detailed source positions.
  3462 
  3463 * Structure Name_Space (formerly NameSpace) now manages uniquely
  3464 identified entries, with some additional information such as source
  3465 position, logical grouping etc.
  3466 
  3467 * Theory and context data is now introduced by the simplified and
  3468 modernized functors Theory_Data, Proof_Data, Generic_Data.  Data needs
  3469 to be pure, but the old TheoryDataFun for mutable data (with explicit
  3470 copy operation) is still available for some time.
  3471 
  3472 * Structure Synchronized (cf. src/Pure/Concurrent/synchronized.ML)
  3473 provides a high-level programming interface to synchronized state
  3474 variables with atomic update.  This works via pure function
  3475 application within a critical section -- its runtime should be as
  3476 short as possible; beware of deadlocks if critical code is nested,
  3477 either directly or indirectly via other synchronized variables!
  3478 
  3479 * Structure Unsynchronized (cf. src/Pure/ML-Systems/unsynchronized.ML)
  3480 wraps raw ML references, explicitly indicating their non-thread-safe
  3481 behaviour.  The Isar toplevel keeps this structure open, to
  3482 accommodate Proof General as well as quick and dirty interactive
  3483 experiments with references.
  3484 
  3485 * PARALLEL_CHOICE and PARALLEL_GOALS provide basic support for
  3486 parallel tactical reasoning.
  3487 
  3488 * Tacticals Subgoal.FOCUS, Subgoal.FOCUS_PREMS, Subgoal.FOCUS_PARAMS
  3489 are similar to SUBPROOF, but are slightly more flexible: only the
  3490 specified parts of the subgoal are imported into the context, and the
  3491 body tactic may introduce new subgoals and schematic variables.
  3492 
  3493 * Old tactical METAHYPS, which does not observe the proof context, has
  3494 been renamed to Old_Goals.METAHYPS and awaits deletion.  Use SUBPROOF
  3495 or Subgoal.FOCUS etc.
  3496 
  3497 * Renamed functor TableFun to Table, and GraphFun to Graph.  (Since
  3498 functors have their own ML name space there is no point to mark them
  3499 separately.)  Minor INCOMPATIBILITY.
  3500 
  3501 * Renamed NamedThmsFun to Named_Thms.  INCOMPATIBILITY.
  3502 
  3503 * Renamed several structures FooBar to Foo_Bar.  Occasional,
  3504 INCOMPATIBILITY.
  3505 
  3506 * Operations of structure Skip_Proof no longer require quick_and_dirty
  3507 mode, which avoids critical setmp.
  3508 
  3509 * Eliminated old Attrib.add_attributes, Method.add_methods and related
  3510 combinators for "args".  INCOMPATIBILITY, need to use simplified
  3511 Attrib/Method.setup introduced in Isabelle2009.
  3512 
  3513 * Proper context for simpset_of, claset_of, clasimpset_of.  May fall
  3514 back on global_simpset_of, global_claset_of, global_clasimpset_of as
  3515 last resort.  INCOMPATIBILITY.
  3516 
  3517 * Display.pretty_thm now requires a proper context (cf. former
  3518 ProofContext.pretty_thm).  May fall back on Display.pretty_thm_global
  3519 or even Display.pretty_thm_without_context as last resort.
  3520 INCOMPATIBILITY.
  3521 
  3522 * Discontinued Display.pretty_ctyp/cterm etc.  INCOMPATIBILITY, use
  3523 Syntax.pretty_typ/term directly, preferably with proper context
  3524 instead of global theory.
  3525 
  3526 
  3527 *** System ***
  3528 
  3529 * Further fine tuning of parallel proof checking, scales up to 8 cores
  3530 (max. speedup factor 5.0).  See also Goal.parallel_proofs in ML and
  3531 usedir option -q.
  3532 
  3533 * Support for additional "Isabelle components" via etc/components, see
  3534 also the system manual.
  3535 
  3536 * The isabelle makeall tool now operates on all components with
  3537 IsaMakefile, not just hardwired "logics".
  3538 
  3539 * Removed "compress" option from isabelle-process and isabelle usedir;
  3540 this is always enabled.
  3541 
  3542 * Discontinued support for Poly/ML 4.x versions.
  3543 
  3544 * Isabelle tool "wwwfind" provides web interface for 'find_theorems'
  3545 on a given logic image.  This requires the lighttpd webserver and is
  3546 currently supported on Linux only.
  3547 
  3548 
  3549 
  3550 New in Isabelle2009 (April 2009)
  3551 --------------------------------
  3552 
  3553 *** General ***
  3554 
  3555 * Simplified main Isabelle executables, with less surprises on
  3556 case-insensitive file-systems (such as Mac OS).
  3557 
  3558   - The main Isabelle tool wrapper is now called "isabelle" instead of
  3559     "isatool."
  3560 
  3561   - The former "isabelle" alias for "isabelle-process" has been
  3562     removed (should rarely occur to regular users).
  3563 
  3564   - The former "isabelle-interface" and its alias "Isabelle" have been
  3565     removed (interfaces are now regular Isabelle tools).
  3566 
  3567 Within scripts and make files, the Isabelle environment variables
  3568 ISABELLE_TOOL and ISABELLE_PROCESS replace old ISATOOL and ISABELLE,
  3569 respectively.  (The latter are still available as legacy feature.)
  3570 
  3571 The old isabelle-interface wrapper could react in confusing ways if
  3572 the interface was uninstalled or changed otherwise.  Individual
  3573 interface tool configuration is now more explicit, see also the
  3574 Isabelle system manual.  In particular, Proof General is now available
  3575 via "isabelle emacs".
  3576 
  3577 INCOMPATIBILITY, need to adapt derivative scripts.  Users may need to
  3578 purge installed copies of Isabelle executables and re-run "isabelle
  3579 install -p ...", or use symlinks.
  3580 
  3581 * The default for ISABELLE_HOME_USER is now ~/.isabelle instead of the
  3582 old ~/isabelle, which was slightly non-standard and apt to cause
  3583 surprises on case-insensitive file-systems (such as Mac OS).
  3584 
  3585 INCOMPATIBILITY, need to move existing ~/isabelle/etc,
  3586 ~/isabelle/heaps, ~/isabelle/browser_info to the new place.  Special
  3587 care is required when using older releases of Isabelle.  Note that
  3588 ISABELLE_HOME_USER can be changed in Isabelle/etc/settings of any
  3589 Isabelle distribution, in order to use the new ~/.isabelle uniformly.
  3590 
  3591 * Proofs of fully specified statements are run in parallel on
  3592 multi-core systems.  A speedup factor of 2.5 to 3.2 can be expected on
  3593 a regular 4-core machine, if the initial heap space is made reasonably
  3594 large (cf. Poly/ML option -H).  (Requires Poly/ML 5.2.1 or later.)
  3595 
  3596 * The main reference manuals ("isar-ref", "implementation", and
  3597 "system") have been updated and extended.  Formally checked references
  3598 as hyperlinks are now available uniformly.
  3599 
  3600 
  3601 *** Pure ***
  3602 
  3603 * Complete re-implementation of locales.  INCOMPATIBILITY in several
  3604 respects.  The most important changes are listed below.  See the
  3605 Tutorial on Locales ("locales" manual) for details.
  3606 
  3607 - In locale expressions, instantiation replaces renaming.  Parameters
  3608 must be declared in a for clause.  To aid compatibility with previous
  3609 parameter inheritance, in locale declarations, parameters that are not
  3610 'touched' (instantiation position "_" or omitted) are implicitly added
  3611 with their syntax at the beginning of the for clause.
  3612 
  3613 - Syntax from abbreviations and definitions in locales is available in
  3614 locale expressions and context elements.  The latter is particularly
  3615 useful in locale declarations.
  3616 
  3617 - More flexible mechanisms to qualify names generated by locale
  3618 expressions.  Qualifiers (prefixes) may be specified in locale
  3619 expressions, and can be marked as mandatory (syntax: "name!:") or
  3620 optional (syntax "name?:").  The default depends for plain "name:"
  3621 depends on the situation where a locale expression is used: in
  3622 commands 'locale' and 'sublocale' prefixes are optional, in
  3623 'interpretation' and 'interpret' prefixes are mandatory.  The old
  3624 implicit qualifiers derived from the parameter names of a locale are
  3625 no longer generated.
  3626 
  3627 - Command "sublocale l < e" replaces "interpretation l < e".  The
  3628 instantiation clause in "interpretation" and "interpret" (square
  3629 brackets) is no longer available.  Use locale expressions.
  3630 
  3631 - When converting proof scripts, mandatory qualifiers in
  3632 'interpretation' and 'interpret' should be retained by default, even
  3633 if this is an INCOMPATIBILITY compared to former behavior.  In the
  3634 worst case, use the "name?:" form for non-mandatory ones.  Qualifiers
  3635 in locale expressions range over a single locale instance only.
  3636 
  3637 - Dropped locale element "includes".  This is a major INCOMPATIBILITY.
  3638 In existing theorem specifications replace the includes element by the
  3639 respective context elements of the included locale, omitting those
  3640 that are already present in the theorem specification.  Multiple
  3641 assume elements of a locale should be replaced by a single one
  3642 involving the locale predicate.  In the proof body, declarations (most
  3643 notably theorems) may be regained by interpreting the respective
  3644 locales in the proof context as required (command "interpret").
  3645 
  3646 If using "includes" in replacement of a target solely because the
  3647 parameter types in the theorem are not as general as in the target,
  3648 consider declaring a new locale with additional type constraints on
  3649 the parameters (context element "constrains").
  3650 
  3651 - Discontinued "locale (open)".  INCOMPATIBILITY.
  3652 
  3653 - Locale interpretation commands no longer attempt to simplify goal.
  3654 INCOMPATIBILITY: in rare situations the generated goal differs.  Use
  3655 methods intro_locales and unfold_locales to clarify.
  3656 
  3657 - Locale interpretation commands no longer accept interpretation
  3658 attributes.  INCOMPATIBILITY.
  3659 
  3660 * Class declaration: so-called "base sort" must not be given in import
  3661 list any longer, but is inferred from the specification.  Particularly
  3662 in HOL, write
  3663 
  3664     class foo = ...
  3665 
  3666 instead of
  3667 
  3668     class foo = type + ...
  3669 
  3670 * Class target: global versions of theorems stemming do not carry a
  3671 parameter prefix any longer.  INCOMPATIBILITY.
  3672 
  3673 * Class 'instance' command no longer accepts attached definitions.
  3674 INCOMPATIBILITY, use proper 'instantiation' target instead.
  3675 
  3676 * Recovered hiding of consts, which was accidentally broken in
  3677 Isabelle2007.  Potential INCOMPATIBILITY, ``hide const c'' really
  3678 makes c inaccessible; consider using ``hide (open) const c'' instead.
  3679 
  3680 * Slightly more coherent Pure syntax, with updated documentation in
  3681 isar-ref manual.  Removed locales meta_term_syntax and
  3682 meta_conjunction_syntax: TERM and &&& (formerly &&) are now permanent,
  3683 INCOMPATIBILITY in rare situations.  Note that &&& should not be used
  3684 directly in regular applications.
  3685 
  3686 * There is a new syntactic category "float_const" for signed decimal
  3687 fractions (e.g. 123.45 or -123.45).
  3688 
  3689 * Removed exotic 'token_translation' command.  INCOMPATIBILITY, use ML
  3690 interface with 'setup' command instead.
  3691 
  3692 * Command 'local_setup' is similar to 'setup', but operates on a local
  3693 theory context.
  3694 
  3695 * The 'axiomatization' command now only works within a global theory
  3696 context.  INCOMPATIBILITY.
  3697 
  3698 * Goal-directed proof now enforces strict proof irrelevance wrt. sort
  3699 hypotheses.  Sorts required in the course of reasoning need to be
  3700 covered by the constraints in the initial statement, completed by the
  3701 type instance information of the background theory.  Non-trivial sort
  3702 hypotheses, which rarely occur in practice, may be specified via
  3703 vacuous propositions of the form SORT_CONSTRAINT('a::c).  For example:
  3704 
  3705   lemma assumes "SORT_CONSTRAINT('a::empty)" shows False ...
  3706 
  3707 The result contains an implicit sort hypotheses as before --
  3708 SORT_CONSTRAINT premises are eliminated as part of the canonical rule
  3709 normalization.
  3710 
  3711 * Generalized Isar history, with support for linear undo, direct state
  3712 addressing etc.
  3713 
  3714 * Changed defaults for unify configuration options:
  3715 
  3716   unify_trace_bound = 50 (formerly 25)
  3717   unify_search_bound = 60 (formerly 30)
  3718 
  3719 * Different bookkeeping for code equations (INCOMPATIBILITY):
  3720 
  3721   a) On theory merge, the last set of code equations for a particular
  3722      constant is taken (in accordance with the policy applied by other
  3723      parts of the code generator framework).
  3724 
  3725   b) Code equations stemming from explicit declarations (e.g. code
  3726      attribute) gain priority over default code equations stemming
  3727      from definition, primrec, fun etc.
  3728 
  3729 * Keyword 'code_exception' now named 'code_abort'.  INCOMPATIBILITY.
  3730 
  3731 * Unified theorem tables for both code generators.  Thus [code
  3732 func] has disappeared and only [code] remains.  INCOMPATIBILITY.
  3733 
  3734 * Command 'find_consts' searches for constants based on type and name
  3735 patterns, e.g.
  3736 
  3737     find_consts "_ => bool"
  3738 
  3739 By default, matching is against subtypes, but it may be restricted to
  3740 the whole type.  Searching by name is possible.  Multiple queries are
  3741 conjunctive and queries may be negated by prefixing them with a
  3742 hyphen:
  3743 
  3744     find_consts strict: "_ => bool" name: "Int" -"int => int"
  3745 
  3746 * New 'find_theorems' criterion "solves" matches theorems that
  3747 directly solve the current goal (modulo higher-order unification).
  3748 
  3749 * Auto solve feature for main theorem statements: whenever a new goal
  3750 is stated, "find_theorems solves" is called; any theorems that could
  3751 solve the lemma directly are listed as part of the goal state.
  3752 Cf. associated options in Proof General Isabelle settings menu,
  3753 enabled by default, with reasonable timeout for pathological cases of
  3754 higher-order unification.
  3755 
  3756 
  3757 *** Document preparation ***
  3758 
  3759 * Antiquotation @{lemma} now imitates a regular terminal proof,
  3760 demanding keyword 'by' and supporting the full method expression
  3761 syntax just like the Isar command 'by'.
  3762 
  3763 
  3764 *** HOL ***
  3765 
  3766 * Integrated main parts of former image HOL-Complex with HOL.  Entry
  3767 points Main and Complex_Main remain as before.
  3768 
  3769 * Logic image HOL-Plain provides a minimal HOL with the most important
  3770 tools available (inductive, datatype, primrec, ...).  This facilitates
  3771 experimentation and tool development.  Note that user applications
  3772 (and library theories) should never refer to anything below theory
  3773 Main, as before.
  3774 
  3775 * Logic image HOL-Main stops at theory Main, and thus facilitates
  3776 experimentation due to shorter build times.
  3777 
  3778 * Logic image HOL-NSA contains theories of nonstandard analysis which
  3779 were previously part of former HOL-Complex.  Entry point Hyperreal
  3780 remains valid, but theories formerly using Complex_Main should now use
  3781 new entry point Hypercomplex.
  3782 
  3783 * Generic ATP manager for Sledgehammer, based on ML threads instead of
  3784 Posix processes.  Avoids potentially expensive forking of the ML
  3785 process.  New thread-based implementation also works on non-Unix
  3786 platforms (Cygwin).  Provers are no longer hardwired, but defined
  3787 within the theory via plain ML wrapper functions.  Basic Sledgehammer
  3788 commands are covered in the isar-ref manual.
  3789 
  3790 * Wrapper scripts for remote SystemOnTPTP service allows to use
  3791 sledgehammer without local ATP installation (Vampire etc.). Other
  3792 provers may be included via suitable ML wrappers, see also
  3793 src/HOL/ATP_Linkup.thy.
  3794 
  3795 * ATP selection (E/Vampire/Spass) is now via Proof General's settings
  3796 menu.
  3797 
  3798 * The metis method no longer fails because the theorem is too trivial
  3799 (contains the empty clause).
  3800 
  3801 * The metis method now fails in the usual manner, rather than raising
  3802 an exception, if it determines that it cannot prove the theorem.
  3803 
  3804 * Method "coherent" implements a prover for coherent logic (see also
  3805 src/Tools/coherent.ML).
  3806 
  3807 * Constants "undefined" and "default" replace "arbitrary".  Usually
  3808 "undefined" is the right choice to replace "arbitrary", though
  3809 logically there is no difference.  INCOMPATIBILITY.
  3810 
  3811 * Command "value" now integrates different evaluation mechanisms.  The
  3812 result of the first successful evaluation mechanism is printed.  In
  3813 square brackets a particular named evaluation mechanisms may be
  3814 specified (currently, [SML], [code] or [nbe]).  See further
  3815 src/HOL/ex/Eval_Examples.thy.
  3816 
  3817 * Normalization by evaluation now allows non-leftlinear equations.
  3818 Declare with attribute [code nbe].
  3819 
  3820 * Methods "case_tac" and "induct_tac" now refer to the very same rules
  3821 as the structured Isar versions "cases" and "induct", cf. the
  3822 corresponding "cases" and "induct" attributes.  Mutual induction rules
  3823 are now presented as a list of individual projections
  3824 (e.g. foo_bar.inducts for types foo and bar); the old format with
  3825 explicit HOL conjunction is no longer supported.  INCOMPATIBILITY, in
  3826 rare situations a different rule is selected --- notably nested tuple
  3827 elimination instead of former prod.exhaust: use explicit (case_tac t
  3828 rule: prod.exhaust) here.
  3829 
  3830 * Attributes "cases", "induct", "coinduct" support "del" option.
  3831 
  3832 * Removed fact "case_split_thm", which duplicates "case_split".
  3833 
  3834 * The option datatype has been moved to a new theory Option.  Renamed
  3835 option_map to Option.map, and o2s to Option.set, INCOMPATIBILITY.
  3836 
  3837 * New predicate "strict_mono" classifies strict functions on partial
  3838 orders.  With strict functions on linear orders, reasoning about
  3839 (in)equalities is facilitated by theorems "strict_mono_eq",
  3840 "strict_mono_less_eq" and "strict_mono_less".
  3841 
  3842 * Some set operations are now proper qualified constants with
  3843 authentic syntax.  INCOMPATIBILITY:
  3844 
  3845     op Int ~>   Set.Int
  3846     op Un ~>    Set.Un
  3847     INTER ~>    Set.INTER
  3848     UNION ~>    Set.UNION
  3849     Inter ~>    Set.Inter
  3850     Union ~>    Set.Union
  3851     {} ~>       Set.empty
  3852     UNIV ~>     Set.UNIV
  3853 
  3854 * Class complete_lattice with operations Inf, Sup, INFI, SUPR now in
  3855 theory Set.
  3856 
  3857 * Auxiliary class "itself" has disappeared -- classes without any
  3858 parameter are treated as expected by the 'class' command.
  3859 
  3860 * Leibnitz's Series for Pi and the arcus tangens and logarithm series.
  3861 
  3862 * Common decision procedures (Cooper, MIR, Ferrack, Approximation,
  3863 Dense_Linear_Order) are now in directory HOL/Decision_Procs.
  3864 
  3865 * Theory src/HOL/Decision_Procs/Approximation provides the new proof
  3866 method "approximation".  It proves formulas on real values by using
  3867 interval arithmetic.  In the formulas are also the transcendental
  3868 functions sin, cos, tan, atan, ln, exp and the constant pi are
  3869 allowed. For examples see
  3870 src/HOL/Descision_Procs/ex/Approximation_Ex.thy.
  3871 
  3872 * Theory "Reflection" now resides in HOL/Library.
  3873 
  3874 * Entry point to Word library now simply named "Word".
  3875 INCOMPATIBILITY.
  3876 
  3877 * Made source layout more coherent with logical distribution
  3878 structure:
  3879 
  3880     src/HOL/Library/RType.thy ~> src/HOL/Typerep.thy
  3881     src/HOL/Library/Code_Message.thy ~> src/HOL/
  3882     src/HOL/Library/GCD.thy ~> src/HOL/
  3883     src/HOL/Library/Order_Relation.thy ~> src/HOL/
  3884     src/HOL/Library/Parity.thy ~> src/HOL/
  3885     src/HOL/Library/Univ_Poly.thy ~> src/HOL/
  3886     src/HOL/Real/ContNotDenum.thy ~> src/HOL/Library/
  3887     src/HOL/Real/Lubs.thy ~> src/HOL/
  3888     src/HOL/Real/PReal.thy ~> src/HOL/
  3889     src/HOL/Real/Rational.thy ~> src/HOL/
  3890     src/HOL/Real/RComplete.thy ~> src/HOL/
  3891     src/HOL/Real/RealDef.thy ~> src/HOL/
  3892     src/HOL/Real/RealPow.thy ~> src/HOL/
  3893     src/HOL/Real/Real.thy ~> src/HOL/
  3894     src/HOL/Complex/Complex_Main.thy ~> src/HOL/
  3895     src/HOL/Complex/Complex.thy ~> src/HOL/
  3896     src/HOL/Complex/FrechetDeriv.thy ~> src/HOL/Library/
  3897     src/HOL/Complex/Fundamental_Theorem_Algebra.thy ~> src/HOL/Library/
  3898     src/HOL/Hyperreal/Deriv.thy ~> src/HOL/
  3899     src/HOL/Hyperreal/Fact.thy ~> src/HOL/
  3900     src/HOL/Hyperreal/Integration.thy ~> src/HOL/
  3901     src/HOL/Hyperreal/Lim.thy ~> src/HOL/
  3902     src/HOL/Hyperreal/Ln.thy ~> src/HOL/
  3903     src/HOL/Hyperreal/Log.thy ~> src/HOL/
  3904     src/HOL/Hyperreal/MacLaurin.thy ~> src/HOL/
  3905     src/HOL/Hyperreal/NthRoot.thy ~> src/HOL/
  3906     src/HOL/Hyperreal/Series.thy ~> src/HOL/
  3907     src/HOL/Hyperreal/SEQ.thy ~> src/HOL/
  3908     src/HOL/Hyperreal/Taylor.thy ~> src/HOL/
  3909     src/HOL/Hyperreal/Transcendental.thy ~> src/HOL/
  3910     src/HOL/Real/Float ~> src/HOL/Library/
  3911     src/HOL/Real/HahnBanach ~> src/HOL/HahnBanach
  3912     src/HOL/Real/RealVector.thy ~> src/HOL/
  3913 
  3914     src/HOL/arith_data.ML ~> src/HOL/Tools
  3915     src/HOL/hologic.ML ~> src/HOL/Tools
  3916     src/HOL/simpdata.ML ~> src/HOL/Tools
  3917     src/HOL/int_arith1.ML ~> src/HOL/Tools/int_arith.ML
  3918     src/HOL/int_factor_simprocs.ML ~> src/HOL/Tools
  3919     src/HOL/nat_simprocs.ML ~> src/HOL/Tools
  3920     src/HOL/Real/float_arith.ML ~> src/HOL/Tools
  3921     src/HOL/Real/float_syntax.ML ~> src/HOL/Tools
  3922     src/HOL/Real/rat_arith.ML ~> src/HOL/Tools
  3923     src/HOL/Real/real_arith.ML ~> src/HOL/Tools
  3924 
  3925     src/HOL/Library/Array.thy ~> src/HOL/Imperative_HOL
  3926     src/HOL/Library/Heap_Monad.thy ~> src/HOL/Imperative_HOL
  3927     src/HOL/Library/Heap.thy ~> src/HOL/Imperative_HOL
  3928     src/HOL/Library/Imperative_HOL.thy ~> src/HOL/Imperative_HOL
  3929     src/HOL/Library/Ref.thy ~> src/HOL/Imperative_HOL
  3930     src/HOL/Library/Relational.thy ~> src/HOL/Imperative_HOL
  3931 
  3932 * If methods "eval" and "evaluation" encounter a structured proof
  3933 state with !!/==>, only the conclusion is evaluated to True (if
  3934 possible), avoiding strange error messages.
  3935 
  3936 * Method "sizechange" automates termination proofs using (a
  3937 modification of) the size-change principle.  Requires SAT solver.  See
  3938 src/HOL/ex/Termination.thy for examples.
  3939 
  3940 * Simplifier: simproc for let expressions now unfolds if bound
  3941 variable occurs at most once in let expression body.  INCOMPATIBILITY.
  3942 
  3943 * Method "arith": Linear arithmetic now ignores all inequalities when
  3944 fast_arith_neq_limit is exceeded, instead of giving up entirely.
  3945 
  3946 * New attribute "arith" for facts that should always be used
  3947 automatically by arithmetic. It is intended to be used locally in
  3948 proofs, e.g.
  3949 
  3950   assumes [arith]: "x > 0"
  3951 
  3952 Global usage is discouraged because of possible performance impact.
  3953 
  3954 * New classes "top" and "bot" with corresponding operations "top" and
  3955 "bot" in theory Orderings; instantiation of class "complete_lattice"
  3956 requires instantiation of classes "top" and "bot".  INCOMPATIBILITY.
  3957 
  3958 * Changed definition lemma "less_fun_def" in order to provide an
  3959 instance for preorders on functions; use lemma "less_le" instead.
  3960 INCOMPATIBILITY.
  3961 
  3962 * Theory Orderings: class "wellorder" moved here, with explicit
  3963 induction rule "less_induct" as assumption.  For instantiation of
  3964 "wellorder" by means of predicate "wf", use rule wf_wellorderI.
  3965 INCOMPATIBILITY.
  3966 
  3967 * Theory Orderings: added class "preorder" as superclass of "order".
  3968 INCOMPATIBILITY: Instantiation proofs for order, linorder
  3969 etc. slightly changed.  Some theorems named order_class.* now named
  3970 preorder_class.*.
  3971 
  3972 * Theory Relation: renamed "refl" to "refl_on", "reflexive" to "refl,
  3973 "diag" to "Id_on".
  3974 
  3975 * Theory Finite_Set: added a new fold combinator of type
  3976 
  3977   ('a => 'b => 'b) => 'b => 'a set => 'b
  3978 
  3979 Occasionally this is more convenient than the old fold combinator
  3980 which is now defined in terms of the new one and renamed to
  3981 fold_image.
  3982 
  3983 * Theories Ring_and_Field and OrderedGroup: The lemmas "group_simps"
  3984 and "ring_simps" have been replaced by "algebra_simps" (which can be
  3985 extended with further lemmas!).  At the moment both still exist but
  3986 the former will disappear at some point.
  3987 
  3988 * Theory Power: Lemma power_Suc is now declared as a simp rule in
  3989 class recpower.  Type-specific simp rules for various recpower types
  3990 have been removed.  INCOMPATIBILITY, rename old lemmas as follows:
  3991 
  3992 rat_power_0    -> power_0
  3993 rat_power_Suc  -> power_Suc
  3994 realpow_0      -> power_0
  3995 realpow_Suc    -> power_Suc
  3996 complexpow_0   -> power_0
  3997 complexpow_Suc -> power_Suc
  3998 power_poly_0   -> power_0
  3999 power_poly_Suc -> power_Suc
  4000 
  4001 * Theories Ring_and_Field and Divides: Definition of "op dvd" has been
  4002 moved to separate class dvd in Ring_and_Field; a couple of lemmas on
  4003 dvd has been generalized to class comm_semiring_1.  Likewise a bunch
  4004 of lemmas from Divides has been generalized from nat to class
  4005 semiring_div.  INCOMPATIBILITY.  This involves the following theorem
  4006 renames resulting from duplicate elimination:
  4007 
  4008     dvd_def_mod ~>          dvd_eq_mod_eq_0
  4009     zero_dvd_iff ~>         dvd_0_left_iff
  4010     dvd_0 ~>                dvd_0_right
  4011     DIVISION_BY_ZERO_DIV ~> div_by_0
  4012     DIVISION_BY_ZERO_MOD ~> mod_by_0
  4013     mult_div ~>             div_mult_self2_is_id
  4014     mult_mod ~>             mod_mult_self2_is_0
  4015 
  4016 * Theory IntDiv: removed many lemmas that are instances of class-based
  4017 generalizations (from Divides and Ring_and_Field).  INCOMPATIBILITY,
  4018 rename old lemmas as follows:
  4019 
  4020 dvd_diff               -> nat_dvd_diff
  4021 dvd_zminus_iff         -> dvd_minus_iff
  4022 mod_add1_eq            -> mod_add_eq
  4023 mod_mult1_eq           -> mod_mult_right_eq
  4024 mod_mult1_eq'          -> mod_mult_left_eq
  4025 mod_mult_distrib_mod   -> mod_mult_eq
  4026 nat_mod_add_left_eq    -> mod_add_left_eq
  4027 nat_mod_add_right_eq   -> mod_add_right_eq
  4028 nat_mod_div_trivial    -> mod_div_trivial
  4029 nat_mod_mod_trivial    -> mod_mod_trivial
  4030 zdiv_zadd_self1        -> div_add_self1
  4031 zdiv_zadd_self2        -> div_add_self2
  4032 zdiv_zmult_self1       -> div_mult_self2_is_id
  4033 zdiv_zmult_self2       -> div_mult_self1_is_id
  4034 zdvd_triv_left         -> dvd_triv_left
  4035 zdvd_triv_right        -> dvd_triv_right
  4036 zdvd_zmult_cancel_disj -> dvd_mult_cancel_left
  4037 zmod_eq0_zdvd_iff      -> dvd_eq_mod_eq_0[symmetric]
  4038 zmod_zadd_left_eq      -> mod_add_left_eq
  4039 zmod_zadd_right_eq     -> mod_add_right_eq
  4040 zmod_zadd_self1        -> mod_add_self1
  4041 zmod_zadd_self2        -> mod_add_self2
  4042 zmod_zadd1_eq          -> mod_add_eq
  4043 zmod_zdiff1_eq         -> mod_diff_eq
  4044 zmod_zdvd_zmod         -> mod_mod_cancel
  4045 zmod_zmod_cancel       -> mod_mod_cancel
  4046 zmod_zmult_self1       -> mod_mult_self2_is_0
  4047 zmod_zmult_self2       -> mod_mult_self1_is_0
  4048 zmod_1                 -> mod_by_1
  4049 zdiv_1                 -> div_by_1
  4050 zdvd_abs1              -> abs_dvd_iff
  4051 zdvd_abs2              -> dvd_abs_iff
  4052 zdvd_refl              -> dvd_refl
  4053 zdvd_trans             -> dvd_trans
  4054 zdvd_zadd              -> dvd_add
  4055 zdvd_zdiff             -> dvd_diff
  4056 zdvd_zminus_iff        -> dvd_minus_iff
  4057 zdvd_zminus2_iff       -> minus_dvd_iff
  4058 zdvd_zmultD            -> dvd_mult_right
  4059 zdvd_zmultD2           -> dvd_mult_left
  4060 zdvd_zmult_mono        -> mult_dvd_mono
  4061 zdvd_0_right           -> dvd_0_right
  4062 zdvd_0_left            -> dvd_0_left_iff
  4063 zdvd_1_left            -> one_dvd
  4064 zminus_dvd_iff         -> minus_dvd_iff
  4065 
  4066 * Theory Rational: 'Fract k 0' now equals '0'.  INCOMPATIBILITY.
  4067 
  4068 * The real numbers offer decimal input syntax: 12.34 is translated
  4069 into 1234/10^2. This translation is not reversed upon output.
  4070 
  4071 * Theory Library/Polynomial defines an abstract type 'a poly of
  4072 univariate polynomials with coefficients of type 'a.  In addition to
  4073 the standard ring operations, it also supports div and mod.  Code
  4074 generation is also supported, using list-style constructors.
  4075 
  4076 * Theory Library/Inner_Product defines a class of real_inner for real
  4077 inner product spaces, with an overloaded operation inner :: 'a => 'a
  4078 => real.  Class real_inner is a subclass of real_normed_vector from
  4079 theory RealVector.
  4080 
  4081 * Theory Library/Product_Vector provides instances for the product
  4082 type 'a * 'b of several classes from RealVector and Inner_Product.
  4083 Definitions of addition, subtraction, scalar multiplication, norms,
  4084 and inner products are included.
  4085 
  4086 * Theory Library/Bit defines the field "bit" of integers modulo 2.  In
  4087 addition to the field operations, numerals and case syntax are also
  4088 supported.
  4089 
  4090 * Theory Library/Diagonalize provides constructive version of Cantor's
  4091 first diagonalization argument.
  4092 
  4093 * Theory Library/GCD: Curried operations gcd, lcm (for nat) and zgcd,
  4094 zlcm (for int); carried together from various gcd/lcm developements in
  4095 the HOL Distribution.  Constants zgcd and zlcm replace former igcd and
  4096 ilcm; corresponding theorems renamed accordingly.  INCOMPATIBILITY,
  4097 may recover tupled syntax as follows:
  4098 
  4099     hide (open) const gcd
  4100     abbreviation gcd where
  4101       "gcd == (%(a, b). GCD.gcd a b)"
  4102     notation (output)
  4103       GCD.gcd ("gcd '(_, _')")
  4104 
  4105 The same works for lcm, zgcd, zlcm.
  4106 
  4107 * Theory Library/Nat_Infinity: added addition, numeral syntax and more
  4108 instantiations for algebraic structures.  Removed some duplicate
  4109 theorems.  Changes in simp rules.  INCOMPATIBILITY.
  4110 
  4111 * ML antiquotation @{code} takes a constant as argument and generates
  4112 corresponding code in background and inserts name of the corresponding
  4113 resulting ML value/function/datatype constructor binding in place.
  4114 All occurrences of @{code} with a single ML block are generated
  4115 simultaneously.  Provides a generic and safe interface for
  4116 instrumentalizing code generation.  See
  4117 src/HOL/Decision_Procs/Ferrack.thy for a more ambitious application.
  4118 In future you ought to refrain from ad-hoc compiling generated SML
  4119 code on the ML toplevel.  Note that (for technical reasons) @{code}
  4120 cannot refer to constants for which user-defined serializations are
  4121 set.  Refer to the corresponding ML counterpart directly in that
  4122 cases.
  4123 
  4124 * Command 'rep_datatype': instead of theorem names the command now
  4125 takes a list of terms denoting the constructors of the type to be
  4126 represented as datatype.  The characteristic theorems have to be
  4127 proven.  INCOMPATIBILITY.  Also observe that the following theorems
  4128 have disappeared in favour of existing ones:
  4129 
  4130     unit_induct                 ~> unit.induct
  4131     prod_induct                 ~> prod.induct
  4132     sum_induct                  ~> sum.induct
  4133     Suc_Suc_eq                  ~> nat.inject
  4134     Suc_not_Zero Zero_not_Suc   ~> nat.distinct
  4135 
  4136 
  4137 *** HOL-Algebra ***
  4138 
  4139 * New locales for orders and lattices where the equivalence relation
  4140 is not restricted to equality.  INCOMPATIBILITY: all order and lattice
  4141 locales use a record structure with field eq for the equivalence.
  4142 
  4143 * New theory of factorial domains.
  4144 
  4145 * Units_l_inv and Units_r_inv are now simp rules by default.
  4146 INCOMPATIBILITY.  Simplifier proof that require deletion of l_inv
  4147 and/or r_inv will now also require deletion of these lemmas.
  4148 
  4149 * Renamed the following theorems, INCOMPATIBILITY:
  4150 
  4151 UpperD ~> Upper_memD
  4152 LowerD ~> Lower_memD
  4153 least_carrier ~> least_closed
  4154 greatest_carrier ~> greatest_closed
  4155 greatest_Lower_above ~> greatest_Lower_below
  4156 one_zero ~> carrier_one_zero
  4157 one_not_zero ~> carrier_one_not_zero  (collision with assumption)
  4158 
  4159 
  4160 *** HOL-Nominal ***
  4161 
  4162 * Nominal datatypes can now contain type-variables.
  4163 
  4164 * Commands 'nominal_inductive' and 'equivariance' work with local
  4165 theory targets.
  4166 
  4167 * Nominal primrec can now works with local theory targets and its
  4168 specification syntax now conforms to the general format as seen in
  4169 'inductive' etc.
  4170 
  4171 * Method "perm_simp" honours the standard simplifier attributes
  4172 (no_asm), (no_asm_use) etc.
  4173 
  4174 * The new predicate #* is defined like freshness, except that on the
  4175 left hand side can be a set or list of atoms.
  4176 
  4177 * Experimental command 'nominal_inductive2' derives strong induction
  4178 principles for inductive definitions.  In contrast to
  4179 'nominal_inductive', which can only deal with a fixed number of
  4180 binders, it can deal with arbitrary expressions standing for sets of
  4181 atoms to be avoided.  The only inductive definition we have at the
  4182 moment that needs this generalisation is the typing rule for Lets in
  4183 the algorithm W:
  4184 
  4185  Gamma |- t1 : T1   (x,close Gamma T1)::Gamma |- t2 : T2   x#Gamma
  4186  -----------------------------------------------------------------
  4187          Gamma |- Let x be t1 in t2 : T2
  4188 
  4189 In this rule one wants to avoid all the binders that are introduced by
  4190 "close Gamma T1".  We are looking for other examples where this
  4191 feature might be useful.  Please let us know.
  4192 
  4193 
  4194 *** HOLCF ***
  4195 
  4196 * Reimplemented the simplification procedure for proving continuity
  4197 subgoals.  The new simproc is extensible; users can declare additional
  4198 continuity introduction rules with the attribute [cont2cont].
  4199 
  4200 * The continuity simproc now uses a different introduction rule for
  4201 solving continuity subgoals on terms with lambda abstractions.  In
  4202 some rare cases the new simproc may fail to solve subgoals that the
  4203 old one could solve, and "simp add: cont2cont_LAM" may be necessary.
  4204 Potential INCOMPATIBILITY.
  4205 
  4206 * Command 'fixrec': specification syntax now conforms to the general
  4207 format as seen in 'inductive' etc.  See src/HOLCF/ex/Fixrec_ex.thy for
  4208 examples.  INCOMPATIBILITY.
  4209 
  4210 
  4211 *** ZF ***
  4212 
  4213 * Proof of Zorn's Lemma for partial orders.
  4214 
  4215 
  4216 *** ML ***
  4217 
  4218 * Multithreading for Poly/ML 5.1/5.2 is no longer supported, only for
  4219 Poly/ML 5.2.1 or later.  Important note: the TimeLimit facility
  4220 depends on multithreading, so timouts will not work before Poly/ML
  4221 5.2.1!
  4222 
  4223 * High-level support for concurrent ML programming, see
  4224 src/Pure/Cuncurrent.  The data-oriented model of "future values" is
  4225 particularly convenient to organize independent functional
  4226 computations.  The concept of "synchronized variables" provides a
  4227 higher-order interface for components with shared state, avoiding the
  4228 delicate details of mutexes and condition variables.  (Requires
  4229 Poly/ML 5.2.1 or later.)
  4230 
  4231 * ML bindings produced via Isar commands are stored within the Isar
  4232 context (theory or proof).  Consequently, commands like 'use' and 'ML'
  4233 become thread-safe and work with undo as expected (concerning
  4234 top-level bindings, not side-effects on global references).
  4235 INCOMPATIBILITY, need to provide proper Isar context when invoking the
  4236 compiler at runtime; really global bindings need to be given outside a
  4237 theory.  (Requires Poly/ML 5.2 or later.)
  4238 
  4239 * Command 'ML_prf' is analogous to 'ML' but works within a proof
  4240 context.  Top-level ML bindings are stored within the proof context in
  4241 a purely sequential fashion, disregarding the nested proof structure.
  4242 ML bindings introduced by 'ML_prf' are discarded at the end of the
  4243 proof.  (Requires Poly/ML 5.2 or later.)
  4244 
  4245 * Simplified ML attribute and method setup, cf. functions Attrib.setup
  4246 and Method.setup, as well as Isar commands 'attribute_setup' and
  4247 'method_setup'.  INCOMPATIBILITY for 'method_setup', need to simplify
  4248 existing code accordingly, or use plain 'setup' together with old
  4249 Method.add_method.
  4250 
  4251 * Simplified ML oracle interface Thm.add_oracle promotes 'a -> cterm
  4252 to 'a -> thm, while results are always tagged with an authentic oracle
  4253 name.  The Isar command 'oracle' is now polymorphic, no argument type
  4254 is specified.  INCOMPATIBILITY, need to simplify existing oracle code
  4255 accordingly.  Note that extra performance may be gained by producing
  4256 the cterm carefully, avoiding slow Thm.cterm_of.
  4257 
  4258 * Simplified interface for defining document antiquotations via
  4259 ThyOutput.antiquotation, ThyOutput.output, and optionally
  4260 ThyOutput.maybe_pretty_source.  INCOMPATIBILITY, need to simplify user
  4261 antiquotations accordingly, see src/Pure/Thy/thy_output.ML for common
  4262 examples.
  4263 
  4264 * More systematic treatment of long names, abstract name bindings, and
  4265 name space operations.  Basic operations on qualified names have been
  4266 move from structure NameSpace to Long_Name, e.g. Long_Name.base_name,
  4267 Long_Name.append.  Old type bstring has been mostly replaced by
  4268 abstract type binding (see structure Binding), which supports precise
  4269 qualification by packages and local theory targets, as well as proper
  4270 tracking of source positions.  INCOMPATIBILITY, need to wrap old
  4271 bstring values into Binding.name, or better pass through abstract
  4272 bindings everywhere.  See further src/Pure/General/long_name.ML,
  4273 src/Pure/General/binding.ML and src/Pure/General/name_space.ML
  4274 
  4275 * Result facts (from PureThy.note_thms, ProofContext.note_thms,
  4276 LocalTheory.note etc.) now refer to the *full* internal name, not the
  4277 bstring as before.  INCOMPATIBILITY, not detected by ML type-checking!
  4278 
  4279 * Disposed old type and term read functions (Sign.read_def_typ,
  4280 Sign.read_typ, Sign.read_def_terms, Sign.read_term,
  4281 Thm.read_def_cterms, Thm.read_cterm etc.).  INCOMPATIBILITY, should
  4282 use regular Syntax.read_typ, Syntax.read_term, Syntax.read_typ_global,
  4283 Syntax.read_term_global etc.; see also OldGoals.read_term as last
  4284 resort for legacy applications.
  4285 
  4286 * Disposed old declarations, tactics, tactic combinators that refer to
  4287 the simpset or claset of an implicit theory (such as Addsimps,
  4288 Simp_tac, SIMPSET).  INCOMPATIBILITY, should use @{simpset} etc. in
  4289 embedded ML text, or local_simpset_of with a proper context passed as
  4290 explicit runtime argument.
  4291 
  4292 * Rules and tactics that read instantiations (read_instantiate,
  4293 res_inst_tac, thin_tac, subgoal_tac etc.) now demand a proper proof
  4294 context, which is required for parsing and type-checking.  Moreover,
  4295 the variables are specified as plain indexnames, not string encodings
  4296 thereof.  INCOMPATIBILITY.
  4297 
  4298 * Generic Toplevel.add_hook interface allows to analyze the result of
  4299 transactions.  E.g. see src/Pure/ProofGeneral/proof_general_pgip.ML
  4300 for theorem dependency output of transactions resulting in a new
  4301 theory state.
  4302 
  4303 * ML antiquotations: block-structured compilation context indicated by
  4304 \<lbrace> ... \<rbrace>; additional antiquotation forms:
  4305 
  4306   @{binding name}                         - basic name binding
  4307   @{let ?pat = term}                      - term abbreviation (HO matching)
  4308   @{note name = fact}                     - fact abbreviation
  4309   @{thm fact}                             - singleton fact (with attributes)
  4310   @{thms fact}                            - general fact (with attributes)
  4311   @{lemma prop by method}                 - singleton goal
  4312   @{lemma prop by meth1 meth2}            - singleton goal
  4313   @{lemma prop1 ... propN by method}      - general goal
  4314   @{lemma prop1 ... propN by meth1 meth2} - general goal
  4315   @{lemma (open) ...}                     - open derivation
  4316 
  4317 
  4318 *** System ***
  4319 
  4320 * The Isabelle "emacs" tool provides a specific interface to invoke
  4321 Proof General / Emacs, with more explicit failure if that is not
  4322 installed (the old isabelle-interface script silently falls back on
  4323 isabelle-process).  The PROOFGENERAL_HOME setting determines the
  4324 installation location of the Proof General distribution.
  4325 
  4326 * Isabelle/lib/classes/Pure.jar provides basic support to integrate
  4327 the Isabelle process into a JVM/Scala application.  See
  4328 Isabelle/lib/jedit/plugin for a minimal example.  (The obsolete Java
  4329 process wrapper has been discontinued.)
  4330 
  4331 * Added homegrown Isabelle font with unicode layout, see lib/fonts.
  4332 
  4333 * Various status messages (with exact source position information) are
  4334 emitted, if proper markup print mode is enabled.  This allows
  4335 user-interface components to provide detailed feedback on internal
  4336 prover operations.
  4337 
  4338 
  4339 
  4340 New in Isabelle2008 (June 2008)
  4341 -------------------------------
  4342 
  4343 *** General ***
  4344 
  4345 * The Isabelle/Isar Reference Manual (isar-ref) has been reorganized
  4346 and updated, with formally checked references as hyperlinks.
  4347 
  4348 * Theory loader: use_thy (and similar operations) no longer set the
  4349 implicit ML context, which was occasionally hard to predict and in
  4350 conflict with concurrency.  INCOMPATIBILITY, use ML within Isar which
  4351 provides a proper context already.
  4352 
  4353 * Theory loader: old-style ML proof scripts being *attached* to a thy
  4354 file are no longer supported.  INCOMPATIBILITY, regular 'uses' and
  4355 'use' within a theory file will do the job.
  4356 
  4357 * Name space merge now observes canonical order, i.e. the second space
  4358 is inserted into the first one, while existing entries in the first
  4359 space take precedence.  INCOMPATIBILITY in rare situations, may try to
  4360 swap theory imports.
  4361 
  4362 * Syntax: symbol \<chi> is now considered a letter.  Potential
  4363 INCOMPATIBILITY in identifier syntax etc.
  4364 
  4365 * Outer syntax: string tokens no longer admit escaped white space,
  4366 which was an accidental (undocumented) feature.  INCOMPATIBILITY, use
  4367 white space without escapes.
  4368 
  4369 * Outer syntax: string tokens may contain arbitrary character codes
  4370 specified via 3 decimal digits (as in SML).  E.g. "foo\095bar" for
  4371 "foo_bar".
  4372 
  4373 
  4374 *** Pure ***
  4375 
  4376 * Context-dependent token translations.  Default setup reverts locally
  4377 fixed variables, and adds hilite markup for undeclared frees.
  4378 
  4379 * Unused theorems can be found using the new command 'unused_thms'.
  4380 There are three ways of invoking it:
  4381 
  4382 (1) unused_thms
  4383      Only finds unused theorems in the current theory.
  4384 
  4385 (2) unused_thms thy_1 ... thy_n -
  4386      Finds unused theorems in the current theory and all of its ancestors,
  4387      excluding the theories thy_1 ... thy_n and all of their ancestors.
  4388 
  4389 (3) unused_thms thy_1 ... thy_n - thy'_1 ... thy'_m
  4390      Finds unused theorems in the theories thy'_1 ... thy'_m and all of
  4391      their ancestors, excluding the theories thy_1 ... thy_n and all of
  4392      their ancestors.
  4393 
  4394 In order to increase the readability of the list produced by
  4395 unused_thms, theorems that have been created by a particular instance
  4396 of a theory command such as 'inductive' or 'function' are considered
  4397 to belong to the same "group", meaning that if at least one theorem in
  4398 this group is used, the other theorems in the same group are no longer
  4399 reported as unused.  Moreover, if all theorems in the group are
  4400 unused, only one theorem in the group is displayed.
  4401 
  4402 Note that proof objects have to be switched on in order for
  4403 unused_thms to work properly (i.e. !proofs must be >= 1, which is
  4404 usually the case when using Proof General with the default settings).
  4405 
  4406 * Authentic naming of facts disallows ad-hoc overwriting of previous
  4407 theorems within the same name space.  INCOMPATIBILITY, need to remove
  4408 duplicate fact bindings, or even accidental fact duplications.  Note
  4409 that tools may maintain dynamically scoped facts systematically, using
  4410 PureThy.add_thms_dynamic.
  4411 
  4412 * Command 'hide' now allows to hide from "fact" name space as well.
  4413 
  4414 * Eliminated destructive theorem database, simpset, claset, and
  4415 clasimpset.  Potential INCOMPATIBILITY, really need to observe linear
  4416 update of theories within ML code.
  4417 
  4418 * Eliminated theory ProtoPure and CPure, leaving just one Pure theory.
  4419 INCOMPATIBILITY, object-logics depending on former Pure require
  4420 additional setup PureThy.old_appl_syntax_setup; object-logics
  4421 depending on former CPure need to refer to Pure.
  4422 
  4423 * Commands 'use' and 'ML' are now purely functional, operating on
  4424 theory/local_theory.  Removed former 'ML_setup' (on theory), use 'ML'
  4425 instead.  Added 'ML_val' as mere diagnostic replacement for 'ML'.
  4426 INCOMPATIBILITY.
  4427 
  4428 * Command 'setup': discontinued implicit version with ML reference.
  4429 
  4430 * Instantiation target allows for simultaneous specification of class
  4431 instance operations together with an instantiation proof.
  4432 Type-checking phase allows to refer to class operations uniformly.
  4433 See src/HOL/Complex/Complex.thy for an Isar example and
  4434 src/HOL/Library/Eval.thy for an ML example.
  4435 
  4436 * Indexing of literal facts: be more serious about including only
  4437 facts from the visible specification/proof context, but not the
  4438 background context (locale etc.).  Affects `prop` notation and method
  4439 "fact".  INCOMPATIBILITY: need to name facts explicitly in rare
  4440 situations.
  4441 
  4442 * Method "cases", "induct", "coinduct": removed obsolete/undocumented
  4443 "(open)" option, which used to expose internal bound variables to the
  4444 proof text.
  4445 
  4446 * Isar statements: removed obsolete case "rule_context".
  4447 INCOMPATIBILITY, better use explicit fixes/assumes.
  4448 
  4449 * Locale proofs: default proof step now includes 'unfold_locales';
  4450 hence 'proof' without argument may be used to unfold locale
  4451 predicates.
  4452 
  4453 
  4454 *** Document preparation ***
  4455 
  4456 * Simplified pdfsetup.sty: color/hyperref is used unconditionally for
  4457 both pdf and dvi (hyperlinks usually work in xdvi as well); removed
  4458 obsolete thumbpdf setup (contemporary PDF viewers do this on the
  4459 spot); renamed link color from "darkblue" to "linkcolor" (default
  4460 value unchanged, can be redefined via \definecolor); no longer sets
  4461 "a4paper" option (unnecessary or even intrusive).
  4462 
  4463 * Antiquotation @{lemma A method} proves proposition A by the given
  4464 method (either a method name or a method name plus (optional) method
  4465 arguments in parentheses) and prints A just like @{prop A}.
  4466 
  4467 
  4468 *** HOL ***
  4469 
  4470 * New primrec package.  Specification syntax conforms in style to
  4471 definition/function/....  No separate induction rule is provided.  The
  4472 "primrec" command distinguishes old-style and new-style specifications
  4473 by syntax.  The former primrec package is now named OldPrimrecPackage.
  4474 When adjusting theories, beware: constants stemming from new-style
  4475 primrec specifications have authentic syntax.
  4476 
  4477 * Metis prover is now an order of magnitude faster, and also works
  4478 with multithreading.
  4479 
  4480 * Metis: the maximum number of clauses that can be produced from a
  4481 theorem is now given by the attribute max_clauses.  Theorems that
  4482 exceed this number are ignored, with a warning printed.
  4483 
  4484 * Sledgehammer no longer produces structured proofs by default. To
  4485 enable, declare [[sledgehammer_full = true]].  Attributes
  4486 reconstruction_modulus, reconstruction_sorts renamed
  4487 sledgehammer_modulus, sledgehammer_sorts.  INCOMPATIBILITY.
  4488 
  4489 * Method "induct_scheme" derives user-specified induction rules
  4490 from well-founded induction and completeness of patterns. This factors
  4491 out some operations that are done internally by the function package
  4492 and makes them available separately.  See
  4493 src/HOL/ex/Induction_Scheme.thy for examples.
  4494 
  4495 * More flexible generation of measure functions for termination
  4496 proofs: Measure functions can be declared by proving a rule of the
  4497 form "is_measure f" and giving it the [measure_function] attribute.
  4498 The "is_measure" predicate is logically meaningless (always true), and
  4499 just guides the heuristic.  To find suitable measure functions, the
  4500 termination prover sets up the goal "is_measure ?f" of the appropriate
  4501 type and generates all solutions by prolog-style backwards proof using
  4502 the declared rules.
  4503 
  4504 This setup also deals with rules like 
  4505 
  4506   "is_measure f ==> is_measure (list_size f)"
  4507 
  4508 which accommodates nested datatypes that recurse through lists.
  4509 Similar rules are predeclared for products and option types.
  4510 
  4511 * Turned the type of sets "'a set" into an abbreviation for "'a => bool"
  4512 
  4513   INCOMPATIBILITIES:
  4514 
  4515   - Definitions of overloaded constants on sets have to be replaced by
  4516     definitions on => and bool.
  4517 
  4518   - Some definitions of overloaded operators on sets can now be proved
  4519     using the definitions of the operators on => and bool.  Therefore,
  4520     the following theorems have been renamed:
  4521 
  4522       subset_def   -> subset_eq
  4523       psubset_def  -> psubset_eq
  4524       set_diff_def -> set_diff_eq
  4525       Compl_def    -> Compl_eq
  4526       Sup_set_def  -> Sup_set_eq
  4527       Inf_set_def  -> Inf_set_eq
  4528       sup_set_def  -> sup_set_eq
  4529       inf_set_def  -> inf_set_eq
  4530 
  4531   - Due to the incompleteness of the HO unification algorithm, some
  4532     rules such as subst may require manual instantiation, if some of
  4533     the unknowns in the rule is a set.
  4534 
  4535   - Higher order unification and forward proofs:
  4536     The proof pattern
  4537 
  4538       have "P (S::'a set)" <...>
  4539       then have "EX S. P S" ..
  4540 
  4541     no longer works (due to the incompleteness of the HO unification
  4542     algorithm) and must be replaced by the pattern
  4543 
  4544       have "EX S. P S"
  4545       proof
  4546         show "P S" <...>
  4547       qed
  4548 
  4549   - Calculational reasoning with subst (or similar rules):
  4550     The proof pattern
  4551 
  4552       have "P (S::'a set)" <...>
  4553       also have "S = T" <...>
  4554       finally have "P T" .
  4555 
  4556     no longer works (for similar reasons as the previous example) and
  4557     must be replaced by something like
  4558 
  4559       have "P (S::'a set)" <...>
  4560       moreover have "S = T" <...>
  4561       ultimately have "P T" by simp
  4562 
  4563   - Tactics or packages written in ML code:
  4564     Code performing pattern matching on types via
  4565 
  4566       Type ("set", [T]) => ...
  4567 
  4568     must be rewritten. Moreover, functions like strip_type or
  4569     binder_types no longer return the right value when applied to a
  4570     type of the form
  4571 
  4572       T1 => ... => Tn => U => bool
  4573 
  4574     rather than
  4575 
  4576       T1 => ... => Tn => U set
  4577 
  4578 * Merged theories Wellfounded_Recursion, Accessible_Part and
  4579 Wellfounded_Relations to theory Wellfounded.
  4580 
  4581 * Explicit class "eq" for executable equality.  INCOMPATIBILITY.
  4582 
  4583 * Class finite no longer treats UNIV as class parameter.  Use class
  4584 enum from theory Library/Enum instead to achieve a similar effect.
  4585 INCOMPATIBILITY.
  4586 
  4587 * Theory List: rule list_induct2 now has explicitly named cases "Nil"
  4588 and "Cons".  INCOMPATIBILITY.
  4589 
  4590 * HOL (and FOL): renamed variables in rules imp_elim and swap.
  4591 Potential INCOMPATIBILITY.
  4592 
  4593 * Theory Product_Type: duplicated lemmas split_Pair_apply and
  4594 injective_fst_snd removed, use split_eta and prod_eqI instead.
  4595 Renamed upd_fst to apfst and upd_snd to apsnd.  INCOMPATIBILITY.
  4596 
  4597 * Theory Nat: removed redundant lemmas that merely duplicate lemmas of
  4598 the same name in theory Orderings:
  4599 
  4600   less_trans
  4601   less_linear
  4602   le_imp_less_or_eq
  4603   le_less_trans
  4604   less_le_trans
  4605   less_not_sym
  4606   less_asym
  4607 
  4608 Renamed less_imp_le to less_imp_le_nat, and less_irrefl to
  4609 less_irrefl_nat.  Potential INCOMPATIBILITY due to more general types
  4610 and different variable names.
  4611 
  4612 * Library/Option_ord.thy: Canonical order on option type.
  4613 
  4614 * Library/RBT.thy: Red-black trees, an efficient implementation of
  4615 finite maps.
  4616 
  4617 * Library/Countable.thy: Type class for countable types.
  4618 
  4619 * Theory Int: The representation of numerals has changed.  The infix
  4620 operator BIT and the bit datatype with constructors B0 and B1 have
  4621 disappeared.  INCOMPATIBILITY, use "Int.Bit0 x" and "Int.Bit1 y" in
  4622 place of "x BIT bit.B0" and "y BIT bit.B1", respectively.  Theorems
  4623 involving BIT, B0, or B1 have been renamed with "Bit0" or "Bit1"
  4624 accordingly.
  4625 
  4626 * Theory Nat: definition of <= and < on natural numbers no longer
  4627 depend on well-founded relations.  INCOMPATIBILITY.  Definitions
  4628 le_def and less_def have disappeared.  Consider lemmas not_less
  4629 [symmetric, where ?'a = nat] and less_eq [symmetric] instead.
  4630 
  4631 * Theory Finite_Set: locales ACf, ACe, ACIf, ACIfSL and ACIfSLlin
  4632 (whose purpose mainly is for various fold_set functionals) have been
  4633 abandoned in favor of the existing algebraic classes
  4634 ab_semigroup_mult, comm_monoid_mult, ab_semigroup_idem_mult,
  4635 lower_semilattice (resp. upper_semilattice) and linorder.
  4636 INCOMPATIBILITY.
  4637 
  4638 * Theory Transitive_Closure: induct and cases rules now declare proper
  4639 case_names ("base" and "step").  INCOMPATIBILITY.
  4640 
  4641 * Theorem Inductive.lfp_ordinal_induct generalized to complete
  4642 lattices.  The form set-specific version is available as
  4643 Inductive.lfp_ordinal_induct_set.
  4644 
  4645 * Renamed theorems "power.simps" to "power_int.simps".
  4646 INCOMPATIBILITY.
  4647 
  4648 * Class semiring_div provides basic abstract properties of semirings
  4649 with division and modulo operations.  Subsumes former class dvd_mod.
  4650 
  4651 * Merged theories IntDef, Numeral and IntArith into unified theory
  4652 Int.  INCOMPATIBILITY.
  4653 
  4654 * Theory Library/Code_Index: type "index" now represents natural
  4655 numbers rather than integers.  INCOMPATIBILITY.
  4656 
  4657 * New class "uminus" with operation "uminus" (split of from class
  4658 "minus" which now only has operation "minus", binary).
  4659 INCOMPATIBILITY.
  4660 
  4661 * Constants "card", "internal_split", "option_map" now with authentic
  4662 syntax.  INCOMPATIBILITY.
  4663 
  4664 * Definitions subset_def, psubset_def, set_diff_def, Compl_def,
  4665 le_bool_def, less_bool_def, le_fun_def, less_fun_def, inf_bool_def,
  4666 sup_bool_def, Inf_bool_def, Sup_bool_def, inf_fun_def, sup_fun_def,
  4667 Inf_fun_def, Sup_fun_def, inf_set_def, sup_set_def, Inf_set_def,
  4668 Sup_set_def, le_def, less_def, option_map_def now with object
  4669 equality.  INCOMPATIBILITY.
  4670 
  4671 * Records. Removed K_record, and replaced it by pure lambda term
  4672 %x. c. The simplifier setup is now more robust against eta expansion.
  4673 INCOMPATIBILITY: in cases explicitly referring to K_record.
  4674 
  4675 * Library/Multiset: {#a, b, c#} abbreviates {#a#} + {#b#} + {#c#}.
  4676 
  4677 * Library/ListVector: new theory of arithmetic vector operations.
  4678 
  4679 * Library/Order_Relation: new theory of various orderings as sets of
  4680 pairs.  Defines preorders, partial orders, linear orders and
  4681 well-orders on sets and on types.
  4682 
  4683 
  4684 *** ZF ***
  4685 
  4686 * Renamed some theories to allow to loading both ZF and HOL in the
  4687 same session:
  4688 
  4689   Datatype  -> Datatype_ZF
  4690   Inductive -> Inductive_ZF
  4691   Int       -> Int_ZF
  4692   IntDiv    -> IntDiv_ZF
  4693   Nat       -> Nat_ZF
  4694   List      -> List_ZF
  4695   Main      -> Main_ZF
  4696 
  4697 INCOMPATIBILITY: ZF theories that import individual theories below
  4698 Main might need to be adapted.  Regular theory Main is still
  4699 available, as trivial extension of Main_ZF.
  4700 
  4701 
  4702 *** ML ***
  4703 
  4704 * ML within Isar: antiquotation @{const name} or @{const
  4705 name(typargs)} produces statically-checked Const term.
  4706 
  4707 * Functor NamedThmsFun: data is available to the user as dynamic fact
  4708 (of the same name).  Removed obsolete print command.
  4709 
  4710 * Removed obsolete "use_legacy_bindings" function.
  4711 
  4712 * The ``print mode'' is now a thread-local value derived from a global
  4713 template (the former print_mode reference), thus access becomes
  4714 non-critical.  The global print_mode reference is for session
  4715 management only; user-code should use print_mode_value,
  4716 print_mode_active, PrintMode.setmp etc.  INCOMPATIBILITY.
  4717 
  4718 * Functions system/system_out provide a robust way to invoke external
  4719 shell commands, with propagation of interrupts (requires Poly/ML
  4720 5.2.1).  Do not use OS.Process.system etc. from the basis library!
  4721 
  4722 
  4723 *** System ***
  4724 
  4725 * Default settings: PROOFGENERAL_OPTIONS no longer impose xemacs ---
  4726 in accordance with Proof General 3.7, which prefers GNU emacs.
  4727 
  4728 * isatool tty runs Isabelle process with plain tty interaction;
  4729 optional line editor may be specified via ISABELLE_LINE_EDITOR
  4730 setting, the default settings attempt to locate "ledit" and "rlwrap".
  4731 
  4732 * isatool browser now works with Cygwin as well, using general
  4733 "javapath" function defined in Isabelle process environment.
  4734 
  4735 * YXML notation provides a simple and efficient alternative to
  4736 standard XML transfer syntax.  See src/Pure/General/yxml.ML and
  4737 isatool yxml as described in the Isabelle system manual.
  4738 
  4739 * JVM class isabelle.IsabelleProcess (located in Isabelle/lib/classes)
  4740 provides general wrapper for managing an Isabelle process in a robust
  4741 fashion, with ``cooked'' output from stdin/stderr.
  4742 
  4743 * Rudimentary Isabelle plugin for jEdit (see Isabelle/lib/jedit),
  4744 based on Isabelle/JVM process wrapper (see Isabelle/lib/classes).
  4745 
  4746 * Removed obsolete THIS_IS_ISABELLE_BUILD feature.  NB: the documented
  4747 way of changing the user's settings is via
  4748 ISABELLE_HOME_USER/etc/settings, which is a fully featured bash
  4749 script.
  4750 
  4751 * Multithreading.max_threads := 0 refers to the number of actual CPU
  4752 cores of the underlying machine, which is a good starting point for
  4753 optimal performance tuning.  The corresponding usedir option -M allows
  4754 "max" as an alias for "0".  WARNING: does not work on certain versions
  4755 of Mac OS (with Poly/ML 5.1).
  4756 
  4757 * isabelle-process: non-ML sessions are run with "nice", to reduce the
  4758 adverse effect of Isabelle flooding interactive front-ends (notably
  4759 ProofGeneral / XEmacs).
  4760 
  4761 
  4762 
  4763 New in Isabelle2007 (November 2007)
  4764 -----------------------------------
  4765 
  4766 *** General ***
  4767 
  4768 * More uniform information about legacy features, notably a
  4769 warning/error of "Legacy feature: ...", depending on the state of the
  4770 tolerate_legacy_features flag (default true). FUTURE INCOMPATIBILITY:
  4771 legacy features will disappear eventually.
  4772 
  4773 * Theory syntax: the header format ``theory A = B + C:'' has been
  4774 discontinued in favour of ``theory A imports B C begin''.  Use isatool
  4775 fixheaders to convert existing theory files.  INCOMPATIBILITY.
  4776 
  4777 * Theory syntax: the old non-Isar theory file format has been
  4778 discontinued altogether.  Note that ML proof scripts may still be used
  4779 with Isar theories; migration is usually quite simple with the ML
  4780 function use_legacy_bindings.  INCOMPATIBILITY.
  4781 
  4782 * Theory syntax: some popular names (e.g. 'class', 'declaration',
  4783 'fun', 'help', 'if') are now keywords.  INCOMPATIBILITY, use double
  4784 quotes.
  4785 
  4786 * Theory loader: be more serious about observing the static theory
  4787 header specifications (including optional directories), but not the
  4788 accidental file locations of previously successful loads.  The strict
  4789 update policy of former update_thy is now already performed by
  4790 use_thy, so the former has been removed; use_thys updates several
  4791 theories simultaneously, just as 'imports' within a theory header
  4792 specification, but without merging the results.  Potential
  4793 INCOMPATIBILITY: may need to refine theory headers and commands
  4794 ROOT.ML which depend on load order.
  4795 
  4796 * Theory loader: optional support for content-based file
  4797 identification, instead of the traditional scheme of full physical
  4798 path plus date stamp; configured by the ISABELLE_FILE_IDENT setting
  4799 (cf. the system manual).  The new scheme allows to work with
  4800 non-finished theories in persistent session images, such that source
  4801 files may be moved later on without requiring reloads.
  4802 
  4803 * Theory loader: old-style ML proof scripts being *attached* to a thy
  4804 file (with the same base name as the theory) are considered a legacy
  4805 feature, which will disappear eventually. Even now, the theory loader
  4806 no longer maintains dependencies on such files.
  4807 
  4808 * Syntax: the scope for resolving ambiguities via type-inference is
  4809 now limited to individual terms, instead of whole simultaneous
  4810 specifications as before. This greatly reduces the complexity of the
  4811 syntax module and improves flexibility by separating parsing and
  4812 type-checking. INCOMPATIBILITY: additional type-constraints (explicit
  4813 'fixes' etc.) are required in rare situations.
  4814 
  4815 * Syntax: constants introduced by new-style packages ('definition',
  4816 'abbreviation' etc.) are passed through the syntax module in
  4817 ``authentic mode''. This means that associated mixfix annotations
  4818 really stick to such constants, independently of potential name space
  4819 ambiguities introduced later on. INCOMPATIBILITY: constants in parse
  4820 trees are represented slightly differently, may need to adapt syntax
  4821 translations accordingly. Use CONST marker in 'translations' and
  4822 @{const_syntax} antiquotation in 'parse_translation' etc.
  4823 
  4824 * Legacy goal package: reduced interface to the bare minimum required
  4825 to keep existing proof scripts running.  Most other user-level
  4826 functions are now part of the OldGoals structure, which is *not* open
  4827 by default (consider isatool expandshort before open OldGoals).
  4828 Removed top_sg, prin, printyp, pprint_term/typ altogether, because
  4829 these tend to cause confusion about the actual goal (!) context being
  4830 used here, which is not necessarily the same as the_context().
  4831 
  4832 * Command 'find_theorems': supports "*" wild-card in "name:"
  4833 criterion; "with_dups" option.  Certain ProofGeneral versions might
  4834 support a specific search form (see ProofGeneral/CHANGES).
  4835 
  4836 * The ``prems limit'' option (cf. ProofContext.prems_limit) is now -1
  4837 by default, which means that "prems" (and also "fixed variables") are
  4838 suppressed from proof state output.  Note that the ProofGeneral
  4839 settings mechanism allows to change and save options persistently, but
  4840 older versions of Isabelle will fail to start up if a negative prems
  4841 limit is imposed.
  4842 
  4843 * Local theory targets may be specified by non-nested blocks of
  4844 ``context/locale/class ... begin'' followed by ``end''.  The body may
  4845 contain definitions, theorems etc., including any derived mechanism
  4846 that has been implemented on top of these primitives.  This concept
  4847 generalizes the existing ``theorem (in ...)'' towards more versatility
  4848 and scalability.
  4849 
  4850 * Proof General interface: proper undo of final 'end' command;
  4851 discontinued Isabelle/classic mode (ML proof scripts).
  4852 
  4853 
  4854 *** Document preparation ***
  4855 
  4856 * Added antiquotation @{theory name} which prints the given name,
  4857 after checking that it refers to a valid ancestor theory in the
  4858 current context.
  4859 
  4860 * Added antiquotations @{ML_type text} and @{ML_struct text} which
  4861 check the given source text as ML type/structure, printing verbatim.
  4862 
  4863 * Added antiquotation @{abbrev "c args"} which prints the abbreviation
  4864 "c args == rhs" given in the current context.  (Any number of
  4865 arguments may be given on the LHS.)
  4866 
  4867 
  4868 *** Pure ***
  4869 
  4870 * The 'class' package offers a combination of axclass and locale to
  4871 achieve Haskell-like type classes in Isabelle.  Definitions and
  4872 theorems within a class context produce both relative results (with
  4873 implicit parameters according to the locale context), and polymorphic
  4874 constants with qualified polymorphism (according to the class
  4875 context).  Within the body context of a 'class' target, a separate
  4876 syntax layer ("user space type system") takes care of converting
  4877 between global polymorphic consts and internal locale representation.
  4878 See src/HOL/ex/Classpackage.thy for examples (as well as main HOL).
  4879 "isatool doc classes" provides a tutorial.
  4880 
  4881 * Generic code generator framework allows to generate executable
  4882 code for ML and Haskell (including Isabelle classes).  A short usage
  4883 sketch:
  4884 
  4885     internal compilation:
  4886         export_code <list of constants (term syntax)> in SML
  4887     writing SML code to a file:
  4888         export_code <list of constants (term syntax)> in SML <filename>
  4889     writing OCaml code to a file:
  4890         export_code <list of constants (term syntax)> in OCaml <filename>
  4891     writing Haskell code to a bunch of files:
  4892         export_code <list of constants (term syntax)> in Haskell <filename>
  4893 
  4894     evaluating closed propositions to True/False using code generation:
  4895         method ``eval''
  4896 
  4897 Reasonable default setup of framework in HOL.
  4898 
  4899 Theorem attributs for selecting and transforming function equations theorems:
  4900 
  4901     [code fun]:        select a theorem as function equation for a specific constant
  4902     [code fun del]:    deselect a theorem as function equation for a specific constant
  4903     [code inline]:     select an equation theorem for unfolding (inlining) in place
  4904     [code inline del]: deselect an equation theorem for unfolding (inlining) in place
  4905 
  4906 User-defined serializations (target in {SML, OCaml, Haskell}):
  4907 
  4908     code_const <and-list of constants (term syntax)>
  4909       {(target) <and-list of const target syntax>}+
  4910 
  4911     code_type <and-list of type constructors>
  4912       {(target) <and-list of type target syntax>}+
  4913 
  4914     code_instance <and-list of instances>
  4915       {(target)}+
  4916         where instance ::= <type constructor> :: <class>
  4917 
  4918     code_class <and_list of classes>
  4919       {(target) <and-list of class target syntax>}+
  4920         where class target syntax ::= <class name> {where {<classop> == <target syntax>}+}?
  4921 
  4922 code_instance and code_class only are effective to target Haskell.
  4923 
  4924 For example usage see src/HOL/ex/Codegenerator.thy and
  4925 src/HOL/ex/Codegenerator_Pretty.thy.  A separate tutorial on code
  4926 generation from Isabelle/HOL theories is available via "isatool doc
  4927 codegen".
  4928 
  4929 * Code generator: consts in 'consts_code' Isar commands are now
  4930 referred to by usual term syntax (including optional type
  4931 annotations).
  4932 
  4933 * Command 'no_translations' removes translation rules from theory
  4934 syntax.
  4935 
  4936 * Overloaded definitions are now actually checked for acyclic
  4937 dependencies.  The overloading scheme is slightly more general than
  4938 that of Haskell98, although Isabelle does not demand an exact
  4939 correspondence to type class and instance declarations.
  4940 INCOMPATIBILITY, use ``defs (unchecked overloaded)'' to admit more
  4941 exotic versions of overloading -- at the discretion of the user!
  4942 
  4943 Polymorphic constants are represented via type arguments, i.e. the
  4944 instantiation that matches an instance against the most general
  4945 declaration given in the signature.  For example, with the declaration
  4946 c :: 'a => 'a => 'a, an instance c :: nat => nat => nat is represented
  4947 as c(nat).  Overloading is essentially simultaneous structural
  4948 recursion over such type arguments.  Incomplete specification patterns
  4949 impose global constraints on all occurrences, e.g. c('a * 'a) on the
  4950 LHS means that more general c('a * 'b) will be disallowed on any RHS.
  4951 Command 'print_theory' outputs the normalized system of recursive
  4952 equations, see section "definitions".
  4953 
  4954 * Configuration options are maintained within the theory or proof
  4955 context (with name and type bool/int/string), providing a very simple
  4956 interface to a poor-man's version of general context data.  Tools may
  4957 declare options in ML (e.g. using Attrib.config_int) and then refer to
  4958 these values using Config.get etc.  Users may change options via an
  4959 associated attribute of the same name.  This form of context
  4960 declaration works particularly well with commands 'declare' or
  4961 'using', for example ``declare [[foo = 42]]''.  Thus it has become
  4962 very easy to avoid global references, which would not observe Isar
  4963 toplevel undo/redo and fail to work with multithreading.
  4964 
  4965 Various global ML references of Pure and HOL have been turned into
  4966 configuration options:
  4967 
  4968   Unify.search_bound		unify_search_bound
  4969   Unify.trace_bound		unify_trace_bound
  4970   Unify.trace_simp		unify_trace_simp
  4971   Unify.trace_types		unify_trace_types
  4972   Simplifier.simp_depth_limit	simp_depth_limit
  4973   Blast.depth_limit		blast_depth_limit
  4974   DatatypeProp.dtK		datatype_distinctness_limit
  4975   fast_arith_neq_limit  	fast_arith_neq_limit
  4976   fast_arith_split_limit	fast_arith_split_limit
  4977 
  4978 * Named collections of theorems may be easily installed as context
  4979 data using the functor NamedThmsFun (see also
  4980 src/Pure/Tools/named_thms.ML).  The user may add or delete facts via
  4981 attributes; there is also a toplevel print command.  This facility is
  4982 just a common case of general context data, which is the preferred way
  4983 for anything more complex than just a list of facts in canonical
  4984 order.
  4985 
  4986 * Isar: command 'declaration' augments a local theory by generic
  4987 declaration functions written in ML.  This enables arbitrary content
  4988 being added to the context, depending on a morphism that tells the
  4989 difference of the original declaration context wrt. the application
  4990 context encountered later on.
  4991 
  4992 * Isar: proper interfaces for simplification procedures.  Command
  4993 'simproc_setup' declares named simprocs (with match patterns, and body
  4994 text in ML).  Attribute "simproc" adds/deletes simprocs in the current
  4995 context.  ML antiquotation @{simproc name} retrieves named simprocs.
  4996 
  4997 * Isar: an extra pair of brackets around attribute declarations
  4998 abbreviates a theorem reference involving an internal dummy fact,
  4999 which will be ignored later --- only the effect of the attribute on
  5000 the background context will persist.  This form of in-place
  5001 declarations is particularly useful with commands like 'declare' and
  5002 'using', for example ``have A using [[simproc a]] by simp''.
  5003 
  5004 * Isar: method "assumption" (and implicit closing of subproofs) now
  5005 takes simple non-atomic goal assumptions into account: after applying
  5006 an assumption as a rule the resulting subgoals are solved by atomic
  5007 assumption steps.  This is particularly useful to finish 'obtain'
  5008 goals, such as "!!x. (!!x. P x ==> thesis) ==> P x ==> thesis",
  5009 without referring to the original premise "!!x. P x ==> thesis" in the
  5010 Isar proof context.  POTENTIAL INCOMPATIBILITY: method "assumption" is
  5011 more permissive.
  5012 
  5013 * Isar: implicit use of prems from the Isar proof context is
  5014 considered a legacy feature.  Common applications like ``have A .''
  5015 may be replaced by ``have A by fact'' or ``note `A`''.  In general,
  5016 referencing facts explicitly here improves readability and
  5017 maintainability of proof texts.
  5018 
  5019 * Isar: improper proof element 'guess' is like 'obtain', but derives
  5020 the obtained context from the course of reasoning!  For example:
  5021 
  5022   assume "EX x y. A x & B y"   -- "any previous fact"
  5023   then guess x and y by clarify
  5024 
  5025 This technique is potentially adventurous, depending on the facts and
  5026 proof tools being involved here.
  5027 
  5028 * Isar: known facts from the proof context may be specified as literal
  5029 propositions, using ASCII back-quote syntax.  This works wherever
  5030 named facts used to be allowed so far, in proof commands, proof
  5031 methods, attributes etc.  Literal facts are retrieved from the context
  5032 according to unification of type and term parameters.  For example,
  5033 provided that "A" and "A ==> B" and "!!x. P x ==> Q x" are known
  5034 theorems in the current context, then these are valid literal facts:
  5035 `A` and `A ==> B` and `!!x. P x ==> Q x" as well as `P a ==> Q a` etc.
  5036 
  5037 There is also a proof method "fact" which does the same composition
  5038 for explicit goal states, e.g. the following proof texts coincide with
  5039 certain special cases of literal facts:
  5040 
  5041   have "A" by fact                 ==  note `A`
  5042   have "A ==> B" by fact           ==  note `A ==> B`
  5043   have "!!x. P x ==> Q x" by fact  ==  note `!!x. P x ==> Q x`
  5044   have "P a ==> Q a" by fact       ==  note `P a ==> Q a`
  5045 
  5046 * Isar: ":" (colon) is no longer a symbolic identifier character in
  5047 outer syntax.  Thus symbolic identifiers may be used without
  5048 additional white space in declarations like this: ``assume *: A''.
  5049 
  5050 * Isar: 'print_facts' prints all local facts of the current context,
  5051 both named and unnamed ones.
  5052 
  5053 * Isar: 'def' now admits simultaneous definitions, e.g.:
  5054 
  5055   def x == "t" and y == "u"
  5056 
  5057 * Isar: added command 'unfolding', which is structurally similar to
  5058 'using', but affects both the goal state and facts by unfolding given
  5059 rewrite rules.  Thus many occurrences of the 'unfold' method or
  5060 'unfolded' attribute may be replaced by first-class proof text.
  5061 
  5062 * Isar: methods 'unfold' / 'fold', attributes 'unfolded' / 'folded',
  5063 and command 'unfolding' now all support object-level equalities
  5064 (potentially conditional).  The underlying notion of rewrite rule is
  5065 analogous to the 'rule_format' attribute, but *not* that of the
  5066 Simplifier (which is usually more generous).
  5067 
  5068 * Isar: the new attribute [rotated n] (default n = 1) rotates the
  5069 premises of a theorem by n. Useful in conjunction with drule.
  5070 
  5071 * Isar: the goal restriction operator [N] (default N = 1) evaluates a
  5072 method expression within a sandbox consisting of the first N
  5073 sub-goals, which need to exist.  For example, ``simp_all [3]''
  5074 simplifies the first three sub-goals, while (rule foo, simp_all)[]
  5075 simplifies all new goals that emerge from applying rule foo to the
  5076 originally first one.
  5077 
  5078 * Isar: schematic goals are no longer restricted to higher-order
  5079 patterns; e.g. ``lemma "?P(?x)" by (rule TrueI)'' now works as
  5080 expected.
  5081 
  5082 * Isar: the conclusion of a long theorem statement is now either
  5083 'shows' (a simultaneous conjunction, as before), or 'obtains'
  5084 (essentially a disjunction of cases with local parameters and
  5085 assumptions).  The latter allows to express general elimination rules
  5086 adequately; in this notation common elimination rules look like this:
  5087 
  5088   lemma exE:    -- "EX x. P x ==> (!!x. P x ==> thesis) ==> thesis"
  5089     assumes "EX x. P x"
  5090     obtains x where "P x"
  5091 
  5092   lemma conjE:  -- "A & B ==> (A ==> B ==> thesis) ==> thesis"
  5093     assumes "A & B"
  5094     obtains A and B
  5095 
  5096   lemma disjE:  -- "A | B ==> (A ==> thesis) ==> (B ==> thesis) ==> thesis"
  5097     assumes "A | B"
  5098     obtains
  5099       A
  5100     | B
  5101 
  5102 The subsequent classical rules even refer to the formal "thesis"
  5103 explicitly:
  5104 
  5105   lemma classical:     -- "(~ thesis ==> thesis) ==> thesis"
  5106     obtains "~ thesis"
  5107 
  5108   lemma Peirce's_Law:  -- "((thesis ==> something) ==> thesis) ==> thesis"
  5109     obtains "thesis ==> something"
  5110 
  5111 The actual proof of an 'obtains' statement is analogous to that of the
  5112 Isar proof element 'obtain', only that there may be several cases.
  5113 Optional case names may be specified in parentheses; these will be
  5114 available both in the present proof and as annotations in the
  5115 resulting rule, for later use with the 'cases' method (cf. attribute
  5116 case_names).
  5117 
  5118 * Isar: the assumptions of a long theorem statement are available as
  5119 "assms" fact in the proof context.  This is more appropriate than the
  5120 (historical) "prems", which refers to all assumptions of the current
  5121 context, including those from the target locale, proof body etc.
  5122 
  5123 * Isar: 'print_statement' prints theorems from the current theory or
  5124 proof context in long statement form, according to the syntax of a
  5125 top-level lemma.
  5126 
  5127 * Isar: 'obtain' takes an optional case name for the local context
  5128 introduction rule (default "that").
  5129 
  5130 * Isar: removed obsolete 'concl is' patterns.  INCOMPATIBILITY, use
  5131 explicit (is "_ ==> ?foo") in the rare cases where this still happens
  5132 to occur.
  5133 
  5134 * Pure: syntax "CONST name" produces a fully internalized constant
  5135 according to the current context.  This is particularly useful for
  5136 syntax translations that should refer to internal constant
  5137 representations independently of name spaces.
  5138 
  5139 * Pure: syntax constant for foo (binder "FOO ") is called "foo_binder"
  5140 instead of "FOO ". This allows multiple binder declarations to coexist
  5141 in the same context.  INCOMPATIBILITY.
  5142 
  5143 * Isar/locales: 'notation' provides a robust interface to the 'syntax'
  5144 primitive that also works in a locale context (both for constants and
  5145 fixed variables). Type declaration and internal syntactic representation
  5146 of given constants retrieved from the context. Likewise, the
  5147 'no_notation' command allows to remove given syntax annotations from the
  5148 current context.
  5149 
  5150 * Isar/locales: new derived specification elements 'axiomatization',
  5151 'definition', 'abbreviation', which support type-inference, admit
  5152 object-level specifications (equality, equivalence).  See also the
  5153 isar-ref manual.  Examples:
  5154 
  5155   axiomatization
  5156     eq  (infix "===" 50) where
  5157     eq_refl: "x === x" and eq_subst: "x === y ==> P x ==> P y"
  5158 
  5159   definition "f x y = x + y + 1"
  5160   definition g where "g x = f x x"
  5161 
  5162   abbreviation
  5163     neq  (infix "=!=" 50) where
  5164     "x =!= y == ~ (x === y)"
  5165 
  5166 These specifications may be also used in a locale context.  Then the
  5167 constants being introduced depend on certain fixed parameters, and the
  5168 constant name is qualified by the locale base name.  An internal
  5169 abbreviation takes care for convenient input and output, making the
  5170 parameters implicit and using the original short name.  See also
  5171 src/HOL/ex/Abstract_NAT.thy for an example of deriving polymorphic
  5172 entities from a monomorphic theory.
  5173 
  5174 Presently, abbreviations are only available 'in' a target locale, but
  5175 not inherited by general import expressions.  Also note that
  5176 'abbreviation' may be used as a type-safe replacement for 'syntax' +
  5177 'translations' in common applications.  The "no_abbrevs" print mode
  5178 prevents folding of abbreviations in term output.
  5179 
  5180 Concrete syntax is attached to specified constants in internal form,
  5181 independently of name spaces.  The parse tree representation is
  5182 slightly different -- use 'notation' instead of raw 'syntax', and
  5183 'translations' with explicit "CONST" markup to accommodate this.
  5184 
  5185 * Pure/Isar: unified syntax for new-style specification mechanisms
  5186 (e.g.  'definition', 'abbreviation', or 'inductive' in HOL) admits
  5187 full type inference and dummy patterns ("_").  For example:
  5188 
  5189   definition "K x _ = x"
  5190 
  5191   inductive conj for A B
  5192   where "A ==> B ==> conj A B"
  5193 
  5194 * Pure: command 'print_abbrevs' prints all constant abbreviations of
  5195 the current context.  Print mode "no_abbrevs" prevents inversion of
  5196 abbreviations on output.
  5197 
  5198 * Isar/locales: improved parameter handling: use of locales "var" and
  5199 "struct" no longer necessary; - parameter renamings are no longer
  5200 required to be injective.  For example, this allows to define
  5201 endomorphisms as locale endom = homom mult mult h.
  5202 
  5203 * Isar/locales: changed the way locales with predicates are defined.
  5204 Instead of accumulating the specification, the imported expression is
  5205 now an interpretation.  INCOMPATIBILITY: different normal form of
  5206 locale expressions.  In particular, in interpretations of locales with
  5207 predicates, goals repesenting already interpreted fragments are not
  5208 removed automatically.  Use methods `intro_locales' and
  5209 `unfold_locales'; see below.
  5210 
  5211 * Isar/locales: new methods `intro_locales' and `unfold_locales'
  5212 provide backward reasoning on locales predicates.  The methods are
  5213 aware of interpretations and discharge corresponding goals.
  5214 `intro_locales' is less aggressive then `unfold_locales' and does not
  5215 unfold predicates to assumptions.
  5216 
  5217 * Isar/locales: the order in which locale fragments are accumulated
  5218 has changed.  This enables to override declarations from fragments due
  5219 to interpretations -- for example, unwanted simp rules.
  5220 
  5221 * Isar/locales: interpretation in theories and proof contexts has been
  5222 extended.  One may now specify (and prove) equations, which are
  5223 unfolded in interpreted theorems.  This is useful for replacing
  5224 defined concepts (constants depending on locale parameters) by
  5225 concepts already existing in the target context.  Example:
  5226 
  5227   interpretation partial_order ["op <= :: [int, int] => bool"]
  5228     where "partial_order.less (op <=) (x::int) y = (x < y)"
  5229 
  5230 Typically, the constant `partial_order.less' is created by a
  5231 definition specification element in the context of locale
  5232 partial_order.
  5233 
  5234 * Method "induct": improved internal context management to support
  5235 local fixes and defines on-the-fly. Thus explicit meta-level
  5236 connectives !!  and ==> are rarely required anymore in inductive goals
  5237 (using object-logic connectives for this purpose has been long
  5238 obsolete anyway). Common proof patterns are explained in
  5239 src/HOL/Induct/Common_Patterns.thy, see also
  5240 src/HOL/Isar_examples/Puzzle.thy and src/HOL/Lambda for realistic
  5241 examples.
  5242 
  5243 * Method "induct": improved handling of simultaneous goals. Instead of
  5244 introducing object-level conjunction, the statement is now split into
  5245 several conclusions, while the corresponding symbolic cases are nested
  5246 accordingly. INCOMPATIBILITY, proofs need to be structured explicitly,
  5247 see src/HOL/Induct/Common_Patterns.thy, for example.
  5248 
  5249 * Method "induct": mutual induction rules are now specified as a list
  5250 of rule sharing the same induction cases. HOL packages usually provide
  5251 foo_bar.inducts for mutually defined items foo and bar (e.g. inductive
  5252 predicates/sets or datatypes). INCOMPATIBILITY, users need to specify
  5253 mutual induction rules differently, i.e. like this:
  5254 
  5255   (induct rule: foo_bar.inducts)
  5256   (induct set: foo bar)
  5257   (induct pred: foo bar)
  5258   (induct type: foo bar)
  5259 
  5260 The ML function ProjectRule.projections turns old-style rules into the
  5261 new format.
  5262 
  5263 * Method "coinduct": dual of induction, see
  5264 src/HOL/Library/Coinductive_List.thy for various examples.
  5265 
  5266 * Method "cases", "induct", "coinduct": the ``(open)'' option is
  5267 considered a legacy feature.
  5268 
  5269 * Attribute "symmetric" produces result with standardized schematic
  5270 variables (index 0).  Potential INCOMPATIBILITY.
  5271 
  5272 * Simplifier: by default the simplifier trace only shows top level
  5273 rewrites now. That is, trace_simp_depth_limit is set to 1 by
  5274 default. Thus there is less danger of being flooded by the trace. The
  5275 trace indicates where parts have been suppressed.
  5276   
  5277 * Provers/classical: removed obsolete classical version of elim_format
  5278 attribute; classical elim/dest rules are now treated uniformly when
  5279 manipulating the claset.
  5280 
  5281 * Provers/classical: stricter checks to ensure that supplied intro,
  5282 dest and elim rules are well-formed; dest and elim rules must have at
  5283 least one premise.
  5284 
  5285 * Provers/classical: attributes dest/elim/intro take an optional
  5286 weight argument for the rule (just as the Pure versions).  Weights are
  5287 ignored by automated tools, but determine the search order of single
  5288 rule steps.
  5289 
  5290 * Syntax: input syntax now supports dummy variable binding "%_. b",
  5291 where the body does not mention the bound variable.  Note that dummy
  5292 patterns implicitly depend on their context of bounds, which makes
  5293 "{_. _}" match any set comprehension as expected.  Potential
  5294 INCOMPATIBILITY -- parse translations need to cope with syntactic
  5295 constant "_idtdummy" in the binding position.
  5296 
  5297 * Syntax: removed obsolete syntactic constant "_K" and its associated
  5298 parse translation.  INCOMPATIBILITY -- use dummy abstraction instead,
  5299 for example "A -> B" => "Pi A (%_. B)".
  5300 
  5301 * Pure: 'class_deps' command visualizes the subclass relation, using
  5302 the graph browser tool.
  5303 
  5304 * Pure: 'print_theory' now suppresses certain internal declarations by
  5305 default; use '!' option for full details.
  5306 
  5307 
  5308 *** HOL ***
  5309 
  5310 * Method "metis" proves goals by applying the Metis general-purpose
  5311 resolution prover (see also http://gilith.com/software/metis/).
  5312 Examples are in the directory MetisExamples.  WARNING: the
  5313 Isabelle/HOL-Metis integration does not yet work properly with
  5314 multi-threading.
  5315   
  5316 * Command 'sledgehammer' invokes external automatic theorem provers as
  5317 background processes.  It generates calls to the "metis" method if
  5318 successful. These can be pasted into the proof.  Users do not have to
  5319 wait for the automatic provers to return.  WARNING: does not really
  5320 work with multi-threading.
  5321 
  5322 * New "auto_quickcheck" feature tests outermost goal statements for
  5323 potential counter-examples.  Controlled by ML references
  5324 auto_quickcheck (default true) and auto_quickcheck_time_limit (default
  5325 5000 milliseconds).  Fails silently if statements is outside of
  5326 executable fragment, or any other codgenerator problem occurs.
  5327 
  5328 * New constant "undefined" with axiom "undefined x = undefined".
  5329 
  5330 * Added class "HOL.eq", allowing for code generation with polymorphic
  5331 equality.
  5332 
  5333 * Some renaming of class constants due to canonical name prefixing in
  5334 the new 'class' package:
  5335 
  5336     HOL.abs ~> HOL.abs_class.abs
  5337     HOL.divide ~> HOL.divide_class.divide
  5338     0 ~> HOL.zero_class.zero
  5339     1 ~> HOL.one_class.one
  5340     op + ~> HOL.plus_class.plus
  5341     op - ~> HOL.minus_class.minus
  5342     uminus ~> HOL.minus_class.uminus
  5343     op * ~> HOL.times_class.times
  5344     op < ~> HOL.ord_class.less
  5345     op <= > HOL.ord_class.less_eq
  5346     Nat.power ~> Power.power_class.power
  5347     Nat.size ~> Nat.size_class.size
  5348     Numeral.number_of ~> Numeral.number_class.number_of
  5349     FixedPoint.Inf ~> Lattices.complete_lattice_class.Inf
  5350     FixedPoint.Sup ~> Lattices.complete_lattice_class.Sup
  5351     Orderings.min ~> Orderings.ord_class.min
  5352     Orderings.max ~> Orderings.ord_class.max
  5353     Divides.op div ~> Divides.div_class.div
  5354     Divides.op mod ~> Divides.div_class.mod
  5355     Divides.op dvd ~> Divides.div_class.dvd
  5356 
  5357 INCOMPATIBILITY.  Adaptions may be required in the following cases:
  5358 
  5359 a) User-defined constants using any of the names "plus", "minus",
  5360 "times", "less" or "less_eq". The standard syntax translations for
  5361 "+", "-" and "*" may go wrong.  INCOMPATIBILITY: use more specific
  5362 names.
  5363 
  5364 b) Variables named "plus", "minus", "times", "less", "less_eq"
  5365 INCOMPATIBILITY: use more specific names.
  5366 
  5367 c) Permutative equations (e.g. "a + b = b + a")
  5368 Since the change of names also changes the order of terms, permutative
  5369 rewrite rules may get applied in a different order. Experience shows
  5370 that this is rarely the case (only two adaptions in the whole Isabelle
  5371 distribution).  INCOMPATIBILITY: rewrite proofs
  5372 
  5373 d) ML code directly refering to constant names
  5374 This in general only affects hand-written proof tactics, simprocs and
  5375 so on.  INCOMPATIBILITY: grep your sourcecode and replace names.
  5376 Consider using @{const_name} antiquotation.
  5377 
  5378 * New class "default" with associated constant "default".
  5379 
  5380 * Function "sgn" is now overloaded and available on int, real, complex
  5381 (and other numeric types), using class "sgn".  Two possible defs of
  5382 sgn are given as equational assumptions in the classes sgn_if and
  5383 sgn_div_norm; ordered_idom now also inherits from sgn_if.
  5384 INCOMPATIBILITY.
  5385 
  5386 * Locale "partial_order" now unified with class "order" (cf. theory
  5387 Orderings), added parameter "less".  INCOMPATIBILITY.
  5388 
  5389 * Renamings in classes "order" and "linorder": facts "refl", "trans" and
  5390 "cases" to "order_refl", "order_trans" and "linorder_cases", to avoid
  5391 clashes with HOL "refl" and "trans".  INCOMPATIBILITY.
  5392 
  5393 * Classes "order" and "linorder": potential INCOMPATIBILITY due to
  5394 changed order of proof goals in instance proofs.
  5395 
  5396 * The transitivity reasoner for partial and linear orders is set up
  5397 for classes "order" and "linorder".  Instances of the reasoner are available
  5398 in all contexts importing or interpreting the corresponding locales.
  5399 Method "order" invokes the reasoner separately; the reasoner
  5400 is also integrated with the Simplifier as a solver.  Diagnostic
  5401 command 'print_orders' shows the available instances of the reasoner
  5402 in the current context.
  5403 
  5404 * Localized monotonicity predicate in theory "Orderings"; integrated
  5405 lemmas max_of_mono and min_of_mono with this predicate.
  5406 INCOMPATIBILITY.
  5407 
  5408 * Formulation of theorem "dense" changed slightly due to integration
  5409 with new class dense_linear_order.
  5410 
  5411 * Uniform lattice theory development in HOL.
  5412 
  5413     constants "meet" and "join" now named "inf" and "sup"
  5414     constant "Meet" now named "Inf"
  5415 
  5416     classes "meet_semilorder" and "join_semilorder" now named
  5417       "lower_semilattice" and "upper_semilattice"
  5418     class "lorder" now named "lattice"
  5419     class "comp_lat" now named "complete_lattice"
  5420 
  5421     Instantiation of lattice classes allows explicit definitions
  5422     for "inf" and "sup" operations (or "Inf" and "Sup" for complete lattices).
  5423 
  5424   INCOMPATIBILITY.  Theorem renames:
  5425 
  5426     meet_left_le            ~> inf_le1
  5427     meet_right_le           ~> inf_le2
  5428     join_left_le            ~> sup_ge1
  5429     join_right_le           ~> sup_ge2
  5430     meet_join_le            ~> inf_sup_ord
  5431     le_meetI                ~> le_infI
  5432     join_leI                ~> le_supI
  5433     le_meet                 ~> le_inf_iff
  5434     le_join                 ~> ge_sup_conv
  5435     meet_idempotent         ~> inf_idem
  5436     join_idempotent         ~> sup_idem
  5437     meet_comm               ~> inf_commute
  5438     join_comm               ~> sup_commute
  5439     meet_leI1               ~> le_infI1
  5440     meet_leI2               ~> le_infI2
  5441     le_joinI1               ~> le_supI1
  5442     le_joinI2               ~> le_supI2
  5443     meet_assoc              ~> inf_assoc
  5444     join_assoc              ~> sup_assoc
  5445     meet_left_comm          ~> inf_left_commute
  5446     meet_left_idempotent    ~> inf_left_idem
  5447     join_left_comm          ~> sup_left_commute
  5448     join_left_idempotent    ~> sup_left_idem
  5449     meet_aci                ~> inf_aci
  5450     join_aci                ~> sup_aci
  5451     le_def_meet             ~> le_iff_inf
  5452     le_def_join             ~> le_iff_sup
  5453     join_absorp2            ~> sup_absorb2
  5454     join_absorp1            ~> sup_absorb1
  5455     meet_absorp1            ~> inf_absorb1
  5456     meet_absorp2            ~> inf_absorb2
  5457     meet_join_absorp        ~> inf_sup_absorb
  5458     join_meet_absorp        ~> sup_inf_absorb
  5459     distrib_join_le         ~> distrib_sup_le
  5460     distrib_meet_le         ~> distrib_inf_le
  5461 
  5462     add_meet_distrib_left   ~> add_inf_distrib_left
  5463     add_join_distrib_left   ~> add_sup_distrib_left
  5464     is_join_neg_meet        ~> is_join_neg_inf
  5465     is_meet_neg_join        ~> is_meet_neg_sup
  5466     add_meet_distrib_right  ~> add_inf_distrib_right
  5467     add_join_distrib_right  ~> add_sup_distrib_right
  5468     add_meet_join_distribs  ~> add_sup_inf_distribs
  5469     join_eq_neg_meet        ~> sup_eq_neg_inf
  5470     meet_eq_neg_join        ~> inf_eq_neg_sup
  5471     add_eq_meet_join        ~> add_eq_inf_sup
  5472     meet_0_imp_0            ~> inf_0_imp_0
  5473     join_0_imp_0            ~> sup_0_imp_0
  5474     meet_0_eq_0             ~> inf_0_eq_0
  5475     join_0_eq_0             ~> sup_0_eq_0
  5476     neg_meet_eq_join        ~> neg_inf_eq_sup
  5477     neg_join_eq_meet        ~> neg_sup_eq_inf
  5478     join_eq_if              ~> sup_eq_if
  5479 
  5480     mono_meet               ~> mono_inf
  5481     mono_join               ~> mono_sup
  5482     meet_bool_eq            ~> inf_bool_eq
  5483     join_bool_eq            ~> sup_bool_eq
  5484     meet_fun_eq             ~> inf_fun_eq
  5485     join_fun_eq             ~> sup_fun_eq
  5486     meet_set_eq             ~> inf_set_eq
  5487     join_set_eq             ~> sup_set_eq
  5488     meet1_iff               ~> inf1_iff
  5489     meet2_iff               ~> inf2_iff
  5490     meet1I                  ~> inf1I
  5491     meet2I                  ~> inf2I
  5492     meet1D1                 ~> inf1D1
  5493     meet2D1                 ~> inf2D1
  5494     meet1D2                 ~> inf1D2
  5495     meet2D2                 ~> inf2D2
  5496     meet1E                  ~> inf1E
  5497     meet2E                  ~> inf2E
  5498     join1_iff               ~> sup1_iff
  5499     join2_iff               ~> sup2_iff
  5500     join1I1                 ~> sup1I1
  5501     join2I1                 ~> sup2I1
  5502     join1I1                 ~> sup1I1
  5503     join2I2                 ~> sup1I2
  5504     join1CI                 ~> sup1CI
  5505     join2CI                 ~> sup2CI
  5506     join1E                  ~> sup1E
  5507     join2E                  ~> sup2E
  5508 
  5509     is_meet_Meet            ~> is_meet_Inf
  5510     Meet_bool_def           ~> Inf_bool_def
  5511     Meet_fun_def            ~> Inf_fun_def
  5512     Meet_greatest           ~> Inf_greatest
  5513     Meet_lower              ~> Inf_lower
  5514     Meet_set_def            ~> Inf_set_def
  5515 
  5516     Sup_def                 ~> Sup_Inf
  5517     Sup_bool_eq             ~> Sup_bool_def
  5518     Sup_fun_eq              ~> Sup_fun_def
  5519     Sup_set_eq              ~> Sup_set_def
  5520 
  5521     listsp_meetI            ~> listsp_infI
  5522     listsp_meet_eq          ~> listsp_inf_eq
  5523 
  5524     meet_min                ~> inf_min
  5525     join_max                ~> sup_max
  5526 
  5527 * Added syntactic class "size"; overloaded constant "size" now has
  5528 type "'a::size ==> bool"
  5529 
  5530 * Internal reorganisation of `size' of datatypes: size theorems
  5531 "foo.size" are no longer subsumed by "foo.simps" (but are still
  5532 simplification rules by default!); theorems "prod.size" now named
  5533 "*.size".
  5534 
  5535 * Class "div" now inherits from class "times" rather than "type".
  5536 INCOMPATIBILITY.
  5537 
  5538 * HOL/Finite_Set: "name-space" locales Lattice, Distrib_lattice,
  5539 Linorder etc.  have disappeared; operations defined in terms of
  5540 fold_set now are named Inf_fin, Sup_fin.  INCOMPATIBILITY.
  5541 
  5542 * HOL/Nat: neq0_conv no longer declared as iff.  INCOMPATIBILITY.
  5543 
  5544 * HOL-Word: New extensive library and type for generic, fixed size
  5545 machine words, with arithmetic, bit-wise, shifting and rotating
  5546 operations, reflection into int, nat, and bool lists, automation for
  5547 linear arithmetic (by automatic reflection into nat or int), including
  5548 lemmas on overflow and monotonicity.  Instantiated to all appropriate
  5549 arithmetic type classes, supporting automatic simplification of
  5550 numerals on all operations.
  5551 
  5552 * Library/Boolean_Algebra: locales for abstract boolean algebras.
  5553 
  5554 * Library/Numeral_Type: numbers as types, e.g. TYPE(32).
  5555 
  5556 * Code generator library theories:
  5557   - Code_Integer represents HOL integers by big integer literals in target
  5558     languages.
  5559   - Code_Char represents HOL characters by character literals in target
  5560     languages.
  5561   - Code_Char_chr like Code_Char, but also offers treatment of character
  5562     codes; includes Code_Integer.
  5563   - Executable_Set allows to generate code for finite sets using lists.
  5564   - Executable_Rat implements rational numbers as triples (sign, enumerator,
  5565     denominator).
  5566   - Executable_Real implements a subset of real numbers, namly those
  5567     representable by rational numbers.
  5568   - Efficient_Nat implements natural numbers by integers, which in general will
  5569     result in higher efficency; pattern matching with 0/Suc is eliminated;
  5570     includes Code_Integer.
  5571   - Code_Index provides an additional datatype index which is mapped to
  5572     target-language built-in integers.
  5573   - Code_Message provides an additional datatype message_string which is isomorphic to
  5574     strings; messages are mapped to target-language strings.
  5575 
  5576 * New package for inductive predicates
  5577 
  5578   An n-ary predicate p with m parameters z_1, ..., z_m can now be defined via
  5579 
  5580     inductive
  5581       p :: "U_1 => ... => U_m => T_1 => ... => T_n => bool"
  5582       for z_1 :: U_1 and ... and z_n :: U_m
  5583     where
  5584       rule_1: "... ==> p z_1 ... z_m t_1_1 ... t_1_n"
  5585     | ...
  5586 
  5587   with full support for type-inference, rather than
  5588 
  5589     consts s :: "U_1 => ... => U_m => (T_1 * ... * T_n) set"
  5590 
  5591     abbreviation p :: "U_1 => ... => U_m => T_1 => ... => T_n => bool"
  5592     where "p z_1 ... z_m x_1 ... x_n == (x_1, ..., x_n) : s z_1 ... z_m"
  5593 
  5594     inductive "s z_1 ... z_m"
  5595     intros
  5596       rule_1: "... ==> (t_1_1, ..., t_1_n) : s z_1 ... z_m"
  5597       ...
  5598 
  5599   For backward compatibility, there is a wrapper allowing inductive
  5600   sets to be defined with the new package via
  5601 
  5602     inductive_set
  5603       s :: "U_1 => ... => U_m => (T_1 * ... * T_n) set"
  5604       for z_1 :: U_1 and ... and z_n :: U_m
  5605     where
  5606       rule_1: "... ==> (t_1_1, ..., t_1_n) : s z_1 ... z_m"
  5607     | ...
  5608 
  5609   or
  5610 
  5611     inductive_set
  5612       s :: "U_1 => ... => U_m => (T_1 * ... * T_n) set"
  5613       and p :: "U_1 => ... => U_m => T_1 => ... => T_n => bool"
  5614       for z_1 :: U_1 and ... and z_n :: U_m
  5615     where
  5616       "p z_1 ... z_m x_1 ... x_n == (x_1, ..., x_n) : s z_1 ... z_m"
  5617     | rule_1: "... ==> p z_1 ... z_m t_1_1 ... t_1_n"
  5618     | ...
  5619 
  5620   if the additional syntax "p ..." is required.
  5621 
  5622   Numerous examples can be found in the subdirectories src/HOL/Auth,
  5623   src/HOL/Bali, src/HOL/Induct, and src/HOL/MicroJava.
  5624 
  5625   INCOMPATIBILITIES:
  5626 
  5627   - Since declaration and definition of inductive sets or predicates
  5628     is no longer separated, abbreviations involving the newly
  5629     introduced sets or predicates must be specified together with the
  5630     introduction rules after the 'where' keyword (see above), rather
  5631     than before the actual inductive definition.
  5632 
  5633   - The variables in induction and elimination rules are now
  5634     quantified in the order of their occurrence in the introduction
  5635     rules, rather than in alphabetical order. Since this may break
  5636     some proofs, these proofs either have to be repaired, e.g. by
  5637     reordering the variables a_i_1 ... a_i_{k_i} in Isar 'case'
  5638     statements of the form
  5639 
  5640       case (rule_i a_i_1 ... a_i_{k_i})
  5641 
  5642     or the old order of quantification has to be restored by explicitly adding
  5643     meta-level quantifiers in the introduction rules, i.e.
  5644 
  5645       | rule_i: "!!a_i_1 ... a_i_{k_i}. ... ==> p z_1 ... z_m t_i_1 ... t_i_n"
  5646 
  5647   - The format of the elimination rules is now
  5648 
  5649       p z_1 ... z_m x_1 ... x_n ==>
  5650         (!!a_1_1 ... a_1_{k_1}. x_1 = t_1_1 ==> ... ==> x_n = t_1_n ==> ... ==> P)
  5651         ==> ... ==> P
  5652 
  5653     for predicates and
  5654 
  5655       (x_1, ..., x_n) : s z_1 ... z_m ==>
  5656         (!!a_1_1 ... a_1_{k_1}. x_1 = t_1_1 ==> ... ==> x_n = t_1_n ==> ... ==> P)
  5657         ==> ... ==> P
  5658 
  5659     for sets rather than
  5660 
  5661       x : s z_1 ... z_m ==>
  5662         (!!a_1_1 ... a_1_{k_1}. x = (t_1_1, ..., t_1_n) ==> ... ==> P)
  5663         ==> ... ==> P
  5664 
  5665     This may require terms in goals to be expanded to n-tuples
  5666     (e.g. using case_tac or simplification with the split_paired_all
  5667     rule) before the above elimination rule is applicable.
  5668 
  5669   - The elimination or case analysis rules for (mutually) inductive
  5670     sets or predicates are now called "p_1.cases" ... "p_k.cases". The
  5671     list of rules "p_1_..._p_k.elims" is no longer available.
  5672 
  5673 * New package "function"/"fun" for general recursive functions,
  5674 supporting mutual and nested recursion, definitions in local contexts,
  5675 more general pattern matching and partiality. See HOL/ex/Fundefs.thy
  5676 for small examples, and the separate tutorial on the function
  5677 package. The old recdef "package" is still available as before, but
  5678 users are encouraged to use the new package.
  5679 
  5680 * Method "lexicographic_order" automatically synthesizes termination
  5681 relations as lexicographic combinations of size measures. 
  5682 
  5683 * Case-expressions allow arbitrary constructor-patterns (including
  5684 "_") and take their order into account, like in functional
  5685 programming.  Internally, this is translated into nested
  5686 case-expressions; missing cases are added and mapped to the predefined
  5687 constant "undefined". In complicated cases printing may no longer show
  5688 the original input but the internal form. Lambda-abstractions allow
  5689 the same form of pattern matching: "% pat1 => e1 | ..." is an
  5690 abbreviation for "%x. case x of pat1 => e1 | ..." where x is a new
  5691 variable.
  5692 
  5693 * IntDef: The constant "int :: nat => int" has been removed; now "int"
  5694 is an abbreviation for "of_nat :: nat => int". The simplification
  5695 rules for "of_nat" have been changed to work like "int" did
  5696 previously.  Potential INCOMPATIBILITY:
  5697   - "of_nat (Suc m)" simplifies to "1 + of_nat m" instead of "of_nat m + 1"
  5698   - of_nat_diff and of_nat_mult are no longer default simp rules
  5699 
  5700 * Method "algebra" solves polynomial equations over (semi)rings using
  5701 Groebner bases. The (semi)ring structure is defined by locales and the
  5702 tool setup depends on that generic context. Installing the method for
  5703 a specific type involves instantiating the locale and possibly adding
  5704 declarations for computation on the coefficients.  The method is
  5705 already instantiated for natural numbers and for the axiomatic class
  5706 of idoms with numerals.  See also the paper by Chaieb and Wenzel at
  5707 CALCULEMUS 2007 for the general principles underlying this
  5708 architecture of context-aware proof-tools.
  5709 
  5710 * Method "ferrack" implements quantifier elimination over
  5711 special-purpose dense linear orders using locales (analogous to
  5712 "algebra"). The method is already installed for class
  5713 {ordered_field,recpower,number_ring} which subsumes real, hyperreal,
  5714 rat, etc.
  5715 
  5716 * Former constant "List.op @" now named "List.append".  Use ML
  5717 antiquotations @{const_name List.append} or @{term " ... @ ... "} to
  5718 circumvent possible incompatibilities when working on ML level.
  5719 
  5720 * primrec: missing cases mapped to "undefined" instead of "arbitrary".
  5721 
  5722 * New function listsum :: 'a list => 'a for arbitrary monoids.
  5723 Special syntax: "SUM x <- xs. f x" (and latex variants)
  5724 
  5725 * New syntax for Haskell-like list comprehension (input only), eg.
  5726 [(x,y). x <- xs, y <- ys, x ~= y], see also src/HOL/List.thy.
  5727 
  5728 * The special syntax for function "filter" has changed from [x :
  5729 xs. P] to [x <- xs. P] to avoid an ambiguity caused by list
  5730 comprehension syntax, and for uniformity.  INCOMPATIBILITY.
  5731 
  5732 * [a..b] is now defined for arbitrary linear orders.  It used to be
  5733 defined on nat only, as an abbreviation for [a..<Suc b]
  5734 INCOMPATIBILITY.
  5735 
  5736 * Renamed lemma "set_take_whileD"  to "set_takeWhileD".
  5737 
  5738 * New functions "sorted" and "sort" in src/HOL/List.thy.
  5739 
  5740 * New lemma collection field_simps (an extension of ring_simps) for
  5741 manipulating (in)equations involving division. Multiplies with all
  5742 denominators that can be proved to be non-zero (in equations) or
  5743 positive/negative (in inequations).
  5744 
  5745 * Lemma collections ring_eq_simps, group_eq_simps and ring_distrib
  5746 have been improved and renamed to ring_simps, group_simps and
  5747 ring_distribs.  Removed lemmas field_xyz in theory Ring_and_Field
  5748 because they were subsumed by lemmas xyz.  INCOMPATIBILITY.
  5749 
  5750 * Theory Library/Commutative_Ring: switched from recdef to function
  5751 package; constants add, mul, pow now curried.  Infix syntax for
  5752 algebraic operations.
  5753 
  5754 * Dropped redundant lemma def_imp_eq in favor of meta_eq_to_obj_eq.
  5755 INCOMPATIBILITY.
  5756 
  5757 * Dropped redundant lemma if_def2 in favor of if_bool_eq_conj.
  5758 INCOMPATIBILITY.
  5759 
  5760 * HOL/records: generalised field-update to take a function on the
  5761 field rather than the new value: r(|A := x|) is translated to A_update
  5762 (K x) r The K-combinator that is internally used is called K_record.
  5763 INCOMPATIBILITY: Usage of the plain update functions has to be
  5764 adapted.
  5765  
  5766 * Class "semiring_0" now contains annihilation axioms x * 0 = 0 and 0
  5767 * x = 0, which are required for a semiring.  Richer structures do not
  5768 inherit from semiring_0 anymore, because this property is a theorem
  5769 there, not an axiom.  INCOMPATIBILITY: In instances of semiring_0,
  5770 there is more to prove, but this is mostly trivial.
  5771 
  5772 * Class "recpower" is generalized to arbitrary monoids, not just
  5773 commutative semirings.  INCOMPATIBILITY: may need to incorporate
  5774 commutativity or semiring properties additionally.
  5775 
  5776 * Constant "List.list_all2" in List.thy now uses authentic syntax.
  5777 INCOMPATIBILITY: translations containing list_all2 may go wrong,
  5778 better use 'abbreviation'.
  5779 
  5780 * Renamed constant "List.op mem" to "List.member".  INCOMPATIBILITY.
  5781 
  5782 * Numeral syntax: type 'bin' which was a mere type copy of 'int' has
  5783 been abandoned in favour of plain 'int'.  INCOMPATIBILITY --
  5784 significant changes for setting up numeral syntax for types:
  5785   - New constants Numeral.pred and Numeral.succ instead
  5786       of former Numeral.bin_pred and Numeral.bin_succ.
  5787   - Use integer operations instead of bin_add, bin_mult and so on.
  5788   - Numeral simplification theorems named Numeral.numeral_simps instead of Bin_simps.
  5789   - ML structure Bin_Simprocs now named Int_Numeral_Base_Simprocs.
  5790 
  5791 See src/HOL/Integ/IntArith.thy for an example setup.
  5792 
  5793 * Command 'normal_form' computes the normal form of a term that may
  5794 contain free variables.  For example ``normal_form "rev [a, b, c]"''
  5795 produces ``[b, c, a]'' (without proof).  This command is suitable for
  5796 heavy-duty computations because the functions are compiled to ML
  5797 first.  Correspondingly, a method "normalization" is provided.  See
  5798 further src/HOL/ex/NormalForm.thy and src/Tools/nbe.ML.
  5799 
  5800 * Alternative iff syntax "A <-> B" for equality on bool (with priority
  5801 25 like -->); output depends on the "iff" print_mode, the default is
  5802 "A = B" (with priority 50).
  5803 
  5804 * Relations less (<) and less_eq (<=) are also available on type bool.
  5805 Modified syntax to disallow nesting without explicit parentheses,
  5806 e.g. "(x < y) < z" or "x < (y < z)", but NOT "x < y < z".  Potential
  5807 INCOMPATIBILITY.
  5808 
  5809 * "LEAST x:A. P" expands to "LEAST x. x:A & P" (input only).
  5810 
  5811 * Relation composition operator "op O" now has precedence 75 and binds
  5812 stronger than union and intersection. INCOMPATIBILITY.
  5813 
  5814 * The old set interval syntax "{m..n(}" (and relatives) has been
  5815 removed.  Use "{m..<n}" (and relatives) instead.
  5816 
  5817 * In the context of the assumption "~(s = t)" the Simplifier rewrites
  5818 "t = s" to False (by simproc "neq").  INCOMPATIBILITY, consider using
  5819 ``declare [[simproc del: neq]]''.
  5820 
  5821 * Simplifier: "m dvd n" where m and n are numbers is evaluated to
  5822 True/False.
  5823 
  5824 * Theorem Cons_eq_map_conv no longer declared as "simp".
  5825 
  5826 * Theorem setsum_mult renamed to setsum_right_distrib.
  5827 
  5828 * Prefer ex1I over ex_ex1I in single-step reasoning, e.g. by the
  5829 ``rule'' method.
  5830 
  5831 * Reimplemented methods "sat" and "satx", with several improvements:
  5832 goals no longer need to be stated as "<prems> ==> False", equivalences
  5833 (i.e. "=" on type bool) are handled, variable names of the form
  5834 "lit_<n>" are no longer reserved, significant speedup.
  5835 
  5836 * Methods "sat" and "satx" can now replay MiniSat proof traces.
  5837 zChaff is still supported as well.
  5838 
  5839 * 'inductive' and 'datatype': provide projections of mutual rules,
  5840 bundled as foo_bar.inducts;
  5841 
  5842 * Library: moved theories Parity, GCD, Binomial, Infinite_Set to
  5843 Library.
  5844 
  5845 * Library: moved theory Accessible_Part to main HOL.
  5846 
  5847 * Library: added theory Coinductive_List of potentially infinite lists
  5848 as greatest fixed-point.
  5849 
  5850 * Library: added theory AssocList which implements (finite) maps as
  5851 association lists.
  5852 
  5853 * Method "evaluation" solves goals (i.e. a boolean expression)
  5854 efficiently by compiling it to ML.  The goal is "proved" (via an
  5855 oracle) if it evaluates to True.
  5856 
  5857 * Linear arithmetic now splits certain operators (e.g. min, max, abs)
  5858 also when invoked by the simplifier.  This results in the Simplifier
  5859 being more powerful on arithmetic goals.  INCOMPATIBILITY.
  5860 Configuration option fast_arith_split_limit=0 recovers the old
  5861 behavior.
  5862 
  5863 * Support for hex (0x20) and binary (0b1001) numerals.
  5864 
  5865 * New method: reify eqs (t), where eqs are equations for an
  5866 interpretation I :: 'a list => 'b => 'c and t::'c is an optional
  5867 parameter, computes a term s::'b and a list xs::'a list and proves the
  5868 theorem I xs s = t. This is also known as reification or quoting. The
  5869 resulting theorem is applied to the subgoal to substitute t with I xs
  5870 s.  If t is omitted, the subgoal itself is reified.
  5871 
  5872 * New method: reflection corr_thm eqs (t). The parameters eqs and (t)
  5873 are as explained above. corr_thm is a theorem for I vs (f t) = I vs t,
  5874 where f is supposed to be a computable function (in the sense of code
  5875 generattion). The method uses reify to compute s and xs as above then
  5876 applies corr_thm and uses normalization by evaluation to "prove" f s =
  5877 r and finally gets the theorem t = r, which is again applied to the
  5878 subgoal. An Example is available in src/HOL/ex/ReflectionEx.thy.
  5879 
  5880 * Reflection: Automatic reification now handels binding, an example is
  5881 available in src/HOL/ex/ReflectionEx.thy
  5882 
  5883 * HOL-Statespace: ``State Spaces: The Locale Way'' introduces a
  5884 command 'statespace' that is similar to 'record', but introduces an
  5885 abstract specification based on the locale infrastructure instead of
  5886 HOL types.  This leads to extra flexibility in composing state spaces,
  5887 in particular multiple inheritance and renaming of components.
  5888 
  5889 
  5890 *** HOL-Complex ***
  5891 
  5892 * Hyperreal: Functions root and sqrt are now defined on negative real
  5893 inputs so that root n (- x) = - root n x and sqrt (- x) = - sqrt x.
  5894 Nonnegativity side conditions have been removed from many lemmas, so
  5895 that more subgoals may now be solved by simplification; potential
  5896 INCOMPATIBILITY.
  5897 
  5898 * Real: new type classes formalize real normed vector spaces and
  5899 algebras, using new overloaded constants scaleR :: real => 'a => 'a
  5900 and norm :: 'a => real.
  5901 
  5902 * Real: constant of_real :: real => 'a::real_algebra_1 injects from
  5903 reals into other types. The overloaded constant Reals :: 'a set is now
  5904 defined as range of_real; potential INCOMPATIBILITY.
  5905 
  5906 * Real: proper support for ML code generation, including 'quickcheck'.
  5907 Reals are implemented as arbitrary precision rationals.
  5908 
  5909 * Hyperreal: Several constants that previously worked only for the
  5910 reals have been generalized, so they now work over arbitrary vector
  5911 spaces. Type annotations may need to be added in some cases; potential
  5912 INCOMPATIBILITY.
  5913 
  5914   Infinitesimal  :: ('a::real_normed_vector) star set
  5915   HFinite        :: ('a::real_normed_vector) star set
  5916   HInfinite      :: ('a::real_normed_vector) star set
  5917   approx         :: ('a::real_normed_vector) star => 'a star => bool
  5918   monad          :: ('a::real_normed_vector) star => 'a star set
  5919   galaxy         :: ('a::real_normed_vector) star => 'a star set
  5920   (NS)LIMSEQ     :: [nat => 'a::real_normed_vector, 'a] => bool
  5921   (NS)convergent :: (nat => 'a::real_normed_vector) => bool
  5922   (NS)Bseq       :: (nat => 'a::real_normed_vector) => bool
  5923   (NS)Cauchy     :: (nat => 'a::real_normed_vector) => bool
  5924   (NS)LIM        :: ['a::real_normed_vector => 'b::real_normed_vector, 'a, 'b] => bool
  5925   is(NS)Cont     :: ['a::real_normed_vector => 'b::real_normed_vector, 'a] => bool
  5926   deriv          :: ['a::real_normed_field => 'a, 'a, 'a] => bool
  5927   sgn            :: 'a::real_normed_vector => 'a
  5928   exp            :: 'a::{recpower,real_normed_field,banach} => 'a
  5929 
  5930 * Complex: Some complex-specific constants are now abbreviations for
  5931 overloaded ones: complex_of_real = of_real, cmod = norm, hcmod =
  5932 hnorm.  Other constants have been entirely removed in favor of the
  5933 polymorphic versions (INCOMPATIBILITY):
  5934 
  5935   approx        <-- capprox
  5936   HFinite       <-- CFinite
  5937   HInfinite     <-- CInfinite
  5938   Infinitesimal <-- CInfinitesimal
  5939   monad         <-- cmonad
  5940   galaxy        <-- cgalaxy
  5941   (NS)LIM       <-- (NS)CLIM, (NS)CRLIM
  5942   is(NS)Cont    <-- is(NS)Contc, is(NS)contCR
  5943   (ns)deriv     <-- (ns)cderiv
  5944 
  5945 
  5946 *** HOL-Algebra ***
  5947 
  5948 * Formalisation of ideals and the quotient construction over rings.
  5949 
  5950 * Order and lattice theory no longer based on records.
  5951 INCOMPATIBILITY.
  5952 
  5953 * Renamed lemmas least_carrier -> least_closed and greatest_carrier ->
  5954 greatest_closed.  INCOMPATIBILITY.
  5955 
  5956 * Method algebra is now set up via an attribute.  For examples see
  5957 Ring.thy.  INCOMPATIBILITY: the method is now weaker on combinations
  5958 of algebraic structures.
  5959 
  5960 * Renamed theory CRing to Ring.
  5961 
  5962 
  5963 *** HOL-Nominal ***
  5964 
  5965 * Substantial, yet incomplete support for nominal datatypes (binding
  5966 structures) based on HOL-Nominal logic.  See src/HOL/Nominal and
  5967 src/HOL/Nominal/Examples.  Prospective users should consult
  5968 http://isabelle.in.tum.de/nominal/
  5969 
  5970 
  5971 *** ML ***
  5972 
  5973 * ML basics: just one true type int, which coincides with IntInf.int
  5974 (even on SML/NJ).
  5975 
  5976 * ML within Isar: antiquotations allow to embed statically-checked
  5977 formal entities in the source, referring to the context available at
  5978 compile-time.  For example:
  5979 
  5980 ML {* @{sort "{zero,one}"} *}
  5981 ML {* @{typ "'a => 'b"} *}
  5982 ML {* @{term "%x. x"} *}
  5983 ML {* @{prop "x == y"} *}
  5984 ML {* @{ctyp "'a => 'b"} *}
  5985 ML {* @{cterm "%x. x"} *}
  5986 ML {* @{cprop "x == y"} *}
  5987 ML {* @{thm asm_rl} *}
  5988 ML {* @{thms asm_rl} *}
  5989 ML {* @{type_name c} *}
  5990 ML {* @{type_syntax c} *}
  5991 ML {* @{const_name c} *}
  5992 ML {* @{const_syntax c} *}
  5993 ML {* @{context} *}
  5994 ML {* @{theory} *}
  5995 ML {* @{theory Pure} *}
  5996 ML {* @{theory_ref} *}
  5997 ML {* @{theory_ref Pure} *}
  5998 ML {* @{simpset} *}
  5999 ML {* @{claset} *}
  6000 ML {* @{clasimpset} *}
  6001 
  6002 The same works for sources being ``used'' within an Isar context.
  6003 
  6004 * ML in Isar: improved error reporting; extra verbosity with
  6005 ML_Context.trace enabled.
  6006 
  6007 * Pure/General/table.ML: the join operations now works via exceptions
  6008 DUP/SAME instead of type option. This is simpler in simple cases, and
  6009 admits slightly more efficient complex applications.
  6010 
  6011 * Pure: 'advanced' translation functions (parse_translation etc.) now
  6012 use Context.generic instead of just theory.
  6013 
  6014 * Pure: datatype Context.generic joins theory/Proof.context and
  6015 provides some facilities for code that works in either kind of
  6016 context, notably GenericDataFun for uniform theory and proof data.
  6017 
  6018 * Pure: simplified internal attribute type, which is now always
  6019 Context.generic * thm -> Context.generic * thm. Global (theory) vs.
  6020 local (Proof.context) attributes have been discontinued, while
  6021 minimizing code duplication. Thm.rule_attribute and
  6022 Thm.declaration_attribute build canonical attributes; see also structure
  6023 Context for further operations on Context.generic, notably
  6024 GenericDataFun. INCOMPATIBILITY, need to adapt attribute type
  6025 declarations and definitions.
  6026 
  6027 * Context data interfaces (Theory/Proof/GenericDataFun): removed
  6028 name/print, uninitialized data defaults to ad-hoc copy of empty value,
  6029 init only required for impure data. INCOMPATIBILITY: empty really need
  6030 to be empty (no dependencies on theory content!)
  6031 
  6032 * Pure/kernel: consts certification ignores sort constraints given in
  6033 signature declarations. (This information is not relevant to the
  6034 logic, but only for type inference.) SIGNIFICANT INTERNAL CHANGE,
  6035 potential INCOMPATIBILITY.
  6036 
  6037 * Pure: axiomatic type classes are now purely definitional, with
  6038 explicit proofs of class axioms and super class relations performed
  6039 internally. See Pure/axclass.ML for the main internal interfaces --
  6040 notably AxClass.define_class supercedes AxClass.add_axclass, and
  6041 AxClass.axiomatize_class/classrel/arity supersede
  6042 Sign.add_classes/classrel/arities.
  6043 
  6044 * Pure/Isar: Args/Attrib parsers operate on Context.generic --
  6045 global/local versions on theory vs. Proof.context have been
  6046 discontinued; Attrib.syntax and Method.syntax have been adapted
  6047 accordingly.  INCOMPATIBILITY, need to adapt parser expressions for
  6048 attributes, methods, etc.
  6049 
  6050 * Pure: several functions of signature "... -> theory -> theory * ..."
  6051 have been reoriented to "... -> theory -> ... * theory" in order to
  6052 allow natural usage in combination with the ||>, ||>>, |-> and
  6053 fold_map combinators.
  6054 
  6055 * Pure: official theorem names (closed derivations) and additional
  6056 comments (tags) are now strictly separate.  Name hints -- which are
  6057 maintained as tags -- may be attached any time without affecting the
  6058 derivation.
  6059 
  6060 * Pure: primitive rule lift_rule now takes goal cterm instead of an
  6061 actual goal state (thm).  Use Thm.lift_rule (Thm.cprem_of st i) to
  6062 achieve the old behaviour.
  6063 
  6064 * Pure: the "Goal" constant is now called "prop", supporting a
  6065 slightly more general idea of ``protecting'' meta-level rule
  6066 statements.
  6067 
  6068 * Pure: Logic.(un)varify only works in a global context, which is now
  6069 enforced instead of silently assumed.  INCOMPATIBILITY, may use
  6070 Logic.legacy_(un)varify as temporary workaround.
  6071 
  6072 * Pure: structure Name provides scalable operations for generating
  6073 internal variable names, notably Name.variants etc.  This replaces
  6074 some popular functions from term.ML:
  6075 
  6076   Term.variant		->  Name.variant
  6077   Term.variantlist	->  Name.variant_list
  6078   Term.invent_names	->  Name.invent_list
  6079 
  6080 Note that low-level renaming rarely occurs in new code -- operations
  6081 from structure Variable are used instead (see below).
  6082 
  6083 * Pure: structure Variable provides fundamental operations for proper
  6084 treatment of fixed/schematic variables in a context.  For example,
  6085 Variable.import introduces fixes for schematics of given facts and
  6086 Variable.export reverses the effect (up to renaming) -- this replaces
  6087 various freeze_thaw operations.
  6088 
  6089 * Pure: structure Goal provides simple interfaces for
  6090 init/conclude/finish and tactical prove operations (replacing former
  6091 Tactic.prove).  Goal.prove is the canonical way to prove results
  6092 within a given context; Goal.prove_global is a degraded version for
  6093 theory level goals, including a global Drule.standard.  Note that
  6094 OldGoals.prove_goalw_cterm has long been obsolete, since it is
  6095 ill-behaved in a local proof context (e.g. with local fixes/assumes or
  6096 in a locale context).
  6097 
  6098 * Pure/Syntax: generic interfaces for parsing (Syntax.parse_term etc.)
  6099 and type checking (Syntax.check_term etc.), with common combinations
  6100 (Syntax.read_term etc.). These supersede former Sign.read_term etc.
  6101 which are considered legacy and await removal.
  6102 
  6103 * Pure/Syntax: generic interfaces for type unchecking
  6104 (Syntax.uncheck_terms etc.) and unparsing (Syntax.unparse_term etc.),
  6105 with common combinations (Syntax.pretty_term, Syntax.string_of_term
  6106 etc.).  Former Sign.pretty_term, Sign.string_of_term etc. are still
  6107 available for convenience, but refer to the very same operations using
  6108 a mere theory instead of a full context.
  6109 
  6110 * Isar: simplified treatment of user-level errors, using exception
  6111 ERROR of string uniformly.  Function error now merely raises ERROR,
  6112 without any side effect on output channels.  The Isar toplevel takes
  6113 care of proper display of ERROR exceptions.  ML code may use plain
  6114 handle/can/try; cat_error may be used to concatenate errors like this:
  6115 
  6116   ... handle ERROR msg => cat_error msg "..."
  6117 
  6118 Toplevel ML code (run directly or through the Isar toplevel) may be
  6119 embedded into the Isar toplevel with exception display/debug like
  6120 this:
  6121 
  6122   Isar.toplevel (fn () => ...)
  6123 
  6124 INCOMPATIBILITY, removed special transform_error facilities, removed
  6125 obsolete variants of user-level exceptions (ERROR_MESSAGE,
  6126 Context.PROOF, ProofContext.CONTEXT, Proof.STATE, ProofHistory.FAIL)
  6127 -- use plain ERROR instead.
  6128 
  6129 * Isar: theory setup now has type (theory -> theory), instead of a
  6130 list.  INCOMPATIBILITY, may use #> to compose setup functions.
  6131 
  6132 * Isar: ML toplevel pretty printer for type Proof.context, subject to
  6133 ProofContext.debug/verbose flags.
  6134 
  6135 * Isar: Toplevel.theory_to_proof admits transactions that modify the
  6136 theory before entering a proof state.  Transactions now always see a
  6137 quasi-functional intermediate checkpoint, both in interactive and
  6138 batch mode.
  6139 
  6140 * Isar: simplified interfaces for outer syntax.  Renamed
  6141 OuterSyntax.add_keywords to OuterSyntax.keywords.  Removed
  6142 OuterSyntax.add_parsers -- this functionality is now included in
  6143 OuterSyntax.command etc.  INCOMPATIBILITY.
  6144 
  6145 * Simplifier: the simpset of a running simplification process now
  6146 contains a proof context (cf. Simplifier.the_context), which is the
  6147 very context that the initial simpset has been retrieved from (by
  6148 simpset_of/local_simpset_of).  Consequently, all plug-in components
  6149 (solver, looper etc.) may depend on arbitrary proof data.
  6150 
  6151 * Simplifier.inherit_context inherits the proof context (plus the
  6152 local bounds) of the current simplification process; any simproc
  6153 etc. that calls the Simplifier recursively should do this!  Removed
  6154 former Simplifier.inherit_bounds, which is already included here --
  6155 INCOMPATIBILITY.  Tools based on low-level rewriting may even have to
  6156 specify an explicit context using Simplifier.context/theory_context.
  6157 
  6158 * Simplifier/Classical Reasoner: more abstract interfaces
  6159 change_simpset/claset for modifying the simpset/claset reference of a
  6160 theory; raw versions simpset/claset_ref etc. have been discontinued --
  6161 INCOMPATIBILITY.
  6162 
  6163 * Provers: more generic wrt. syntax of object-logics, avoid hardwired
  6164 "Trueprop" etc.
  6165 
  6166 
  6167 *** System ***
  6168 
  6169 * settings: the default heap location within ISABELLE_HOME_USER now
  6170 includes ISABELLE_IDENTIFIER.  This simplifies use of multiple
  6171 Isabelle installations.
  6172 
  6173 * isabelle-process: option -S (secure mode) disables some critical
  6174 operations, notably runtime compilation and evaluation of ML source
  6175 code.
  6176 
  6177 * Basic Isabelle mode for jEdit, see Isabelle/lib/jedit/.
  6178 
  6179 * Support for parallel execution, using native multicore support of
  6180 Poly/ML 5.1.  The theory loader exploits parallelism when processing
  6181 independent theories, according to the given theory header
  6182 specifications. The maximum number of worker threads is specified via
  6183 usedir option -M or the "max-threads" setting in Proof General. A
  6184 speedup factor of 1.5--3.5 can be expected on a 4-core machine, and up
  6185 to 6 on a 8-core machine.  User-code needs to observe certain
  6186 guidelines for thread-safe programming, see appendix A in the Isar
  6187 Implementation manual.
  6188 
  6189 
  6190 
  6191 New in Isabelle2005 (October 2005)
  6192 ----------------------------------
  6193 
  6194 *** General ***
  6195 
  6196 * Theory headers: the new header syntax for Isar theories is
  6197 
  6198   theory <name>
  6199   imports <theory1> ... <theoryN>
  6200   uses <file1> ... <fileM>
  6201   begin
  6202 
  6203 where the 'uses' part is optional.  The previous syntax
  6204 
  6205   theory <name> = <theory1> + ... + <theoryN>:
  6206 
  6207 will disappear in the next release.  Use isatool fixheaders to convert
  6208 existing theory files.  Note that there is no change in ancient
  6209 non-Isar theories now, but these will disappear soon.
  6210 
  6211 * Theory loader: parent theories can now also be referred to via
  6212 relative and absolute paths.
  6213 
  6214 * Command 'find_theorems' searches for a list of criteria instead of a
  6215 list of constants. Known criteria are: intro, elim, dest, name:string,
  6216 simp:term, and any term. Criteria can be preceded by '-' to select
  6217 theorems that do not match. Intro, elim, dest select theorems that
  6218 match the current goal, name:s selects theorems whose fully qualified
  6219 name contain s, and simp:term selects all simplification rules whose
  6220 lhs match term.  Any other term is interpreted as pattern and selects
  6221 all theorems matching the pattern. Available in ProofGeneral under
  6222 'ProofGeneral -> Find Theorems' or C-c C-f.  Example:
  6223 
  6224   C-c C-f (100) "(_::nat) + _ + _" intro -name: "HOL."
  6225 
  6226 prints the last 100 theorems matching the pattern "(_::nat) + _ + _",
  6227 matching the current goal as introduction rule and not having "HOL."
  6228 in their name (i.e. not being defined in theory HOL).
  6229 
  6230 * Command 'thms_containing' has been discontinued in favour of
  6231 'find_theorems'; INCOMPATIBILITY.
  6232 
  6233 * Communication with Proof General is now 8bit clean, which means that
  6234 Unicode text in UTF-8 encoding may be used within theory texts (both
  6235 formal and informal parts).  Cf. option -U of the Isabelle Proof
  6236 General interface.  Here are some simple examples (cf. src/HOL/ex):
  6237 
  6238   http://isabelle.in.tum.de/library/HOL/ex/Hebrew.html
  6239   http://isabelle.in.tum.de/library/HOL/ex/Chinese.html
  6240