NEWS
author wenzelm
Fri Jul 06 16:20:54 2012 +0200 (2012-07-06)
changeset 48205 09c2a3d9aa22
parent 48171 28a6d67c93f0
child 48206 937b53a339f0
permissions -rw-r--r--
discontinued obsolete attribute "COMP";
     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 * Discontinued obsolete attribute "COMP".  Potential INCOMPATIBILITY,
    19 use regular rule composition via "OF" / "THEN", or explicit proof
    20 structure instead.  Note that Isabelle/ML provides a variety of
    21 operators like COMP, INCR_COMP, COMP_INCR, which need to be applied
    22 with some care where this is really required.
    23 
    24 
    25 *** Document preparation ***
    26 
    27 * Default for \<euro> is now based on eurosym package, instead of
    28 slightly exotic babel/greek.
    29 
    30 
    31 *** System ***
    32 
    33 * Discontinued support for Poly/ML 5.2.1, which was the last version
    34 without exception positions and advanced ML compiler/toplevel
    35 configuration.
    36 
    37 
    38 *** HOL ***
    39 
    40 * Simproc for rewriting set comprehensions into pointfree expressions
    41 
    42 * Quickcheck:
    43 
    44   - added an optimisation for equality premises.
    45     It is switched on by default, and can be switched off by setting
    46     the configuration quickcheck_optimise_equality to false.    
    47 
    48 * The SMT solver Z3 has now by default a restricted set of directly
    49 supported features. For the full set of features (div/mod, nonlinear
    50 arithmetic, datatypes/records) with potential proof reconstruction
    51 failures, enable the configuration option "z3_with_extensions".
    52 Minor INCOMPATIBILITY.
    53 
    54 * Sledgehammer:
    55 
    56   - Rationalized type encodings ("type_enc" option).
    57 
    58 
    59 
    60 New in Isabelle2012 (May 2012)
    61 ------------------------------
    62 
    63 *** General ***
    64 
    65 * Prover IDE (PIDE) improvements:
    66 
    67   - more robust Sledgehammer integration (as before the sledgehammer
    68     command-line needs to be typed into the source buffer)
    69   - markup for bound variables
    70   - markup for types of term variables (displayed as tooltips)
    71   - support for user-defined Isar commands within the running session
    72   - improved support for Unicode outside original 16bit range
    73     e.g. glyph for \<A> (thanks to jEdit 4.5.1)
    74 
    75 * Forward declaration of outer syntax keywords within the theory
    76 header -- minor INCOMPATIBILITY for user-defined commands.  Allow new
    77 commands to be used in the same theory where defined.
    78 
    79 * Auxiliary contexts indicate block structure for specifications with
    80 additional parameters and assumptions.  Such unnamed contexts may be
    81 nested within other targets, like 'theory', 'locale', 'class',
    82 'instantiation' etc.  Results from the local context are generalized
    83 accordingly and applied to the enclosing target context.  Example:
    84 
    85   context
    86     fixes x y z :: 'a
    87     assumes xy: "x = y" and yz: "y = z"
    88   begin
    89 
    90   lemma my_trans: "x = z" using xy yz by simp
    91 
    92   end
    93 
    94   thm my_trans
    95 
    96 The most basic application is to factor-out context elements of
    97 several fixes/assumes/shows theorem statements, e.g. see
    98 ~~/src/HOL/Isar_Examples/Group_Context.thy
    99 
   100 Any other local theory specification element works within the "context
   101 ... begin ... end" block as well.
   102 
   103 * Bundled declarations associate attributed fact expressions with a
   104 given name in the context.  These may be later included in other
   105 contexts.  This allows to manage context extensions casually, without
   106 the logical dependencies of locales and locale interpretation.  See
   107 commands 'bundle', 'include', 'including' etc. in the isar-ref manual.
   108 
   109 * Commands 'lemmas' and 'theorems' allow local variables using 'for'
   110 declaration, and results are standardized before being stored.  Thus
   111 old-style "standard" after instantiation or composition of facts
   112 becomes obsolete.  Minor INCOMPATIBILITY, due to potential change of
   113 indices of schematic variables.
   114 
   115 * Rule attributes in local theory declarations (e.g. locale or class)
   116 are now statically evaluated: the resulting theorem is stored instead
   117 of the original expression.  INCOMPATIBILITY in rare situations, where
   118 the historic accident of dynamic re-evaluation in interpretations
   119 etc. was exploited.
   120 
   121 * New tutorial "Programming and Proving in Isabelle/HOL"
   122 ("prog-prove").  It completely supersedes "A Tutorial Introduction to
   123 Structured Isar Proofs" ("isar-overview"), which has been removed.  It
   124 also supersedes "Isabelle/HOL, A Proof Assistant for Higher-Order
   125 Logic" as the recommended beginners tutorial, but does not cover all
   126 of the material of that old tutorial.
   127 
   128 * Updated and extended reference manuals: "isar-ref",
   129 "implementation", "system"; reduced remaining material in old "ref"
   130 manual.
   131 
   132 
   133 *** Pure ***
   134 
   135 * Command 'definition' no longer exports the foundational "raw_def"
   136 into the user context.  Minor INCOMPATIBILITY, may use the regular
   137 "def" result with attribute "abs_def" to imitate the old version.
   138 
   139 * Attribute "abs_def" turns an equation of the form "f x y == t" into
   140 "f == %x y. t", which ensures that "simp" or "unfold" steps always
   141 expand it.  This also works for object-logic equality.  (Formerly
   142 undocumented feature.)
   143 
   144 * Sort constraints are now propagated in simultaneous statements, just
   145 like type constraints.  INCOMPATIBILITY in rare situations, where
   146 distinct sorts used to be assigned accidentally.  For example:
   147 
   148   lemma "P (x::'a::foo)" and "Q (y::'a::bar)"  -- "now illegal"
   149 
   150   lemma "P (x::'a)" and "Q (y::'a::bar)"
   151     -- "now uniform 'a::bar instead of default sort for first occurrence (!)"
   152 
   153 * Rule composition via attribute "OF" (or ML functions OF/MRS) is more
   154 tolerant against multiple unifiers, as long as the final result is
   155 unique.  (As before, rules are composed in canonical right-to-left
   156 order to accommodate newly introduced premises.)
   157 
   158 * Renamed some inner syntax categories:
   159 
   160     num ~> num_token
   161     xnum ~> xnum_token
   162     xstr ~> str_token
   163 
   164 Minor INCOMPATIBILITY.  Note that in practice "num_const" or
   165 "num_position" etc. are mainly used instead (which also include
   166 position information via constraints).
   167 
   168 * Simplified configuration options for syntax ambiguity: see
   169 "syntax_ambiguity_warning" and "syntax_ambiguity_limit" in isar-ref
   170 manual.  Minor INCOMPATIBILITY.
   171 
   172 * Discontinued configuration option "syntax_positions": atomic terms
   173 in parse trees are always annotated by position constraints.
   174 
   175 * Old code generator for SML and its commands 'code_module',
   176 'code_library', 'consts_code', 'types_code' have been discontinued.
   177 Use commands of the generic code generator instead.  INCOMPATIBILITY.
   178 
   179 * Redundant attribute "code_inline" has been discontinued. Use
   180 "code_unfold" instead.  INCOMPATIBILITY.
   181 
   182 * Dropped attribute "code_unfold_post" in favor of the its dual
   183 "code_abbrev", which yields a common pattern in definitions like
   184 
   185   definition [code_abbrev]: "f = t"
   186 
   187 INCOMPATIBILITY.
   188 
   189 * Obsolete 'types' command has been discontinued.  Use 'type_synonym'
   190 instead.  INCOMPATIBILITY.
   191 
   192 * Discontinued old "prems" fact, which used to refer to the accidental
   193 collection of foundational premises in the context (already marked as
   194 legacy since Isabelle2011).
   195 
   196 
   197 *** HOL ***
   198 
   199 * Type 'a set is now a proper type constructor (just as before
   200 Isabelle2008).  Definitions mem_def and Collect_def have disappeared.
   201 Non-trivial INCOMPATIBILITY.  For developments keeping predicates and
   202 sets separate, it is often sufficient to rephrase some set S that has
   203 been accidentally used as predicates by "%x. x : S", and some
   204 predicate P that has been accidentally used as set by "{x. P x}".
   205 Corresponding proofs in a first step should be pruned from any
   206 tinkering with former theorems mem_def and Collect_def as far as
   207 possible.
   208 
   209 For developments which deliberately mix predicates and sets, a
   210 planning step is necessary to determine what should become a predicate
   211 and what a set.  It can be helpful to carry out that step in
   212 Isabelle2011-1 before jumping right into the current release.
   213 
   214 * Code generation by default implements sets as container type rather
   215 than predicates.  INCOMPATIBILITY.
   216 
   217 * New type synonym 'a rel = ('a * 'a) set
   218 
   219 * The representation of numerals has changed.  Datatype "num"
   220 represents strictly positive binary numerals, along with functions
   221 "numeral :: num => 'a" and "neg_numeral :: num => 'a" to represent
   222 positive and negated numeric literals, respectively.  See also
   223 definitions in ~~/src/HOL/Num.thy.  Potential INCOMPATIBILITY, some
   224 user theories may require adaptations as follows:
   225 
   226   - Theorems with number_ring or number_semiring constraints: These
   227     classes are gone; use comm_ring_1 or comm_semiring_1 instead.
   228 
   229   - Theories defining numeric types: Remove number, number_semiring,
   230     and number_ring instances. Defer all theorems about numerals until
   231     after classes one and semigroup_add have been instantiated.
   232 
   233   - Numeral-only simp rules: Replace each rule having a "number_of v"
   234     pattern with two copies, one for numeral and one for neg_numeral.
   235 
   236   - Theorems about subclasses of semiring_1 or ring_1: These classes
   237     automatically support numerals now, so more simp rules and
   238     simprocs may now apply within the proof.
   239 
   240   - Definitions and theorems using old constructors Pls/Min/Bit0/Bit1:
   241     Redefine using other integer operations.
   242 
   243 * Transfer: New package intended to generalize the existing
   244 "descending" method and related theorem attributes from the Quotient
   245 package.  (Not all functionality is implemented yet, but future
   246 development will focus on Transfer as an eventual replacement for the
   247 corresponding parts of the Quotient package.)
   248 
   249   - transfer_rule attribute: Maintains a collection of transfer rules,
   250     which relate constants at two different types. Transfer rules may
   251     relate different type instances of the same polymorphic constant,
   252     or they may relate an operation on a raw type to a corresponding
   253     operation on an abstract type (quotient or subtype). For example:
   254 
   255     ((A ===> B) ===> list_all2 A ===> list_all2 B) map map
   256     (cr_int ===> cr_int ===> cr_int) (%(x,y) (u,v). (x+u, y+v)) plus_int
   257 
   258   - transfer method: Replaces a subgoal on abstract types with an
   259     equivalent subgoal on the corresponding raw types. Constants are
   260     replaced with corresponding ones according to the transfer rules.
   261     Goals are generalized over all free variables by default; this is
   262     necessary for variables whose types change, but can be overridden
   263     for specific variables with e.g. "transfer fixing: x y z".  The
   264     variant transfer' method allows replacing a subgoal with one that
   265     is logically stronger (rather than equivalent).
   266 
   267   - relator_eq attribute: Collects identity laws for relators of
   268     various type constructors, e.g. "list_all2 (op =) = (op =)".  The
   269     transfer method uses these lemmas to infer transfer rules for
   270     non-polymorphic constants on the fly.
   271 
   272   - transfer_prover method: Assists with proving a transfer rule for a
   273     new constant, provided the constant is defined in terms of other
   274     constants that already have transfer rules. It should be applied
   275     after unfolding the constant definitions.
   276 
   277   - HOL/ex/Transfer_Int_Nat.thy: Example theory demonstrating transfer
   278     from type nat to type int.
   279 
   280 * Lifting: New package intended to generalize the quotient_definition
   281 facility of the Quotient package; designed to work with Transfer.
   282 
   283   - lift_definition command: Defines operations on an abstract type in
   284     terms of a corresponding operation on a representation
   285     type.  Example syntax:
   286 
   287     lift_definition dlist_insert :: "'a => 'a dlist => 'a dlist"
   288       is List.insert
   289 
   290     Users must discharge a respectfulness proof obligation when each
   291     constant is defined. (For a type copy, i.e. a typedef with UNIV,
   292     the proof is discharged automatically.) The obligation is
   293     presented in a user-friendly, readable form; a respectfulness
   294     theorem in the standard format and a transfer rule are generated
   295     by the package.
   296 
   297   - Integration with code_abstype: For typedefs (e.g. subtypes
   298     corresponding to a datatype invariant, such as dlist),
   299     lift_definition generates a code certificate theorem and sets up
   300     code generation for each constant.
   301 
   302   - setup_lifting command: Sets up the Lifting package to work with a
   303     user-defined type. The user must provide either a quotient theorem
   304     or a type_definition theorem.  The package configures transfer
   305     rules for equality and quantifiers on the type, and sets up the
   306     lift_definition command to work with the type.
   307 
   308   - Usage examples: See Quotient_Examples/Lift_DList.thy,
   309     Quotient_Examples/Lift_RBT.thy, Quotient_Examples/Lift_FSet.thy,
   310     Word/Word.thy and Library/Float.thy.
   311 
   312 * Quotient package:
   313 
   314   - The 'quotient_type' command now supports a 'morphisms' option with
   315     rep and abs functions, similar to typedef.
   316 
   317   - 'quotient_type' sets up new types to work with the Lifting and
   318     Transfer packages, as with 'setup_lifting'.
   319 
   320   - The 'quotient_definition' command now requires the user to prove a
   321     respectfulness property at the point where the constant is
   322     defined, similar to lift_definition; INCOMPATIBILITY.
   323 
   324   - Renamed predicate 'Quotient' to 'Quotient3', and renamed theorems
   325     accordingly, INCOMPATIBILITY.
   326 
   327 * New diagnostic command 'find_unused_assms' to find potentially
   328 superfluous assumptions in theorems using Quickcheck.
   329 
   330 * Quickcheck:
   331 
   332   - Quickcheck returns variable assignments as counterexamples, which
   333     allows to reveal the underspecification of functions under test.
   334     For example, refuting "hd xs = x", it presents the variable
   335     assignment xs = [] and x = a1 as a counterexample, assuming that
   336     any property is false whenever "hd []" occurs in it.
   337 
   338     These counterexample are marked as potentially spurious, as
   339     Quickcheck also returns "xs = []" as a counterexample to the
   340     obvious theorem "hd xs = hd xs".
   341 
   342     After finding a potentially spurious counterexample, Quickcheck
   343     continues searching for genuine ones.
   344 
   345     By default, Quickcheck shows potentially spurious and genuine
   346     counterexamples. The option "genuine_only" sets quickcheck to only
   347     show genuine counterexamples.
   348 
   349   - The command 'quickcheck_generator' creates random and exhaustive
   350     value generators for a given type and operations.
   351 
   352     It generates values by using the operations as if they were
   353     constructors of that type.
   354 
   355   - Support for multisets.
   356 
   357   - Added "use_subtype" options.
   358 
   359   - Added "quickcheck_locale" configuration to specify how to process
   360     conjectures in a locale context.
   361 
   362 * Nitpick: Fixed infinite loop caused by the 'peephole_optim' option
   363 and affecting 'rat' and 'real'.
   364 
   365 * Sledgehammer:
   366   - Integrated more tightly with SPASS, as described in the ITP 2012
   367     paper "More SPASS with Isabelle".
   368   - Made it try "smt" as a fallback if "metis" fails or times out.
   369   - Added support for the following provers: Alt-Ergo (via Why3 and
   370     TFF1), iProver, iProver-Eq.
   371   - Replaced remote E-SInE with remote Satallax in the default setup.
   372   - Sped up the minimizer.
   373   - Added "lam_trans", "uncurry_aliases", and "minimize" options.
   374   - Renamed "slicing" ("no_slicing") option to "slice" ("dont_slice").
   375   - Renamed "sound" option to "strict".
   376 
   377 * Metis: Added possibility to specify lambda translations scheme as a
   378 parenthesized argument (e.g., "by (metis (lifting) ...)").
   379 
   380 * SMT: Renamed "smt_fixed" option to "smt_read_only_certificates".
   381 
   382 * Command 'try0': Renamed from 'try_methods'. INCOMPATIBILITY.
   383 
   384 * New "case_product" attribute to generate a case rule doing multiple
   385 case distinctions at the same time.  E.g.
   386 
   387   list.exhaust [case_product nat.exhaust]
   388 
   389 produces a rule which can be used to perform case distinction on both
   390 a list and a nat.
   391 
   392 * New "eventually_elim" method as a generalized variant of the
   393 eventually_elim* rules.  Supports structured proofs.
   394 
   395 * Typedef with implicit set definition is considered legacy.  Use
   396 "typedef (open)" form instead, which will eventually become the
   397 default.
   398 
   399 * Record: code generation can be switched off manually with
   400 
   401   declare [[record_coden = false]]  -- "default true"
   402 
   403 * Datatype: type parameters allow explicit sort constraints.
   404 
   405 * Concrete syntax for case expressions includes constraints for source
   406 positions, and thus produces Prover IDE markup for its bindings.
   407 INCOMPATIBILITY for old-style syntax translations that augment the
   408 pattern notation; e.g. see src/HOL/HOLCF/One.thy for translations of
   409 one_case.
   410 
   411 * Clarified attribute "mono_set": pure declaration without modifying
   412 the result of the fact expression.
   413 
   414 * More default pred/set conversions on a couple of relation operations
   415 and predicates.  Added powers of predicate relations.  Consolidation
   416 of some relation theorems:
   417 
   418   converse_def ~> converse_unfold
   419   rel_comp_def ~> relcomp_unfold
   420   symp_def ~> (modified, use symp_def and sym_def instead)
   421   transp_def ~> transp_trans
   422   Domain_def ~> Domain_unfold
   423   Range_def ~> Domain_converse [symmetric]
   424 
   425 Generalized theorems INF_INT_eq, INF_INT_eq2, SUP_UN_eq, SUP_UN_eq2.
   426 
   427 See theory "Relation" for examples for making use of pred/set
   428 conversions by means of attributes "to_set" and "to_pred".
   429 
   430 INCOMPATIBILITY.
   431 
   432 * Renamed facts about the power operation on relations, i.e., relpow
   433 to match the constant's name:
   434 
   435   rel_pow_1 ~> relpow_1
   436   rel_pow_0_I ~> relpow_0_I
   437   rel_pow_Suc_I ~> relpow_Suc_I
   438   rel_pow_Suc_I2 ~> relpow_Suc_I2
   439   rel_pow_0_E ~> relpow_0_E
   440   rel_pow_Suc_E ~> relpow_Suc_E
   441   rel_pow_E ~> relpow_E
   442   rel_pow_Suc_D2 ~> relpow_Suc_D2
   443   rel_pow_Suc_E2 ~> relpow_Suc_E2
   444   rel_pow_Suc_D2' ~> relpow_Suc_D2'
   445   rel_pow_E2 ~> relpow_E2
   446   rel_pow_add ~> relpow_add
   447   rel_pow_commute ~> relpow
   448   rel_pow_empty ~> relpow_empty:
   449   rtrancl_imp_UN_rel_pow ~> rtrancl_imp_UN_relpow
   450   rel_pow_imp_rtrancl ~> relpow_imp_rtrancl
   451   rtrancl_is_UN_rel_pow ~> rtrancl_is_UN_relpow
   452   rtrancl_imp_rel_pow ~> rtrancl_imp_relpow
   453   rel_pow_fun_conv ~> relpow_fun_conv
   454   rel_pow_finite_bounded1 ~> relpow_finite_bounded1
   455   rel_pow_finite_bounded ~> relpow_finite_bounded
   456   rtrancl_finite_eq_rel_pow ~> rtrancl_finite_eq_relpow
   457   trancl_finite_eq_rel_pow ~> trancl_finite_eq_relpow
   458   single_valued_rel_pow ~> single_valued_relpow
   459 
   460 INCOMPATIBILITY.
   461 
   462 * Theory Relation: Consolidated constant name for relation composition
   463 and corresponding theorem names:
   464 
   465   - Renamed constant rel_comp to relcomp.
   466 
   467   - Dropped abbreviation pred_comp. Use relcompp instead.
   468 
   469   - Renamed theorems:
   470 
   471     rel_compI ~> relcompI
   472     rel_compEpair ~> relcompEpair
   473     rel_compE ~> relcompE
   474     pred_comp_rel_comp_eq ~> relcompp_relcomp_eq
   475     rel_comp_empty1 ~> relcomp_empty1
   476     rel_comp_mono ~> relcomp_mono
   477     rel_comp_subset_Sigma ~> relcomp_subset_Sigma
   478     rel_comp_distrib ~> relcomp_distrib
   479     rel_comp_distrib2 ~> relcomp_distrib2
   480     rel_comp_UNION_distrib ~> relcomp_UNION_distrib
   481     rel_comp_UNION_distrib2 ~> relcomp_UNION_distrib2
   482     single_valued_rel_comp ~> single_valued_relcomp
   483     rel_comp_def ~> relcomp_unfold
   484     converse_rel_comp ~> converse_relcomp
   485     pred_compI ~> relcomppI
   486     pred_compE ~> relcomppE
   487     pred_comp_bot1 ~> relcompp_bot1
   488     pred_comp_bot2 ~> relcompp_bot2
   489     transp_pred_comp_less_eq ~> transp_relcompp_less_eq
   490     pred_comp_mono ~> relcompp_mono
   491     pred_comp_distrib ~> relcompp_distrib
   492     pred_comp_distrib2 ~> relcompp_distrib2
   493     converse_pred_comp ~> converse_relcompp
   494 
   495     finite_rel_comp ~> finite_relcomp
   496 
   497     set_rel_comp ~> set_relcomp
   498 
   499 INCOMPATIBILITY.
   500 
   501 * Theory Divides: Discontinued redundant theorems about div and mod.
   502 INCOMPATIBILITY, use the corresponding generic theorems instead.
   503 
   504   DIVISION_BY_ZERO ~> div_by_0, mod_by_0
   505   zdiv_self ~> div_self
   506   zmod_self ~> mod_self
   507   zdiv_zero ~> div_0
   508   zmod_zero ~> mod_0
   509   zdiv_zmod_equality ~> div_mod_equality2
   510   zdiv_zmod_equality2 ~> div_mod_equality
   511   zmod_zdiv_trivial ~> mod_div_trivial
   512   zdiv_zminus_zminus ~> div_minus_minus
   513   zmod_zminus_zminus ~> mod_minus_minus
   514   zdiv_zminus2 ~> div_minus_right
   515   zmod_zminus2 ~> mod_minus_right
   516   zdiv_minus1_right ~> div_minus1_right
   517   zmod_minus1_right ~> mod_minus1_right
   518   zdvd_mult_div_cancel ~> dvd_mult_div_cancel
   519   zmod_zmult1_eq ~> mod_mult_right_eq
   520   zpower_zmod ~> power_mod
   521   zdvd_zmod ~> dvd_mod
   522   zdvd_zmod_imp_zdvd ~> dvd_mod_imp_dvd
   523   mod_mult_distrib ~> mult_mod_left
   524   mod_mult_distrib2 ~> mult_mod_right
   525 
   526 * Removed redundant theorems nat_mult_2 and nat_mult_2_right; use
   527 generic mult_2 and mult_2_right instead. INCOMPATIBILITY.
   528 
   529 * Finite_Set.fold now qualified.  INCOMPATIBILITY.
   530 
   531 * Consolidated theorem names concerning fold combinators:
   532 
   533   inf_INFI_fold_inf ~> inf_INF_fold_inf
   534   sup_SUPR_fold_sup ~> sup_SUP_fold_sup
   535   INFI_fold_inf ~> INF_fold_inf
   536   SUPR_fold_sup ~> SUP_fold_sup
   537   union_set ~> union_set_fold
   538   minus_set ~> minus_set_fold
   539   INFI_set_fold ~> INF_set_fold
   540   SUPR_set_fold ~> SUP_set_fold
   541   INF_code ~> INF_set_foldr
   542   SUP_code ~> SUP_set_foldr
   543   foldr.simps ~> foldr.simps (in point-free formulation)
   544   foldr_fold_rev ~> foldr_conv_fold
   545   foldl_fold ~> foldl_conv_fold
   546   foldr_foldr ~> foldr_conv_foldl
   547   foldl_foldr ~> foldl_conv_foldr
   548   fold_set_remdups ~> fold_set_fold_remdups
   549   fold_set ~> fold_set_fold
   550   fold1_set ~> fold1_set_fold
   551 
   552 INCOMPATIBILITY.
   553 
   554 * Dropped rarely useful theorems concerning fold combinators:
   555 foldl_apply, foldl_fun_comm, foldl_rev, fold_weak_invariant,
   556 rev_foldl_cons, fold_set_remdups, fold_set, fold_set1,
   557 concat_conv_foldl, foldl_weak_invariant, foldl_invariant,
   558 foldr_invariant, foldl_absorb0, foldl_foldr1_lemma, foldl_foldr1,
   559 listsum_conv_fold, listsum_foldl, sort_foldl_insort, foldl_assoc,
   560 foldr_conv_foldl, start_le_sum, elem_le_sum, sum_eq_0_conv.
   561 INCOMPATIBILITY.  For the common phrases "%xs. List.foldr plus xs 0"
   562 and "List.foldl plus 0", prefer "List.listsum".  Otherwise it can be
   563 useful to boil down "List.foldr" and "List.foldl" to "List.fold" by
   564 unfolding "foldr_conv_fold" and "foldl_conv_fold".
   565 
   566 * Dropped lemmas minus_set_foldr, union_set_foldr, union_coset_foldr,
   567 inter_coset_foldr, Inf_fin_set_foldr, Sup_fin_set_foldr,
   568 Min_fin_set_foldr, Max_fin_set_foldr, Inf_set_foldr, Sup_set_foldr,
   569 INF_set_foldr, SUP_set_foldr.  INCOMPATIBILITY.  Prefer corresponding
   570 lemmas over fold rather than foldr, or make use of lemmas
   571 fold_conv_foldr and fold_rev.
   572 
   573 * Congruence rules Option.map_cong and Option.bind_cong for recursion
   574 through option types.
   575 
   576 * "Transitive_Closure.ntrancl": bounded transitive closure on
   577 relations.
   578 
   579 * Constant "Set.not_member" now qualified.  INCOMPATIBILITY.
   580 
   581 * Theory Int: Discontinued many legacy theorems specific to type int.
   582 INCOMPATIBILITY, use the corresponding generic theorems instead.
   583 
   584   zminus_zminus ~> minus_minus
   585   zminus_0 ~> minus_zero
   586   zminus_zadd_distrib ~> minus_add_distrib
   587   zadd_commute ~> add_commute
   588   zadd_assoc ~> add_assoc
   589   zadd_left_commute ~> add_left_commute
   590   zadd_ac ~> add_ac
   591   zmult_ac ~> mult_ac
   592   zadd_0 ~> add_0_left
   593   zadd_0_right ~> add_0_right
   594   zadd_zminus_inverse2 ~> left_minus
   595   zmult_zminus ~> mult_minus_left
   596   zmult_commute ~> mult_commute
   597   zmult_assoc ~> mult_assoc
   598   zadd_zmult_distrib ~> left_distrib
   599   zadd_zmult_distrib2 ~> right_distrib
   600   zdiff_zmult_distrib ~> left_diff_distrib
   601   zdiff_zmult_distrib2 ~> right_diff_distrib
   602   zmult_1 ~> mult_1_left
   603   zmult_1_right ~> mult_1_right
   604   zle_refl ~> order_refl
   605   zle_trans ~> order_trans
   606   zle_antisym ~> order_antisym
   607   zle_linear ~> linorder_linear
   608   zless_linear ~> linorder_less_linear
   609   zadd_left_mono ~> add_left_mono
   610   zadd_strict_right_mono ~> add_strict_right_mono
   611   zadd_zless_mono ~> add_less_le_mono
   612   int_0_less_1 ~> zero_less_one
   613   int_0_neq_1 ~> zero_neq_one
   614   zless_le ~> less_le
   615   zpower_zadd_distrib ~> power_add
   616   zero_less_zpower_abs_iff ~> zero_less_power_abs_iff
   617   zero_le_zpower_abs ~> zero_le_power_abs
   618 
   619 * Theory Deriv: Renamed
   620 
   621   DERIV_nonneg_imp_nonincreasing ~> DERIV_nonneg_imp_nondecreasing
   622 
   623 * Theory Library/Multiset: Improved code generation of multisets.
   624 
   625 * Theory HOL/Library/Set_Algebras: Addition and multiplication on sets
   626 are expressed via type classes again. The special syntax
   627 \<oplus>/\<otimes> has been replaced by plain +/*. Removed constant
   628 setsum_set, which is now subsumed by Big_Operators.setsum.
   629 INCOMPATIBILITY.
   630 
   631 * Theory HOL/Library/Diagonalize has been removed. INCOMPATIBILITY,
   632 use theory HOL/Library/Nat_Bijection instead.
   633 
   634 * Theory HOL/Library/RBT_Impl: Backing implementation of red-black
   635 trees is now inside a type class context.  Names of affected
   636 operations and lemmas have been prefixed by rbt_.  INCOMPATIBILITY for
   637 theories working directly with raw red-black trees, adapt the names as
   638 follows:
   639 
   640   Operations:
   641   bulkload -> rbt_bulkload
   642   del_from_left -> rbt_del_from_left
   643   del_from_right -> rbt_del_from_right
   644   del -> rbt_del
   645   delete -> rbt_delete
   646   ins -> rbt_ins
   647   insert -> rbt_insert
   648   insertw -> rbt_insert_with
   649   insert_with_key -> rbt_insert_with_key
   650   map_entry -> rbt_map_entry
   651   lookup -> rbt_lookup
   652   sorted -> rbt_sorted
   653   tree_greater -> rbt_greater
   654   tree_less -> rbt_less
   655   tree_less_symbol -> rbt_less_symbol
   656   union -> rbt_union
   657   union_with -> rbt_union_with
   658   union_with_key -> rbt_union_with_key
   659 
   660   Lemmas:
   661   balance_left_sorted -> balance_left_rbt_sorted
   662   balance_left_tree_greater -> balance_left_rbt_greater
   663   balance_left_tree_less -> balance_left_rbt_less
   664   balance_right_sorted -> balance_right_rbt_sorted
   665   balance_right_tree_greater -> balance_right_rbt_greater
   666   balance_right_tree_less -> balance_right_rbt_less
   667   balance_sorted -> balance_rbt_sorted
   668   balance_tree_greater -> balance_rbt_greater
   669   balance_tree_less -> balance_rbt_less
   670   bulkload_is_rbt -> rbt_bulkload_is_rbt
   671   combine_sorted -> combine_rbt_sorted
   672   combine_tree_greater -> combine_rbt_greater
   673   combine_tree_less -> combine_rbt_less
   674   delete_in_tree -> rbt_delete_in_tree
   675   delete_is_rbt -> rbt_delete_is_rbt
   676   del_from_left_tree_greater -> rbt_del_from_left_rbt_greater
   677   del_from_left_tree_less -> rbt_del_from_left_rbt_less
   678   del_from_right_tree_greater -> rbt_del_from_right_rbt_greater
   679   del_from_right_tree_less -> rbt_del_from_right_rbt_less
   680   del_in_tree -> rbt_del_in_tree
   681   del_inv1_inv2 -> rbt_del_inv1_inv2
   682   del_sorted -> rbt_del_rbt_sorted
   683   del_tree_greater -> rbt_del_rbt_greater
   684   del_tree_less -> rbt_del_rbt_less
   685   dom_lookup_Branch -> dom_rbt_lookup_Branch
   686   entries_lookup -> entries_rbt_lookup
   687   finite_dom_lookup -> finite_dom_rbt_lookup
   688   insert_sorted -> rbt_insert_rbt_sorted
   689   insertw_is_rbt -> rbt_insertw_is_rbt
   690   insertwk_is_rbt -> rbt_insertwk_is_rbt
   691   insertwk_sorted -> rbt_insertwk_rbt_sorted
   692   insertw_sorted -> rbt_insertw_rbt_sorted
   693   ins_sorted -> ins_rbt_sorted
   694   ins_tree_greater -> ins_rbt_greater
   695   ins_tree_less -> ins_rbt_less
   696   is_rbt_sorted -> is_rbt_rbt_sorted
   697   lookup_balance -> rbt_lookup_balance
   698   lookup_bulkload -> rbt_lookup_rbt_bulkload
   699   lookup_delete -> rbt_lookup_rbt_delete
   700   lookup_Empty -> rbt_lookup_Empty
   701   lookup_from_in_tree -> rbt_lookup_from_in_tree
   702   lookup_in_tree -> rbt_lookup_in_tree
   703   lookup_ins -> rbt_lookup_ins
   704   lookup_insert -> rbt_lookup_rbt_insert
   705   lookup_insertw -> rbt_lookup_rbt_insertw
   706   lookup_insertwk -> rbt_lookup_rbt_insertwk
   707   lookup_keys -> rbt_lookup_keys
   708   lookup_map -> rbt_lookup_map
   709   lookup_map_entry -> rbt_lookup_rbt_map_entry
   710   lookup_tree_greater -> rbt_lookup_rbt_greater
   711   lookup_tree_less -> rbt_lookup_rbt_less
   712   lookup_union -> rbt_lookup_rbt_union
   713   map_entry_color_of -> rbt_map_entry_color_of
   714   map_entry_inv1 -> rbt_map_entry_inv1
   715   map_entry_inv2 -> rbt_map_entry_inv2
   716   map_entry_is_rbt -> rbt_map_entry_is_rbt
   717   map_entry_sorted -> rbt_map_entry_rbt_sorted
   718   map_entry_tree_greater -> rbt_map_entry_rbt_greater
   719   map_entry_tree_less -> rbt_map_entry_rbt_less
   720   map_tree_greater -> map_rbt_greater
   721   map_tree_less -> map_rbt_less
   722   map_sorted -> map_rbt_sorted
   723   paint_sorted -> paint_rbt_sorted
   724   paint_lookup -> paint_rbt_lookup
   725   paint_tree_greater -> paint_rbt_greater
   726   paint_tree_less -> paint_rbt_less
   727   sorted_entries -> rbt_sorted_entries
   728   tree_greater_eq_trans -> rbt_greater_eq_trans
   729   tree_greater_nit -> rbt_greater_nit
   730   tree_greater_prop -> rbt_greater_prop
   731   tree_greater_simps -> rbt_greater_simps
   732   tree_greater_trans -> rbt_greater_trans
   733   tree_less_eq_trans -> rbt_less_eq_trans
   734   tree_less_nit -> rbt_less_nit
   735   tree_less_prop -> rbt_less_prop
   736   tree_less_simps -> rbt_less_simps
   737   tree_less_trans -> rbt_less_trans
   738   tree_ord_props -> rbt_ord_props
   739   union_Branch -> rbt_union_Branch
   740   union_is_rbt -> rbt_union_is_rbt
   741   unionw_is_rbt -> rbt_unionw_is_rbt
   742   unionwk_is_rbt -> rbt_unionwk_is_rbt
   743   unionwk_sorted -> rbt_unionwk_rbt_sorted
   744 
   745 * Theory HOL/Library/Float: Floating point numbers are now defined as
   746 a subset of the real numbers.  All operations are defined using the
   747 lifing-framework and proofs use the transfer method.  INCOMPATIBILITY.
   748 
   749   Changed Operations:
   750   float_abs -> abs
   751   float_nprt -> nprt
   752   float_pprt -> pprt
   753   pow2 -> use powr
   754   round_down -> float_round_down
   755   round_up -> float_round_up
   756   scale -> exponent
   757 
   758   Removed Operations:
   759   ceiling_fl, lb_mult, lb_mod, ub_mult, ub_mod
   760 
   761   Renamed Lemmas:
   762   abs_float_def -> Float.compute_float_abs
   763   bitlen_ge0 -> bitlen_nonneg
   764   bitlen.simps -> Float.compute_bitlen
   765   float_components -> Float_mantissa_exponent
   766   float_divl.simps -> Float.compute_float_divl
   767   float_divr.simps -> Float.compute_float_divr
   768   float_eq_odd -> mult_powr_eq_mult_powr_iff
   769   float_power -> real_of_float_power
   770   lapprox_posrat_def -> Float.compute_lapprox_posrat
   771   lapprox_rat.simps -> Float.compute_lapprox_rat
   772   le_float_def' -> Float.compute_float_le
   773   le_float_def -> less_eq_float.rep_eq
   774   less_float_def' -> Float.compute_float_less
   775   less_float_def -> less_float.rep_eq
   776   normfloat_def -> Float.compute_normfloat
   777   normfloat_imp_odd_or_zero -> mantissa_not_dvd and mantissa_noteq_0
   778   normfloat -> normfloat_def
   779   normfloat_unique -> use normfloat_def
   780   number_of_float_Float -> Float.compute_float_numeral, Float.compute_float_neg_numeral
   781   one_float_def -> Float.compute_float_one
   782   plus_float_def -> Float.compute_float_plus
   783   rapprox_posrat_def -> Float.compute_rapprox_posrat
   784   rapprox_rat.simps -> Float.compute_rapprox_rat
   785   real_of_float_0 -> zero_float.rep_eq
   786   real_of_float_1 -> one_float.rep_eq
   787   real_of_float_abs -> abs_float.rep_eq
   788   real_of_float_add -> plus_float.rep_eq
   789   real_of_float_minus -> uminus_float.rep_eq
   790   real_of_float_mult -> times_float.rep_eq
   791   real_of_float_simp -> Float.rep_eq
   792   real_of_float_sub -> minus_float.rep_eq
   793   round_down.simps -> Float.compute_float_round_down
   794   round_up.simps -> Float.compute_float_round_up
   795   times_float_def -> Float.compute_float_times
   796   uminus_float_def -> Float.compute_float_uminus
   797   zero_float_def -> Float.compute_float_zero
   798 
   799   Lemmas not necessary anymore, use the transfer method:
   800   bitlen_B0, bitlen_B1, bitlen_ge1, bitlen_Min, bitlen_Pls, float_divl,
   801   float_divr, float_le_simp, float_less1_mantissa_bound,
   802   float_less_simp, float_less_zero, float_le_zero,
   803   float_pos_less1_e_neg, float_pos_m_pos, float_split, float_split2,
   804   floor_pos_exp, lapprox_posrat, lapprox_posrat_bottom, lapprox_rat,
   805   lapprox_rat_bottom, normalized_float, rapprox_posrat,
   806   rapprox_posrat_le1, rapprox_rat, real_of_float_ge0_exp,
   807   real_of_float_neg_exp, real_of_float_nge0_exp, round_down floor_fl,
   808   round_up, zero_le_float, zero_less_float
   809 
   810 * New theory HOL/Library/DAList provides an abstract type for
   811 association lists with distinct keys.
   812 
   813 * Session HOL/IMP: Added new theory of abstract interpretation of
   814 annotated commands.
   815 
   816 * Session HOL-Import: Re-implementation from scratch is faster,
   817 simpler, and more scalable.  Requires a proof bundle, which is
   818 available as an external component.  Discontinued old (and mostly
   819 dead) Importer for HOL4 and HOL Light.  INCOMPATIBILITY.
   820 
   821 * Session HOL-Word: Discontinued many redundant theorems specific to
   822 type 'a word. INCOMPATIBILITY, use the corresponding generic theorems
   823 instead.
   824 
   825   word_sub_alt ~> word_sub_wi
   826   word_add_alt ~> word_add_def
   827   word_mult_alt ~> word_mult_def
   828   word_minus_alt ~> word_minus_def
   829   word_0_alt ~> word_0_wi
   830   word_1_alt ~> word_1_wi
   831   word_add_0 ~> add_0_left
   832   word_add_0_right ~> add_0_right
   833   word_mult_1 ~> mult_1_left
   834   word_mult_1_right ~> mult_1_right
   835   word_add_commute ~> add_commute
   836   word_add_assoc ~> add_assoc
   837   word_add_left_commute ~> add_left_commute
   838   word_mult_commute ~> mult_commute
   839   word_mult_assoc ~> mult_assoc
   840   word_mult_left_commute ~> mult_left_commute
   841   word_left_distrib ~> left_distrib
   842   word_right_distrib ~> right_distrib
   843   word_left_minus ~> left_minus
   844   word_diff_0_right ~> diff_0_right
   845   word_diff_self ~> diff_self
   846   word_sub_def ~> diff_minus
   847   word_diff_minus ~> diff_minus
   848   word_add_ac ~> add_ac
   849   word_mult_ac ~> mult_ac
   850   word_plus_ac0 ~> add_0_left add_0_right add_ac
   851   word_times_ac1 ~> mult_1_left mult_1_right mult_ac
   852   word_order_trans ~> order_trans
   853   word_order_refl ~> order_refl
   854   word_order_antisym ~> order_antisym
   855   word_order_linear ~> linorder_linear
   856   lenw1_zero_neq_one ~> zero_neq_one
   857   word_number_of_eq ~> number_of_eq
   858   word_of_int_add_hom ~> wi_hom_add
   859   word_of_int_sub_hom ~> wi_hom_sub
   860   word_of_int_mult_hom ~> wi_hom_mult
   861   word_of_int_minus_hom ~> wi_hom_neg
   862   word_of_int_succ_hom ~> wi_hom_succ
   863   word_of_int_pred_hom ~> wi_hom_pred
   864   word_of_int_0_hom ~> word_0_wi
   865   word_of_int_1_hom ~> word_1_wi
   866 
   867 * Session HOL-Word: New proof method "word_bitwise" for splitting
   868 machine word equalities and inequalities into logical circuits,
   869 defined in HOL/Word/WordBitwise.thy.  Supports addition, subtraction,
   870 multiplication, shifting by constants, bitwise operators and numeric
   871 constants.  Requires fixed-length word types, not 'a word.  Solves
   872 many standard word identities outright and converts more into first
   873 order problems amenable to blast or similar.  See also examples in
   874 HOL/Word/Examples/WordExamples.thy.
   875 
   876 * Session HOL-Probability: Introduced the type "'a measure" to
   877 represent measures, this replaces the records 'a algebra and 'a
   878 measure_space.  The locales based on subset_class now have two
   879 locale-parameters the space \<Omega> and the set of measurable sets M.
   880 The product of probability spaces uses now the same constant as the
   881 finite product of sigma-finite measure spaces "PiM :: ('i => 'a)
   882 measure".  Most constants are defined now outside of locales and gain
   883 an additional parameter, like null_sets, almost_eventually or \<mu>'.
   884 Measure space constructions for distributions and densities now got
   885 their own constants distr and density.  Instead of using locales to
   886 describe measure spaces with a finite space, the measure count_space
   887 and point_measure is introduced.  INCOMPATIBILITY.
   888 
   889   Renamed constants:
   890   measure -> emeasure
   891   finite_measure.\<mu>' -> measure
   892   product_algebra_generator -> prod_algebra
   893   product_prob_space.emb -> prod_emb
   894   product_prob_space.infprod_algebra -> PiM
   895 
   896   Removed locales:
   897   completeable_measure_space
   898   finite_measure_space
   899   finite_prob_space
   900   finite_product_finite_prob_space
   901   finite_product_sigma_algebra
   902   finite_sigma_algebra
   903   measure_space
   904   pair_finite_prob_space
   905   pair_finite_sigma_algebra
   906   pair_finite_space
   907   pair_sigma_algebra
   908   product_sigma_algebra
   909 
   910   Removed constants:
   911   conditional_space
   912   distribution -> use distr measure, or distributed predicate
   913   image_space
   914   joint_distribution -> use distr measure, or distributed predicate
   915   pair_measure_generator
   916   product_prob_space.infprod_algebra -> use PiM
   917   subvimage
   918 
   919   Replacement theorems:
   920   finite_additivity_sufficient -> ring_of_sets.countably_additiveI_finite
   921   finite_measure.empty_measure -> measure_empty
   922   finite_measure.finite_continuity_from_above -> finite_measure.finite_Lim_measure_decseq
   923   finite_measure.finite_continuity_from_below -> finite_measure.finite_Lim_measure_incseq
   924   finite_measure.finite_measure_countably_subadditive -> finite_measure.finite_measure_subadditive_countably
   925   finite_measure.finite_measure_eq -> finite_measure.emeasure_eq_measure
   926   finite_measure.finite_measure -> finite_measure.emeasure_finite
   927   finite_measure.finite_measure_finite_singleton -> finite_measure.finite_measure_eq_setsum_singleton
   928   finite_measure.positive_measure' -> measure_nonneg
   929   finite_measure.real_measure -> finite_measure.emeasure_real
   930   finite_product_prob_space.finite_measure_times -> finite_product_prob_space.finite_measure_PiM_emb
   931   finite_product_sigma_algebra.in_P -> sets_PiM_I_finite
   932   finite_product_sigma_algebra.P_empty -> space_PiM_empty, sets_PiM_empty
   933   information_space.conditional_entropy_eq -> information_space.conditional_entropy_simple_distributed
   934   information_space.conditional_entropy_positive -> information_space.conditional_entropy_nonneg_simple
   935   information_space.conditional_mutual_information_eq_mutual_information -> information_space.conditional_mutual_information_eq_mutual_information_simple
   936   information_space.conditional_mutual_information_generic_positive -> information_space.conditional_mutual_information_nonneg_simple
   937   information_space.conditional_mutual_information_positive -> information_space.conditional_mutual_information_nonneg_simple
   938   information_space.entropy_commute -> information_space.entropy_commute_simple
   939   information_space.entropy_eq -> information_space.entropy_simple_distributed
   940   information_space.entropy_generic_eq -> information_space.entropy_simple_distributed
   941   information_space.entropy_positive -> information_space.entropy_nonneg_simple
   942   information_space.entropy_uniform_max -> information_space.entropy_uniform
   943   information_space.KL_eq_0_imp -> information_space.KL_eq_0_iff_eq
   944   information_space.KL_eq_0 -> information_space.KL_same_eq_0
   945   information_space.KL_ge_0 -> information_space.KL_nonneg
   946   information_space.mutual_information_eq -> information_space.mutual_information_simple_distributed
   947   information_space.mutual_information_positive -> information_space.mutual_information_nonneg_simple
   948   Int_stable_cuboids -> Int_stable_atLeastAtMost
   949   Int_stable_product_algebra_generator -> positive_integral
   950   measure_preserving -> equality "distr M N f = N" "f : measurable M N"
   951   measure_space.additive -> emeasure_additive
   952   measure_space.AE_iff_null_set -> AE_iff_null
   953   measure_space.almost_everywhere_def -> eventually_ae_filter
   954   measure_space.almost_everywhere_vimage -> AE_distrD
   955   measure_space.continuity_from_above -> INF_emeasure_decseq
   956   measure_space.continuity_from_above_Lim -> Lim_emeasure_decseq
   957   measure_space.continuity_from_below_Lim -> Lim_emeasure_incseq
   958   measure_space.continuity_from_below -> SUP_emeasure_incseq
   959   measure_space_density -> emeasure_density
   960   measure_space.density_is_absolutely_continuous -> absolutely_continuousI_density
   961   measure_space.integrable_vimage -> integrable_distr
   962   measure_space.integral_translated_density -> integral_density
   963   measure_space.integral_vimage -> integral_distr
   964   measure_space.measure_additive -> plus_emeasure
   965   measure_space.measure_compl -> emeasure_compl
   966   measure_space.measure_countable_increasing -> emeasure_countable_increasing
   967   measure_space.measure_countably_subadditive -> emeasure_subadditive_countably
   968   measure_space.measure_decseq -> decseq_emeasure
   969   measure_space.measure_Diff -> emeasure_Diff
   970   measure_space.measure_Diff_null_set -> emeasure_Diff_null_set
   971   measure_space.measure_eq_0 -> emeasure_eq_0
   972   measure_space.measure_finitely_subadditive -> emeasure_subadditive_finite
   973   measure_space.measure_finite_singleton -> emeasure_eq_setsum_singleton
   974   measure_space.measure_incseq -> incseq_emeasure
   975   measure_space.measure_insert -> emeasure_insert
   976   measure_space.measure_mono -> emeasure_mono
   977   measure_space.measure_not_negative -> emeasure_not_MInf
   978   measure_space.measure_preserving_Int_stable -> measure_eqI_generator_eq
   979   measure_space.measure_setsum -> setsum_emeasure
   980   measure_space.measure_setsum_split -> setsum_emeasure_cover
   981   measure_space.measure_space_vimage -> emeasure_distr
   982   measure_space.measure_subadditive_finite -> emeasure_subadditive_finite
   983   measure_space.measure_subadditive -> subadditive
   984   measure_space.measure_top -> emeasure_space
   985   measure_space.measure_UN_eq_0 -> emeasure_UN_eq_0
   986   measure_space.measure_Un_null_set -> emeasure_Un_null_set
   987   measure_space.positive_integral_translated_density -> positive_integral_density
   988   measure_space.positive_integral_vimage -> positive_integral_distr
   989   measure_space.real_continuity_from_above -> Lim_measure_decseq
   990   measure_space.real_continuity_from_below -> Lim_measure_incseq
   991   measure_space.real_measure_countably_subadditive -> measure_subadditive_countably
   992   measure_space.real_measure_Diff -> measure_Diff
   993   measure_space.real_measure_finite_Union -> measure_finite_Union
   994   measure_space.real_measure_setsum_singleton -> measure_eq_setsum_singleton
   995   measure_space.real_measure_subadditive -> measure_subadditive
   996   measure_space.real_measure_Union -> measure_Union
   997   measure_space.real_measure_UNION -> measure_UNION
   998   measure_space.simple_function_vimage -> simple_function_comp
   999   measure_space.simple_integral_vimage -> simple_integral_distr
  1000   measure_space.simple_integral_vimage -> simple_integral_distr
  1001   measure_unique_Int_stable -> measure_eqI_generator_eq
  1002   measure_unique_Int_stable_vimage -> measure_eqI_generator_eq
  1003   pair_sigma_algebra.measurable_cut_fst -> sets_Pair1
  1004   pair_sigma_algebra.measurable_cut_snd -> sets_Pair2
  1005   pair_sigma_algebra.measurable_pair_image_fst -> measurable_Pair1
  1006   pair_sigma_algebra.measurable_pair_image_snd -> measurable_Pair2
  1007   pair_sigma_algebra.measurable_product_swap -> measurable_pair_swap_iff
  1008   pair_sigma_algebra.pair_sigma_algebra_measurable -> measurable_pair_swap
  1009   pair_sigma_algebra.pair_sigma_algebra_swap_measurable -> measurable_pair_swap'
  1010   pair_sigma_algebra.sets_swap -> sets_pair_swap
  1011   pair_sigma_finite.measure_cut_measurable_fst -> pair_sigma_finite.measurable_emeasure_Pair1
  1012   pair_sigma_finite.measure_cut_measurable_snd -> pair_sigma_finite.measurable_emeasure_Pair2
  1013   pair_sigma_finite.measure_preserving_swap -> pair_sigma_finite.distr_pair_swap
  1014   pair_sigma_finite.pair_measure_alt2 -> pair_sigma_finite.emeasure_pair_measure_alt2
  1015   pair_sigma_finite.pair_measure_alt -> pair_sigma_finite.emeasure_pair_measure_alt
  1016   pair_sigma_finite.pair_measure_times -> pair_sigma_finite.emeasure_pair_measure_Times
  1017   prob_space.indep_distribution_eq_measure -> prob_space.indep_vars_iff_distr_eq_PiM
  1018   prob_space.indep_var_distributionD -> prob_space.indep_var_distribution_eq
  1019   prob_space.measure_space_1 -> prob_space.emeasure_space_1
  1020   prob_space.prob_space_vimage -> prob_space_distr
  1021   prob_space.random_variable_restrict -> measurable_restrict
  1022   prob_space_unique_Int_stable -> measure_eqI_prob_space
  1023   product_algebraE -> prod_algebraE_all
  1024   product_algebra_generator_der -> prod_algebra_eq_finite
  1025   product_algebra_generator_into_space -> prod_algebra_sets_into_space
  1026   product_algebraI -> sets_PiM_I_finite
  1027   product_measure_exists -> product_sigma_finite.sigma_finite
  1028   product_prob_space.finite_index_eq_finite_product -> product_prob_space.sets_PiM_generator
  1029   product_prob_space.finite_measure_infprod_emb_Pi -> product_prob_space.measure_PiM_emb
  1030   product_prob_space.infprod_spec -> product_prob_space.emeasure_PiM_emb_not_empty
  1031   product_prob_space.measurable_component -> measurable_component_singleton
  1032   product_prob_space.measurable_emb -> measurable_prod_emb
  1033   product_prob_space.measurable_into_infprod_algebra -> measurable_PiM_single
  1034   product_prob_space.measurable_singleton_infprod -> measurable_component_singleton
  1035   product_prob_space.measure_emb -> emeasure_prod_emb
  1036   product_prob_space.measure_preserving_restrict -> product_prob_space.distr_restrict
  1037   product_sigma_algebra.product_algebra_into_space -> space_closed
  1038   product_sigma_finite.measure_fold -> product_sigma_finite.distr_merge
  1039   product_sigma_finite.measure_preserving_component_singelton -> product_sigma_finite.distr_singleton
  1040   product_sigma_finite.measure_preserving_merge -> product_sigma_finite.distr_merge
  1041   sequence_space.measure_infprod -> sequence_space.measure_PiM_countable
  1042   sets_product_algebra -> sets_PiM
  1043   sigma_algebra.measurable_sigma -> measurable_measure_of
  1044   sigma_finite_measure.disjoint_sigma_finite -> sigma_finite_disjoint
  1045   sigma_finite_measure.RN_deriv_vimage -> sigma_finite_measure.RN_deriv_distr
  1046   sigma_product_algebra_sigma_eq -> sigma_prod_algebra_sigma_eq
  1047   space_product_algebra -> space_PiM
  1048 
  1049 * Session HOL-TPTP: support to parse and import TPTP problems (all
  1050 languages) into Isabelle/HOL.
  1051 
  1052 
  1053 *** FOL ***
  1054 
  1055 * New "case_product" attribute (see HOL).
  1056 
  1057 
  1058 *** ZF ***
  1059 
  1060 * Greater support for structured proofs involving induction or case
  1061 analysis.
  1062 
  1063 * Much greater use of mathematical symbols.
  1064 
  1065 * Removal of many ML theorem bindings.  INCOMPATIBILITY.
  1066 
  1067 
  1068 *** ML ***
  1069 
  1070 * Antiquotation @{keyword "name"} produces a parser for outer syntax
  1071 from a minor keyword introduced via theory header declaration.
  1072 
  1073 * Antiquotation @{command_spec "name"} produces the
  1074 Outer_Syntax.command_spec from a major keyword introduced via theory
  1075 header declaration; it can be passed to Outer_Syntax.command etc.
  1076 
  1077 * Local_Theory.define no longer hard-wires default theorem name
  1078 "foo_def", but retains the binding as given.  If that is Binding.empty
  1079 / Attrib.empty_binding, the result is not registered as user-level
  1080 fact.  The Local_Theory.define_internal variant allows to specify a
  1081 non-empty name (used for the foundation in the background theory),
  1082 while omitting the fact binding in the user-context.  Potential
  1083 INCOMPATIBILITY for derived definitional packages: need to specify
  1084 naming policy for primitive definitions more explicitly.
  1085 
  1086 * Renamed Thm.capply to Thm.apply, and Thm.cabs to Thm.lambda in
  1087 conformance with similar operations in structure Term and Logic.
  1088 
  1089 * Antiquotation @{attributes [...]} embeds attribute source
  1090 representation into the ML text, which is particularly useful with
  1091 declarations like Local_Theory.note.
  1092 
  1093 * Structure Proof_Context follows standard naming scheme.  Old
  1094 ProofContext has been discontinued.  INCOMPATIBILITY.
  1095 
  1096 * Refined Local_Theory.declaration {syntax, pervasive}, with subtle
  1097 change of semantics: update is applied to auxiliary local theory
  1098 context as well.
  1099 
  1100 * Modernized some old-style infix operations:
  1101 
  1102   addeqcongs    ~> Simplifier.add_eqcong
  1103   deleqcongs    ~> Simplifier.del_eqcong
  1104   addcongs      ~> Simplifier.add_cong
  1105   delcongs      ~> Simplifier.del_cong
  1106   setmksimps    ~> Simplifier.set_mksimps
  1107   setmkcong     ~> Simplifier.set_mkcong
  1108   setmksym      ~> Simplifier.set_mksym
  1109   setmkeqTrue   ~> Simplifier.set_mkeqTrue
  1110   settermless   ~> Simplifier.set_termless
  1111   setsubgoaler  ~> Simplifier.set_subgoaler
  1112   addsplits     ~> Splitter.add_split
  1113   delsplits     ~> Splitter.del_split
  1114 
  1115 
  1116 *** System ***
  1117 
  1118 * USER_HOME settings variable points to cross-platform user home
  1119 directory, which coincides with HOME on POSIX systems only.  Likewise,
  1120 the Isabelle path specification "~" now expands to $USER_HOME, instead
  1121 of former $HOME.  A different default for USER_HOME may be set
  1122 explicitly in shell environment, before Isabelle settings are
  1123 evaluated.  Minor INCOMPATIBILITY: need to adapt Isabelle path where
  1124 the generic user home was intended.
  1125 
  1126 * ISABELLE_HOME_WINDOWS refers to ISABELLE_HOME in windows file name
  1127 notation, which is useful for the jEdit file browser, for example.
  1128 
  1129 * ISABELLE_JDK_HOME settings variable points to JDK with javac and jar
  1130 (not just JRE).
  1131 
  1132 
  1133 
  1134 New in Isabelle2011-1 (October 2011)
  1135 ------------------------------------
  1136 
  1137 *** General ***
  1138 
  1139 * Improved Isabelle/jEdit Prover IDE (PIDE), which can be invoked as
  1140 "isabelle jedit" or "ISABELLE_HOME/Isabelle" on the command line.
  1141 
  1142   - Management of multiple theory files directly from the editor
  1143     buffer store -- bypassing the file-system (no requirement to save
  1144     files for checking).
  1145 
  1146   - Markup of formal entities within the text buffer, with semantic
  1147     highlighting, tooltips and hyperlinks to jump to defining source
  1148     positions.
  1149 
  1150   - Improved text rendering, with sub/superscripts in the source
  1151     buffer (including support for copy/paste wrt. output panel, HTML
  1152     theory output and other non-Isabelle text boxes).
  1153 
  1154   - Refined scheduling of proof checking and printing of results,
  1155     based on interactive editor view.  (Note: jEdit folding and
  1156     narrowing allows to restrict buffer perspectives explicitly.)
  1157 
  1158   - Reduced CPU performance requirements, usable on machines with few
  1159     cores.
  1160 
  1161   - Reduced memory requirements due to pruning of unused document
  1162     versions (garbage collection).
  1163 
  1164 See also ~~/src/Tools/jEdit/README.html for further information,
  1165 including some remaining limitations.
  1166 
  1167 * Theory loader: source files are exclusively located via the master
  1168 directory of each theory node (where the .thy file itself resides).
  1169 The global load path (such as src/HOL/Library) has been discontinued.
  1170 Note that the path element ~~ may be used to reference theories in the
  1171 Isabelle home folder -- for instance, "~~/src/HOL/Library/FuncSet".
  1172 INCOMPATIBILITY.
  1173 
  1174 * Theory loader: source files are identified by content via SHA1
  1175 digests.  Discontinued former path/modtime identification and optional
  1176 ISABELLE_FILE_IDENT plugin scripts.
  1177 
  1178 * Parallelization of nested Isar proofs is subject to
  1179 Goal.parallel_proofs_threshold (default 100).  See also isabelle
  1180 usedir option -Q.
  1181 
  1182 * Name space: former unsynchronized references are now proper
  1183 configuration options, with more conventional names:
  1184 
  1185   long_names   ~> names_long
  1186   short_names  ~> names_short
  1187   unique_names ~> names_unique
  1188 
  1189 Minor INCOMPATIBILITY, need to declare options in context like this:
  1190 
  1191   declare [[names_unique = false]]
  1192 
  1193 * Literal facts `prop` may contain dummy patterns, e.g. `_ = _`.  Note
  1194 that the result needs to be unique, which means fact specifications
  1195 may have to be refined after enriching a proof context.
  1196 
  1197 * Attribute "case_names" has been refined: the assumptions in each case
  1198 can be named now by following the case name with [name1 name2 ...].
  1199 
  1200 * Isabelle/Isar reference manual has been updated and extended:
  1201   - "Synopsis" provides a catalog of main Isar language concepts.
  1202   - Formal references in syntax diagrams, via @{rail} antiquotation.
  1203   - Updated material from classic "ref" manual, notably about
  1204     "Classical Reasoner".
  1205 
  1206 
  1207 *** HOL ***
  1208 
  1209 * Class bot and top require underlying partial order rather than
  1210 preorder: uniqueness of bot and top is guaranteed.  INCOMPATIBILITY.
  1211 
  1212 * Class complete_lattice: generalized a couple of lemmas from sets;
  1213 generalized theorems INF_cong and SUP_cong.  New type classes for
  1214 complete boolean algebras and complete linear orders.  Lemmas
  1215 Inf_less_iff, less_Sup_iff, INF_less_iff, less_SUP_iff now reside in
  1216 class complete_linorder.
  1217 
  1218 Changed proposition of lemmas Inf_bool_def, Sup_bool_def, Inf_fun_def,
  1219 Sup_fun_def, Inf_apply, Sup_apply.
  1220 
  1221 Removed redundant lemmas (the right hand side gives hints how to
  1222 replace them for (metis ...), or (simp only: ...) proofs):
  1223 
  1224   Inf_singleton ~> Inf_insert [where A="{}", unfolded Inf_empty inf_top_right]
  1225   Sup_singleton ~> Sup_insert [where A="{}", unfolded Sup_empty sup_bot_right]
  1226   Inf_binary ~> Inf_insert, Inf_empty, and inf_top_right
  1227   Sup_binary ~> Sup_insert, Sup_empty, and sup_bot_right
  1228   Int_eq_Inter ~> Inf_insert, Inf_empty, and inf_top_right
  1229   Un_eq_Union ~> Sup_insert, Sup_empty, and sup_bot_right
  1230   Inter_def ~> INF_def, image_def
  1231   Union_def ~> SUP_def, image_def
  1232   INT_eq ~> INF_def, and image_def
  1233   UN_eq ~> SUP_def, and image_def
  1234   INF_subset ~> INF_superset_mono [OF _ order_refl]
  1235 
  1236 More consistent and comprehensive names:
  1237 
  1238   INTER_eq_Inter_image ~> INF_def
  1239   UNION_eq_Union_image ~> SUP_def
  1240   INFI_def ~> INF_def
  1241   SUPR_def ~> SUP_def
  1242   INF_leI ~> INF_lower
  1243   INF_leI2 ~> INF_lower2
  1244   le_INFI ~> INF_greatest
  1245   le_SUPI ~> SUP_upper
  1246   le_SUPI2 ~> SUP_upper2
  1247   SUP_leI ~> SUP_least
  1248   INFI_bool_eq ~> INF_bool_eq
  1249   SUPR_bool_eq ~> SUP_bool_eq
  1250   INFI_apply ~> INF_apply
  1251   SUPR_apply ~> SUP_apply
  1252   INTER_def ~> INTER_eq
  1253   UNION_def ~> UNION_eq
  1254 
  1255 INCOMPATIBILITY.
  1256 
  1257 * Renamed theory Complete_Lattice to Complete_Lattices.
  1258 INCOMPATIBILITY.
  1259 
  1260 * Theory Complete_Lattices: lemmas Inf_eq_top_iff, INF_eq_top_iff,
  1261 INF_image, Inf_insert, INF_top, Inf_top_conv, INF_top_conv, SUP_bot,
  1262 Sup_bot_conv, SUP_bot_conv, Sup_eq_top_iff, SUP_eq_top_iff, SUP_image,
  1263 Sup_insert are now declared as [simp].  INCOMPATIBILITY.
  1264 
  1265 * Theory Lattice: lemmas compl_inf_bot, compl_le_comp_iff,
  1266 compl_sup_top, inf_idem, inf_left_idem, inf_sup_absorb, sup_idem,
  1267 sup_inf_absob, sup_left_idem are now declared as [simp].  Minor
  1268 INCOMPATIBILITY.
  1269 
  1270 * Added syntactic classes "inf" and "sup" for the respective
  1271 constants.  INCOMPATIBILITY: Changes in the argument order of the
  1272 (mostly internal) locale predicates for some derived classes.
  1273 
  1274 * Theorem collections ball_simps and bex_simps do not contain theorems
  1275 referring to UNION any longer; these have been moved to collection
  1276 UN_ball_bex_simps.  INCOMPATIBILITY.
  1277 
  1278 * Theory Archimedean_Field: floor now is defined as parameter of a
  1279 separate type class floor_ceiling.
  1280 
  1281 * Theory Finite_Set: more coherent development of fold_set locales:
  1282 
  1283     locale fun_left_comm ~> locale comp_fun_commute
  1284     locale fun_left_comm_idem ~> locale comp_fun_idem
  1285 
  1286 Both use point-free characterization; interpretation proofs may need
  1287 adjustment.  INCOMPATIBILITY.
  1288 
  1289 * Theory Limits: Type "'a net" has been renamed to "'a filter", in
  1290 accordance with standard mathematical terminology. INCOMPATIBILITY.
  1291 
  1292 * Theory Complex_Main: The locale interpretations for the
  1293 bounded_linear and bounded_bilinear locales have been removed, in
  1294 order to reduce the number of duplicate lemmas. Users must use the
  1295 original names for distributivity theorems, potential INCOMPATIBILITY.
  1296 
  1297   divide.add ~> add_divide_distrib
  1298   divide.diff ~> diff_divide_distrib
  1299   divide.setsum ~> setsum_divide_distrib
  1300   mult.add_right ~> right_distrib
  1301   mult.diff_right ~> right_diff_distrib
  1302   mult_right.setsum ~> setsum_right_distrib
  1303   mult_left.diff ~> left_diff_distrib
  1304 
  1305 * Theory Complex_Main: Several redundant theorems have been removed or
  1306 replaced by more general versions. INCOMPATIBILITY.
  1307 
  1308   real_diff_def ~> minus_real_def
  1309   real_divide_def ~> divide_real_def
  1310   real_less_def ~> less_le
  1311   real_abs_def ~> abs_real_def
  1312   real_sgn_def ~> sgn_real_def
  1313   real_mult_commute ~> mult_commute
  1314   real_mult_assoc ~> mult_assoc
  1315   real_mult_1 ~> mult_1_left
  1316   real_add_mult_distrib ~> left_distrib
  1317   real_zero_not_eq_one ~> zero_neq_one
  1318   real_mult_inverse_left ~> left_inverse
  1319   INVERSE_ZERO ~> inverse_zero
  1320   real_le_refl ~> order_refl
  1321   real_le_antisym ~> order_antisym
  1322   real_le_trans ~> order_trans
  1323   real_le_linear ~> linear
  1324   real_le_eq_diff ~> le_iff_diff_le_0
  1325   real_add_left_mono ~> add_left_mono
  1326   real_mult_order ~> mult_pos_pos
  1327   real_mult_less_mono2 ~> mult_strict_left_mono
  1328   real_of_int_real_of_nat ~> real_of_int_of_nat_eq
  1329   real_0_le_divide_iff ~> zero_le_divide_iff
  1330   realpow_two_disj ~> power2_eq_iff
  1331   real_squared_diff_one_factored ~> square_diff_one_factored
  1332   realpow_two_diff ~> square_diff_square_factored
  1333   reals_complete2 ~> complete_real
  1334   real_sum_squared_expand ~> power2_sum
  1335   exp_ln_eq ~> ln_unique
  1336   expi_add ~> exp_add
  1337   expi_zero ~> exp_zero
  1338   lemma_DERIV_subst ~> DERIV_cong
  1339   LIMSEQ_Zfun_iff ~> tendsto_Zfun_iff
  1340   LIMSEQ_const ~> tendsto_const
  1341   LIMSEQ_norm ~> tendsto_norm
  1342   LIMSEQ_add ~> tendsto_add
  1343   LIMSEQ_minus ~> tendsto_minus
  1344   LIMSEQ_minus_cancel ~> tendsto_minus_cancel
  1345   LIMSEQ_diff ~> tendsto_diff
  1346   bounded_linear.LIMSEQ ~> bounded_linear.tendsto
  1347   bounded_bilinear.LIMSEQ ~> bounded_bilinear.tendsto
  1348   LIMSEQ_mult ~> tendsto_mult
  1349   LIMSEQ_inverse ~> tendsto_inverse
  1350   LIMSEQ_divide ~> tendsto_divide
  1351   LIMSEQ_pow ~> tendsto_power
  1352   LIMSEQ_setsum ~> tendsto_setsum
  1353   LIMSEQ_setprod ~> tendsto_setprod
  1354   LIMSEQ_norm_zero ~> tendsto_norm_zero_iff
  1355   LIMSEQ_rabs_zero ~> tendsto_rabs_zero_iff
  1356   LIMSEQ_imp_rabs ~> tendsto_rabs
  1357   LIMSEQ_add_minus ~> tendsto_add [OF _ tendsto_minus]
  1358   LIMSEQ_add_const ~> tendsto_add [OF _ tendsto_const]
  1359   LIMSEQ_diff_const ~> tendsto_diff [OF _ tendsto_const]
  1360   LIMSEQ_Complex ~> tendsto_Complex
  1361   LIM_ident ~> tendsto_ident_at
  1362   LIM_const ~> tendsto_const
  1363   LIM_add ~> tendsto_add
  1364   LIM_add_zero ~> tendsto_add_zero
  1365   LIM_minus ~> tendsto_minus
  1366   LIM_diff ~> tendsto_diff
  1367   LIM_norm ~> tendsto_norm
  1368   LIM_norm_zero ~> tendsto_norm_zero
  1369   LIM_norm_zero_cancel ~> tendsto_norm_zero_cancel
  1370   LIM_norm_zero_iff ~> tendsto_norm_zero_iff
  1371   LIM_rabs ~> tendsto_rabs
  1372   LIM_rabs_zero ~> tendsto_rabs_zero
  1373   LIM_rabs_zero_cancel ~> tendsto_rabs_zero_cancel
  1374   LIM_rabs_zero_iff ~> tendsto_rabs_zero_iff
  1375   LIM_compose ~> tendsto_compose
  1376   LIM_mult ~> tendsto_mult
  1377   LIM_scaleR ~> tendsto_scaleR
  1378   LIM_of_real ~> tendsto_of_real
  1379   LIM_power ~> tendsto_power
  1380   LIM_inverse ~> tendsto_inverse
  1381   LIM_sgn ~> tendsto_sgn
  1382   isCont_LIM_compose ~> isCont_tendsto_compose
  1383   bounded_linear.LIM ~> bounded_linear.tendsto
  1384   bounded_linear.LIM_zero ~> bounded_linear.tendsto_zero
  1385   bounded_bilinear.LIM ~> bounded_bilinear.tendsto
  1386   bounded_bilinear.LIM_prod_zero ~> bounded_bilinear.tendsto_zero
  1387   bounded_bilinear.LIM_left_zero ~> bounded_bilinear.tendsto_left_zero
  1388   bounded_bilinear.LIM_right_zero ~> bounded_bilinear.tendsto_right_zero
  1389   LIM_inverse_fun ~> tendsto_inverse [OF tendsto_ident_at]
  1390 
  1391 * Theory Complex_Main: The definition of infinite series was
  1392 generalized.  Now it is defined on the type class {topological_space,
  1393 comm_monoid_add}.  Hence it is useable also for extended real numbers.
  1394 
  1395 * Theory Complex_Main: The complex exponential function "expi" is now
  1396 a type-constrained abbreviation for "exp :: complex => complex"; thus
  1397 several polymorphic lemmas about "exp" are now applicable to "expi".
  1398 
  1399 * Code generation:
  1400 
  1401   - Theory Library/Code_Char_ord provides native ordering of
  1402     characters in the target language.
  1403 
  1404   - Commands code_module and code_library are legacy, use export_code
  1405     instead.
  1406 
  1407   - Method "evaluation" is legacy, use method "eval" instead.
  1408 
  1409   - Legacy evaluator "SML" is deactivated by default.  May be
  1410     reactivated by the following theory command:
  1411 
  1412       setup {* Value.add_evaluator ("SML", Codegen.eval_term) *}
  1413 
  1414 * Declare ext [intro] by default.  Rare INCOMPATIBILITY.
  1415 
  1416 * New proof method "induction" that gives induction hypotheses the
  1417 name "IH", thus distinguishing them from further hypotheses that come
  1418 from rule induction.  The latter are still called "hyps".  Method
  1419 "induction" is a thin wrapper around "induct" and follows the same
  1420 syntax.
  1421 
  1422 * Method "fastsimp" has been renamed to "fastforce", but "fastsimp" is
  1423 still available as a legacy feature for some time.
  1424 
  1425 * Nitpick:
  1426   - Added "need" and "total_consts" options.
  1427   - Reintroduced "show_skolems" option by popular demand.
  1428   - Renamed attribute: nitpick_def ~> nitpick_unfold.
  1429     INCOMPATIBILITY.
  1430 
  1431 * Sledgehammer:
  1432   - Use quasi-sound (and efficient) translations by default.
  1433   - Added support for the following provers: E-ToFoF, LEO-II,
  1434     Satallax, SNARK, Waldmeister, and Z3 with TPTP syntax.
  1435   - Automatically preplay and minimize proofs before showing them if
  1436     this can be done within reasonable time.
  1437   - sledgehammer available_provers ~> sledgehammer supported_provers.
  1438     INCOMPATIBILITY.
  1439   - Added "preplay_timeout", "slicing", "type_enc", "sound",
  1440     "max_mono_iters", and "max_new_mono_instances" options.
  1441   - Removed "explicit_apply" and "full_types" options as well as "Full
  1442     Types" Proof General menu item. INCOMPATIBILITY.
  1443 
  1444 * Metis:
  1445   - Removed "metisF" -- use "metis" instead. INCOMPATIBILITY.
  1446   - Obsoleted "metisFT" -- use "metis (full_types)" instead.
  1447     INCOMPATIBILITY.
  1448 
  1449 * Command 'try':
  1450   - Renamed 'try_methods' and added "simp:", "intro:", "dest:", and
  1451     "elim:" options. INCOMPATIBILITY.
  1452   - Introduced 'try' that not only runs 'try_methods' but also
  1453     'solve_direct', 'sledgehammer', 'quickcheck', and 'nitpick'.
  1454 
  1455 * Quickcheck:
  1456   - Added "eval" option to evaluate terms for the found counterexample
  1457     (currently only supported by the default (exhaustive) tester).
  1458   - Added post-processing of terms to obtain readable counterexamples
  1459     (currently only supported by the default (exhaustive) tester).
  1460   - New counterexample generator quickcheck[narrowing] enables
  1461     narrowing-based testing.  Requires the Glasgow Haskell compiler
  1462     with its installation location defined in the Isabelle settings
  1463     environment as ISABELLE_GHC.
  1464   - Removed quickcheck tester "SML" based on the SML code generator
  1465     (formly in HOL/Library).
  1466 
  1467 * Function package: discontinued option "tailrec".  INCOMPATIBILITY,
  1468 use 'partial_function' instead.
  1469 
  1470 * Theory Library/Extended_Reals replaces now the positive extended
  1471 reals found in probability theory. This file is extended by
  1472 Multivariate_Analysis/Extended_Real_Limits.
  1473 
  1474 * Theory Library/Old_Recdef: old 'recdef' package has been moved here,
  1475 from where it must be imported explicitly if it is really required.
  1476 INCOMPATIBILITY.
  1477 
  1478 * Theory Library/Wfrec: well-founded recursion combinator "wfrec" has
  1479 been moved here.  INCOMPATIBILITY.
  1480 
  1481 * Theory Library/Saturated provides type of numbers with saturated
  1482 arithmetic.
  1483 
  1484 * Theory Library/Product_Lattice defines a pointwise ordering for the
  1485 product type 'a * 'b, and provides instance proofs for various order
  1486 and lattice type classes.
  1487 
  1488 * Theory Library/Countable now provides the "countable_datatype" proof
  1489 method for proving "countable" class instances for datatypes.
  1490 
  1491 * Theory Library/Cset_Monad allows do notation for computable sets
  1492 (cset) via the generic monad ad-hoc overloading facility.
  1493 
  1494 * Library: Theories of common data structures are split into theories
  1495 for implementation, an invariant-ensuring type, and connection to an
  1496 abstract type. INCOMPATIBILITY.
  1497 
  1498   - RBT is split into RBT and RBT_Mapping.
  1499   - AssocList is split and renamed into AList and AList_Mapping.
  1500   - DList is split into DList_Impl, DList, and DList_Cset.
  1501   - Cset is split into Cset and List_Cset.
  1502 
  1503 * Theory Library/Nat_Infinity has been renamed to
  1504 Library/Extended_Nat, with name changes of the following types and
  1505 constants:
  1506 
  1507   type inat   ~> type enat
  1508   Fin         ~> enat
  1509   Infty       ~> infinity (overloaded)
  1510   iSuc        ~> eSuc
  1511   the_Fin     ~> the_enat
  1512 
  1513 Every theorem name containing "inat", "Fin", "Infty", or "iSuc" has
  1514 been renamed accordingly. INCOMPATIBILITY.
  1515 
  1516 * Session Multivariate_Analysis: The euclidean_space type class now
  1517 fixes a constant "Basis :: 'a set" consisting of the standard
  1518 orthonormal basis for the type. Users now have the option of
  1519 quantifying over this set instead of using the "basis" function, e.g.
  1520 "ALL x:Basis. P x" vs "ALL i<DIM('a). P (basis i)".
  1521 
  1522 * Session Multivariate_Analysis: Type "('a, 'b) cart" has been renamed
  1523 to "('a, 'b) vec" (the syntax "'a ^ 'b" remains unaffected). Constants
  1524 "Cart_nth" and "Cart_lambda" have been respectively renamed to
  1525 "vec_nth" and "vec_lambda"; theorems mentioning those names have
  1526 changed to match. Definition theorems for overloaded constants now use
  1527 the standard "foo_vec_def" naming scheme. A few other theorems have
  1528 been renamed as follows (INCOMPATIBILITY):
  1529 
  1530   Cart_eq          ~> vec_eq_iff
  1531   dist_nth_le_cart ~> dist_vec_nth_le
  1532   tendsto_vector   ~> vec_tendstoI
  1533   Cauchy_vector    ~> vec_CauchyI
  1534 
  1535 * Session Multivariate_Analysis: Several duplicate theorems have been
  1536 removed, and other theorems have been renamed or replaced with more
  1537 general versions. INCOMPATIBILITY.
  1538 
  1539   finite_choice ~> finite_set_choice
  1540   eventually_conjI ~> eventually_conj
  1541   eventually_and ~> eventually_conj_iff
  1542   eventually_false ~> eventually_False
  1543   setsum_norm ~> norm_setsum
  1544   Lim_sequentially ~> LIMSEQ_def
  1545   Lim_ident_at ~> LIM_ident
  1546   Lim_const ~> tendsto_const
  1547   Lim_cmul ~> tendsto_scaleR [OF tendsto_const]
  1548   Lim_neg ~> tendsto_minus
  1549   Lim_add ~> tendsto_add
  1550   Lim_sub ~> tendsto_diff
  1551   Lim_mul ~> tendsto_scaleR
  1552   Lim_vmul ~> tendsto_scaleR [OF _ tendsto_const]
  1553   Lim_null_norm ~> tendsto_norm_zero_iff [symmetric]
  1554   Lim_linear ~> bounded_linear.tendsto
  1555   Lim_component ~> tendsto_euclidean_component
  1556   Lim_component_cart ~> tendsto_vec_nth
  1557   Lim_inner ~> tendsto_inner [OF tendsto_const]
  1558   dot_lsum ~> inner_setsum_left
  1559   dot_rsum ~> inner_setsum_right
  1560   continuous_cmul ~> continuous_scaleR [OF continuous_const]
  1561   continuous_neg ~> continuous_minus
  1562   continuous_sub ~> continuous_diff
  1563   continuous_vmul ~> continuous_scaleR [OF _ continuous_const]
  1564   continuous_mul ~> continuous_scaleR
  1565   continuous_inv ~> continuous_inverse
  1566   continuous_at_within_inv ~> continuous_at_within_inverse
  1567   continuous_at_inv ~> continuous_at_inverse
  1568   continuous_at_norm ~> continuous_norm [OF continuous_at_id]
  1569   continuous_at_infnorm ~> continuous_infnorm [OF continuous_at_id]
  1570   continuous_at_component ~> continuous_component [OF continuous_at_id]
  1571   continuous_on_neg ~> continuous_on_minus
  1572   continuous_on_sub ~> continuous_on_diff
  1573   continuous_on_cmul ~> continuous_on_scaleR [OF continuous_on_const]
  1574   continuous_on_vmul ~> continuous_on_scaleR [OF _ continuous_on_const]
  1575   continuous_on_mul ~> continuous_on_scaleR
  1576   continuous_on_mul_real ~> continuous_on_mult
  1577   continuous_on_inner ~> continuous_on_inner [OF continuous_on_const]
  1578   continuous_on_norm ~> continuous_on_norm [OF continuous_on_id]
  1579   continuous_on_inverse ~> continuous_on_inv
  1580   uniformly_continuous_on_neg ~> uniformly_continuous_on_minus
  1581   uniformly_continuous_on_sub ~> uniformly_continuous_on_diff
  1582   subset_interior ~> interior_mono
  1583   subset_closure ~> closure_mono
  1584   closure_univ ~> closure_UNIV
  1585   real_arch_lt ~> reals_Archimedean2
  1586   real_arch ~> reals_Archimedean3
  1587   real_abs_norm ~> abs_norm_cancel
  1588   real_abs_sub_norm ~> norm_triangle_ineq3
  1589   norm_cauchy_schwarz_abs ~> Cauchy_Schwarz_ineq2
  1590 
  1591 * Session HOL-Probability:
  1592   - Caratheodory's extension lemma is now proved for ring_of_sets.
  1593   - Infinite products of probability measures are now available.
  1594   - Sigma closure is independent, if the generator is independent
  1595   - Use extended reals instead of positive extended
  1596     reals. INCOMPATIBILITY.
  1597 
  1598 * Session HOLCF: Discontinued legacy theorem names, INCOMPATIBILITY.
  1599 
  1600   expand_fun_below ~> fun_below_iff
  1601   below_fun_ext ~> fun_belowI
  1602   expand_cfun_eq ~> cfun_eq_iff
  1603   ext_cfun ~> cfun_eqI
  1604   expand_cfun_below ~> cfun_below_iff
  1605   below_cfun_ext ~> cfun_belowI
  1606   monofun_fun_fun ~> fun_belowD
  1607   monofun_fun_arg ~> monofunE
  1608   monofun_lub_fun ~> adm_monofun [THEN admD]
  1609   cont_lub_fun ~> adm_cont [THEN admD]
  1610   cont2cont_Rep_CFun ~> cont2cont_APP
  1611   cont_Rep_CFun_app ~> cont_APP_app
  1612   cont_Rep_CFun_app_app ~> cont_APP_app_app
  1613   cont_cfun_fun ~> cont_Rep_cfun1 [THEN contE]
  1614   cont_cfun_arg ~> cont_Rep_cfun2 [THEN contE]
  1615   contlub_cfun ~> lub_APP [symmetric]
  1616   contlub_LAM ~> lub_LAM [symmetric]
  1617   thelubI ~> lub_eqI
  1618   UU_I ~> bottomI
  1619   lift_distinct1 ~> lift.distinct(1)
  1620   lift_distinct2 ~> lift.distinct(2)
  1621   Def_not_UU ~> lift.distinct(2)
  1622   Def_inject ~> lift.inject
  1623   below_UU_iff ~> below_bottom_iff
  1624   eq_UU_iff ~> eq_bottom_iff
  1625 
  1626 
  1627 *** Document preparation ***
  1628 
  1629 * Antiquotation @{rail} layouts railroad syntax diagrams, see also
  1630 isar-ref manual, both for description and actual application of the
  1631 same.
  1632 
  1633 * Antiquotation @{value} evaluates the given term and presents its
  1634 result.
  1635 
  1636 * Antiquotations: term style "isub" provides ad-hoc conversion of
  1637 variables x1, y23 into subscripted form x\<^isub>1,
  1638 y\<^isub>2\<^isub>3.
  1639 
  1640 * Predefined LaTeX macros for Isabelle symbols \<bind> and \<then>
  1641 (e.g. see ~~/src/HOL/Library/Monad_Syntax.thy).
  1642 
  1643 * Localized \isabellestyle switch can be used within blocks or groups
  1644 like this:
  1645 
  1646   \isabellestyle{it}  %preferred default
  1647   {\isabellestylett @{text "typewriter stuff"}}
  1648 
  1649 * Discontinued special treatment of hard tabulators.  Implicit
  1650 tab-width is now defined as 1.  Potential INCOMPATIBILITY for visual
  1651 layouts.
  1652 
  1653 
  1654 *** ML ***
  1655 
  1656 * The inner syntax of sort/type/term/prop supports inlined YXML
  1657 representations within quoted string tokens.  By encoding logical
  1658 entities via Term_XML (in ML or Scala) concrete syntax can be
  1659 bypassed, which is particularly useful for producing bits of text
  1660 under external program control.
  1661 
  1662 * Antiquotations for ML and document preparation are managed as theory
  1663 data, which requires explicit setup.
  1664 
  1665 * Isabelle_Process.is_active allows tools to check if the official
  1666 process wrapper is running (Isabelle/Scala/jEdit) or the old TTY loop
  1667 (better known as Proof General).
  1668 
  1669 * Structure Proof_Context follows standard naming scheme.  Old
  1670 ProofContext is still available for some time as legacy alias.
  1671 
  1672 * Structure Timing provides various operations for timing; supersedes
  1673 former start_timing/end_timing etc.
  1674 
  1675 * Path.print is the official way to show file-system paths to users
  1676 (including quotes etc.).
  1677 
  1678 * Inner syntax: identifiers in parse trees of generic categories
  1679 "logic", "aprop", "idt" etc. carry position information (disguised as
  1680 type constraints).  Occasional INCOMPATIBILITY with non-compliant
  1681 translations that choke on unexpected type constraints.  Positions can
  1682 be stripped in ML translations via Syntax.strip_positions /
  1683 Syntax.strip_positions_ast, or via the syntax constant
  1684 "_strip_positions" within parse trees.  As last resort, positions can
  1685 be disabled via the configuration option Syntax.positions, which is
  1686 called "syntax_positions" in Isar attribute syntax.
  1687 
  1688 * Discontinued special status of various ML structures that contribute
  1689 to structure Syntax (Ast, Lexicon, Mixfix, Parser, Printer etc.): less
  1690 pervasive content, no inclusion in structure Syntax.  INCOMPATIBILITY,
  1691 refer directly to Ast.Constant, Lexicon.is_identifier,
  1692 Syntax_Trans.mk_binder_tr etc.
  1693 
  1694 * Typed print translation: discontinued show_sorts argument, which is
  1695 already available via context of "advanced" translation.
  1696 
  1697 * Refined PARALLEL_GOALS tactical: degrades gracefully for schematic
  1698 goal states; body tactic needs to address all subgoals uniformly.
  1699 
  1700 * Slightly more special eq_list/eq_set, with shortcut involving
  1701 pointer equality (assumes that eq relation is reflexive).
  1702 
  1703 * Classical tactics use proper Proof.context instead of historic types
  1704 claset/clasimpset.  Old-style declarations like addIs, addEs, addDs
  1705 operate directly on Proof.context.  Raw type claset retains its use as
  1706 snapshot of the classical context, which can be recovered via
  1707 (put_claset HOL_cs) etc.  Type clasimpset has been discontinued.
  1708 INCOMPATIBILITY, classical tactics and derived proof methods require
  1709 proper Proof.context.
  1710 
  1711 
  1712 *** System ***
  1713 
  1714 * Discontinued support for Poly/ML 5.2, which was the last version
  1715 without proper multithreading and TimeLimit implementation.
  1716 
  1717 * Discontinued old lib/scripts/polyml-platform, which has been
  1718 obsolete since Isabelle2009-2.
  1719 
  1720 * Various optional external tools are referenced more robustly and
  1721 uniformly by explicit Isabelle settings as follows:
  1722 
  1723   ISABELLE_CSDP   (formerly CSDP_EXE)
  1724   ISABELLE_GHC    (formerly EXEC_GHC or GHC_PATH)
  1725   ISABELLE_OCAML  (formerly EXEC_OCAML)
  1726   ISABELLE_SWIPL  (formerly EXEC_SWIPL)
  1727   ISABELLE_YAP    (formerly EXEC_YAP)
  1728 
  1729 Note that automated detection from the file-system or search path has
  1730 been discontinued.  INCOMPATIBILITY.
  1731 
  1732 * Scala layer provides JVM method invocation service for static
  1733 methods of type (String)String, see Invoke_Scala.method in ML.  For
  1734 example:
  1735 
  1736   Invoke_Scala.method "java.lang.System.getProperty" "java.home"
  1737 
  1738 Together with YXML.string_of_body/parse_body and XML.Encode/Decode
  1739 this allows to pass structured values between ML and Scala.
  1740 
  1741 * The IsabelleText fonts includes some further glyphs to support the
  1742 Prover IDE.  Potential INCOMPATIBILITY: users who happen to have
  1743 installed a local copy (which is normally *not* required) need to
  1744 delete or update it from ~~/lib/fonts/.
  1745 
  1746 
  1747 
  1748 New in Isabelle2011 (January 2011)
  1749 ----------------------------------
  1750 
  1751 *** General ***
  1752 
  1753 * Experimental Prover IDE based on Isabelle/Scala and jEdit (see
  1754 src/Tools/jEdit).  This also serves as IDE for Isabelle/ML, with
  1755 useful tooltips and hyperlinks produced from its static analysis.  The
  1756 bundled component provides an executable Isabelle tool that can be run
  1757 like this:
  1758 
  1759   Isabelle2011/bin/isabelle jedit
  1760 
  1761 * Significantly improved Isabelle/Isar implementation manual.
  1762 
  1763 * System settings: ISABELLE_HOME_USER now includes ISABELLE_IDENTIFIER
  1764 (and thus refers to something like $HOME/.isabelle/Isabelle2011),
  1765 while the default heap location within that directory lacks that extra
  1766 suffix.  This isolates multiple Isabelle installations from each
  1767 other, avoiding problems with old settings in new versions.
  1768 INCOMPATIBILITY, need to copy/upgrade old user settings manually.
  1769 
  1770 * Source files are always encoded as UTF-8, instead of old-fashioned
  1771 ISO-Latin-1.  INCOMPATIBILITY.  Isabelle LaTeX documents might require
  1772 the following package declarations:
  1773 
  1774   \usepackage[utf8]{inputenc}
  1775   \usepackage{textcomp}
  1776 
  1777 * Explicit treatment of UTF-8 sequences as Isabelle symbols, such that
  1778 a Unicode character is treated as a single symbol, not a sequence of
  1779 non-ASCII bytes as before.  Since Isabelle/ML string literals may
  1780 contain symbols without further backslash escapes, Unicode can now be
  1781 used here as well.  Recall that Symbol.explode in ML provides a
  1782 consistent view on symbols, while raw explode (or String.explode)
  1783 merely give a byte-oriented representation.
  1784 
  1785 * Theory loader: source files are primarily located via the master
  1786 directory of each theory node (where the .thy file itself resides).
  1787 The global load path is still partially available as legacy feature.
  1788 Minor INCOMPATIBILITY due to subtle change in file lookup: use
  1789 explicit paths, relatively to the theory.
  1790 
  1791 * Special treatment of ML file names has been discontinued.
  1792 Historically, optional extensions .ML or .sml were added on demand --
  1793 at the cost of clarity of file dependencies.  Recall that Isabelle/ML
  1794 files exclusively use the .ML extension.  Minor INCOMPATIBILTY.
  1795 
  1796 * Various options that affect pretty printing etc. are now properly
  1797 handled within the context via configuration options, instead of
  1798 unsynchronized references or print modes.  There are both ML Config.T
  1799 entities and Isar declaration attributes to access these.
  1800 
  1801   ML (Config.T)                 Isar (attribute)
  1802 
  1803   eta_contract                  eta_contract
  1804   show_brackets                 show_brackets
  1805   show_sorts                    show_sorts
  1806   show_types                    show_types
  1807   show_question_marks           show_question_marks
  1808   show_consts                   show_consts
  1809   show_abbrevs                  show_abbrevs
  1810 
  1811   Syntax.ast_trace              syntax_ast_trace
  1812   Syntax.ast_stat               syntax_ast_stat
  1813   Syntax.ambiguity_level        syntax_ambiguity_level
  1814 
  1815   Goal_Display.goals_limit      goals_limit
  1816   Goal_Display.show_main_goal   show_main_goal
  1817 
  1818   Method.rule_trace             rule_trace
  1819 
  1820   Thy_Output.display            thy_output_display
  1821   Thy_Output.quotes             thy_output_quotes
  1822   Thy_Output.indent             thy_output_indent
  1823   Thy_Output.source             thy_output_source
  1824   Thy_Output.break              thy_output_break
  1825 
  1826 Note that corresponding "..._default" references in ML may only be
  1827 changed globally at the ROOT session setup, but *not* within a theory.
  1828 The option "show_abbrevs" supersedes the former print mode
  1829 "no_abbrevs" with inverted meaning.
  1830 
  1831 * More systematic naming of some configuration options.
  1832 INCOMPATIBILITY.
  1833 
  1834   trace_simp  ~>  simp_trace
  1835   debug_simp  ~>  simp_debug
  1836 
  1837 * Support for real valued configuration options, using simplistic
  1838 floating-point notation that coincides with the inner syntax for
  1839 float_token.
  1840 
  1841 * Support for real valued preferences (with approximative PGIP type):
  1842 front-ends need to accept "pgint" values in float notation.
  1843 INCOMPATIBILITY.
  1844 
  1845 * The IsabelleText font now includes Cyrillic, Hebrew, Arabic from
  1846 DejaVu Sans.
  1847 
  1848 * Discontinued support for Poly/ML 5.0 and 5.1 versions.
  1849 
  1850 
  1851 *** Pure ***
  1852 
  1853 * Command 'type_synonym' (with single argument) replaces somewhat
  1854 outdated 'types', which is still available as legacy feature for some
  1855 time.
  1856 
  1857 * Command 'nonterminal' (with 'and' separated list of arguments)
  1858 replaces somewhat outdated 'nonterminals'.  INCOMPATIBILITY.
  1859 
  1860 * Command 'notepad' replaces former 'example_proof' for
  1861 experimentation in Isar without any result.  INCOMPATIBILITY.
  1862 
  1863 * Locale interpretation commands 'interpret' and 'sublocale' accept
  1864 lists of equations to map definitions in a locale to appropriate
  1865 entities in the context of the interpretation.  The 'interpretation'
  1866 command already provided this functionality.
  1867 
  1868 * Diagnostic command 'print_dependencies' prints the locale instances
  1869 that would be activated if the specified expression was interpreted in
  1870 the current context.  Variant "print_dependencies!" assumes a context
  1871 without interpretations.
  1872 
  1873 * Diagnostic command 'print_interps' prints interpretations in proofs
  1874 in addition to interpretations in theories.
  1875 
  1876 * Discontinued obsolete 'global' and 'local' commands to manipulate
  1877 the theory name space.  Rare INCOMPATIBILITY.  The ML functions
  1878 Sign.root_path and Sign.local_path may be applied directly where this
  1879 feature is still required for historical reasons.
  1880 
  1881 * Discontinued obsolete 'constdefs' command.  INCOMPATIBILITY, use
  1882 'definition' instead.
  1883 
  1884 * The "prems" fact, which refers to the accidental collection of
  1885 foundational premises in the context, is now explicitly marked as
  1886 legacy feature and will be discontinued soon.  Consider using "assms"
  1887 of the head statement or reference facts by explicit names.
  1888 
  1889 * Document antiquotations @{class} and @{type} print classes and type
  1890 constructors.
  1891 
  1892 * Document antiquotation @{file} checks file/directory entries within
  1893 the local file system.
  1894 
  1895 
  1896 *** HOL ***
  1897 
  1898 * Coercive subtyping: functions can be declared as coercions and type
  1899 inference will add them as necessary upon input of a term.  Theory
  1900 Complex_Main declares real :: nat => real and real :: int => real as
  1901 coercions. A coercion function f is declared like this:
  1902 
  1903   declare [[coercion f]]
  1904 
  1905 To lift coercions through type constructors (e.g. from nat => real to
  1906 nat list => real list), map functions can be declared, e.g.
  1907 
  1908   declare [[coercion_map map]]
  1909 
  1910 Currently coercion inference is activated only in theories including
  1911 real numbers, i.e. descendants of Complex_Main.  This is controlled by
  1912 the configuration option "coercion_enabled", e.g. it can be enabled in
  1913 other theories like this:
  1914 
  1915   declare [[coercion_enabled]]
  1916 
  1917 * Command 'partial_function' provides basic support for recursive
  1918 function definitions over complete partial orders.  Concrete instances
  1919 are provided for i) the option type, ii) tail recursion on arbitrary
  1920 types, and iii) the heap monad of Imperative_HOL.  See
  1921 src/HOL/ex/Fundefs.thy and src/HOL/Imperative_HOL/ex/Linked_Lists.thy
  1922 for examples.
  1923 
  1924 * Function package: f.psimps rules are no longer implicitly declared
  1925 as [simp].  INCOMPATIBILITY.
  1926 
  1927 * Datatype package: theorems generated for executable equality (class
  1928 "eq") carry proper names and are treated as default code equations.
  1929 
  1930 * Inductive package: now offers command 'inductive_simps' to
  1931 automatically derive instantiated and simplified equations for
  1932 inductive predicates, similar to 'inductive_cases'.
  1933 
  1934 * Command 'enriched_type' allows to register properties of the
  1935 functorial structure of types.
  1936 
  1937 * Improved infrastructure for term evaluation using code generator
  1938 techniques, in particular static evaluation conversions.
  1939 
  1940 * Code generator: Scala (2.8 or higher) has been added to the target
  1941 languages.
  1942 
  1943 * Code generator: globbing constant expressions "*" and "Theory.*"
  1944 have been replaced by the more idiomatic "_" and "Theory._".
  1945 INCOMPATIBILITY.
  1946 
  1947 * Code generator: export_code without explicit file declaration prints
  1948 to standard output.  INCOMPATIBILITY.
  1949 
  1950 * Code generator: do not print function definitions for case
  1951 combinators any longer.
  1952 
  1953 * Code generator: simplification with rules determined with
  1954 src/Tools/Code/code_simp.ML and method "code_simp".
  1955 
  1956 * Code generator for records: more idiomatic representation of record
  1957 types.  Warning: records are not covered by ancient SML code
  1958 generation any longer.  INCOMPATIBILITY.  In cases of need, a suitable
  1959 rep_datatype declaration helps to succeed then:
  1960 
  1961   record 'a foo = ...
  1962   ...
  1963   rep_datatype foo_ext ...
  1964 
  1965 * Records: logical foundation type for records does not carry a
  1966 '_type' suffix any longer (obsolete due to authentic syntax).
  1967 INCOMPATIBILITY.
  1968 
  1969 * Quickcheck now by default uses exhaustive testing instead of random
  1970 testing.  Random testing can be invoked by "quickcheck [random]",
  1971 exhaustive testing by "quickcheck [exhaustive]".
  1972 
  1973 * Quickcheck instantiates polymorphic types with small finite
  1974 datatypes by default. This enables a simple execution mechanism to
  1975 handle quantifiers and function equality over the finite datatypes.
  1976 
  1977 * Quickcheck random generator has been renamed from "code" to
  1978 "random".  INCOMPATIBILITY.
  1979 
  1980 * Quickcheck now has a configurable time limit which is set to 30
  1981 seconds by default. This can be changed by adding [timeout = n] to the
  1982 quickcheck command. The time limit for Auto Quickcheck is still set
  1983 independently.
  1984 
  1985 * Quickcheck in locales considers interpretations of that locale for
  1986 counter example search.
  1987 
  1988 * Sledgehammer:
  1989   - Added "smt" and "remote_smt" provers based on the "smt" proof
  1990     method. See the Sledgehammer manual for details ("isabelle doc
  1991     sledgehammer").
  1992   - Renamed commands:
  1993     sledgehammer atp_info ~> sledgehammer running_provers
  1994     sledgehammer atp_kill ~> sledgehammer kill_provers
  1995     sledgehammer available_atps ~> sledgehammer available_provers
  1996     INCOMPATIBILITY.
  1997   - Renamed options:
  1998     sledgehammer [atps = ...] ~> sledgehammer [provers = ...]
  1999     sledgehammer [atp = ...] ~> sledgehammer [prover = ...]
  2000     sledgehammer [timeout = 77 s] ~> sledgehammer [timeout = 77]
  2001     (and "ms" and "min" are no longer supported)
  2002     INCOMPATIBILITY.
  2003 
  2004 * Nitpick:
  2005   - Renamed options:
  2006     nitpick [timeout = 77 s] ~> nitpick [timeout = 77]
  2007     nitpick [tac_timeout = 777 ms] ~> nitpick [tac_timeout = 0.777]
  2008     INCOMPATIBILITY.
  2009   - Added support for partial quotient types.
  2010   - Added local versions of the "Nitpick.register_xxx" functions.
  2011   - Added "whack" option.
  2012   - Allow registration of quotient types as codatatypes.
  2013   - Improved "merge_type_vars" option to merge more types.
  2014   - Removed unsound "fast_descrs" option.
  2015   - Added custom symmetry breaking for datatypes, making it possible to reach
  2016     higher cardinalities.
  2017   - Prevent the expansion of too large definitions.
  2018 
  2019 * Proof methods "metis" and "meson" now have configuration options
  2020 "meson_trace", "metis_trace", and "metis_verbose" that can be enabled
  2021 to diagnose these tools. E.g.
  2022 
  2023     using [[metis_trace = true]]
  2024 
  2025 * Auto Solve: Renamed "Auto Solve Direct".  The tool is now available
  2026 manually as command 'solve_direct'.
  2027 
  2028 * The default SMT solver Z3 must be enabled explicitly (due to
  2029 licensing issues) by setting the environment variable
  2030 Z3_NON_COMMERCIAL in etc/settings of the component, for example.  For
  2031 commercial applications, the SMT solver CVC3 is provided as fall-back;
  2032 changing the SMT solver is done via the configuration option
  2033 "smt_solver".
  2034 
  2035 * Remote SMT solvers need to be referred to by the "remote_" prefix,
  2036 i.e. "remote_cvc3" and "remote_z3".
  2037 
  2038 * Added basic SMT support for datatypes, records, and typedefs using
  2039 the oracle mode (no proofs).  Direct support of pairs has been dropped
  2040 in exchange (pass theorems fst_conv snd_conv pair_collapse to the SMT
  2041 support for a similar behavior).  Minor INCOMPATIBILITY.
  2042 
  2043 * Changed SMT configuration options:
  2044   - Renamed:
  2045     z3_proofs ~> smt_oracle (with inverted meaning)
  2046     z3_trace_assms ~> smt_trace_used_facts
  2047     INCOMPATIBILITY.
  2048   - Added:
  2049     smt_verbose
  2050     smt_random_seed
  2051     smt_datatypes
  2052     smt_infer_triggers
  2053     smt_monomorph_limit
  2054     cvc3_options
  2055     remote_cvc3_options
  2056     remote_z3_options
  2057     yices_options
  2058 
  2059 * Boogie output files (.b2i files) need to be declared in the theory
  2060 header.
  2061 
  2062 * Simplification procedure "list_to_set_comprehension" rewrites list
  2063 comprehensions applied to List.set to set comprehensions.  Occasional
  2064 INCOMPATIBILITY, may be deactivated like this:
  2065 
  2066   declare [[simproc del: list_to_set_comprehension]]
  2067 
  2068 * Removed old version of primrec package.  INCOMPATIBILITY.
  2069 
  2070 * Removed simplifier congruence rule of "prod_case", as has for long
  2071 been the case with "split".  INCOMPATIBILITY.
  2072 
  2073 * String.literal is a type, but not a datatype.  INCOMPATIBILITY.
  2074 
  2075 * Removed [split_format ... and ... and ...] version of
  2076 [split_format].  Potential INCOMPATIBILITY.
  2077 
  2078 * Predicate "sorted" now defined inductively, with nice induction
  2079 rules.  INCOMPATIBILITY: former sorted.simps now named sorted_simps.
  2080 
  2081 * Constant "contents" renamed to "the_elem", to free the generic name
  2082 contents for other uses.  INCOMPATIBILITY.
  2083 
  2084 * Renamed class eq and constant eq (for code generation) to class
  2085 equal and constant equal, plus renaming of related facts and various
  2086 tuning.  INCOMPATIBILITY.
  2087 
  2088 * Dropped type classes mult_mono and mult_mono1.  INCOMPATIBILITY.
  2089 
  2090 * Removed output syntax "'a ~=> 'b" for "'a => 'b option".
  2091 INCOMPATIBILITY.
  2092 
  2093 * Renamed theory Fset to Cset, type Fset.fset to Cset.set, in order to
  2094 avoid confusion with finite sets.  INCOMPATIBILITY.
  2095 
  2096 * Abandoned locales equiv, congruent and congruent2 for equivalence
  2097 relations.  INCOMPATIBILITY: use equivI rather than equiv_intro (same
  2098 for congruent(2)).
  2099 
  2100 * Some previously unqualified names have been qualified:
  2101 
  2102   types
  2103     bool ~> HOL.bool
  2104     nat ~> Nat.nat
  2105 
  2106   constants
  2107     Trueprop ~> HOL.Trueprop
  2108     True ~> HOL.True
  2109     False ~> HOL.False
  2110     op & ~> HOL.conj
  2111     op | ~> HOL.disj
  2112     op --> ~> HOL.implies
  2113     op = ~> HOL.eq
  2114     Not ~> HOL.Not
  2115     The ~> HOL.The
  2116     All ~> HOL.All
  2117     Ex ~> HOL.Ex
  2118     Ex1 ~> HOL.Ex1
  2119     Let ~> HOL.Let
  2120     If ~> HOL.If
  2121     Ball ~> Set.Ball
  2122     Bex ~> Set.Bex
  2123     Suc ~> Nat.Suc
  2124     Pair ~> Product_Type.Pair
  2125     fst ~> Product_Type.fst
  2126     snd ~> Product_Type.snd
  2127     curry ~> Product_Type.curry
  2128     op : ~> Set.member
  2129     Collect ~> Set.Collect
  2130 
  2131 INCOMPATIBILITY.
  2132 
  2133 * More canonical naming convention for some fundamental definitions:
  2134 
  2135     bot_bool_eq ~> bot_bool_def
  2136     top_bool_eq ~> top_bool_def
  2137     inf_bool_eq ~> inf_bool_def
  2138     sup_bool_eq ~> sup_bool_def
  2139     bot_fun_eq  ~> bot_fun_def
  2140     top_fun_eq  ~> top_fun_def
  2141     inf_fun_eq  ~> inf_fun_def
  2142     sup_fun_eq  ~> sup_fun_def
  2143 
  2144 INCOMPATIBILITY.
  2145 
  2146 * More stylized fact names:
  2147 
  2148   expand_fun_eq ~> fun_eq_iff
  2149   expand_set_eq ~> set_eq_iff
  2150   set_ext       ~> set_eqI
  2151   nat_number    ~> eval_nat_numeral
  2152 
  2153 INCOMPATIBILITY.
  2154 
  2155 * Refactoring of code-generation specific operations in theory List:
  2156 
  2157   constants
  2158     null ~> List.null
  2159 
  2160   facts
  2161     mem_iff ~> member_def
  2162     null_empty ~> null_def
  2163 
  2164 INCOMPATIBILITY.  Note that these were not supposed to be used
  2165 regularly unless for striking reasons; their main purpose was code
  2166 generation.
  2167 
  2168 Various operations from the Haskell prelude are used for generating
  2169 Haskell code.
  2170 
  2171 * Term "bij f" is now an abbreviation of "bij_betw f UNIV UNIV".  Term
  2172 "surj f" is now an abbreviation of "range f = UNIV".  The theorems
  2173 bij_def and surj_def are unchanged.  INCOMPATIBILITY.
  2174 
  2175 * Abolished some non-alphabetic type names: "prod" and "sum" replace
  2176 "*" and "+" respectively.  INCOMPATIBILITY.
  2177 
  2178 * Name "Plus" of disjoint sum operator "<+>" is now hidden.  Write
  2179 "Sum_Type.Plus" instead.
  2180 
  2181 * Constant "split" has been merged with constant "prod_case"; names of
  2182 ML functions, facts etc. involving split have been retained so far,
  2183 though.  INCOMPATIBILITY.
  2184 
  2185 * Dropped old infix syntax "_ mem _" for List.member; use "_ : set _"
  2186 instead.  INCOMPATIBILITY.
  2187 
  2188 * Removed lemma "Option.is_none_none" which duplicates "is_none_def".
  2189 INCOMPATIBILITY.
  2190 
  2191 * Former theory Library/Enum is now part of the HOL-Main image.
  2192 INCOMPATIBILITY: all constants of the Enum theory now have to be
  2193 referred to by its qualified name.
  2194 
  2195   enum    ~>  Enum.enum
  2196   nlists  ~>  Enum.nlists
  2197   product ~>  Enum.product
  2198 
  2199 * Theory Library/Monad_Syntax provides do-syntax for monad types.
  2200 Syntax in Library/State_Monad has been changed to avoid ambiguities.
  2201 INCOMPATIBILITY.
  2202 
  2203 * Theory Library/SetsAndFunctions has been split into
  2204 Library/Function_Algebras and Library/Set_Algebras; canonical names
  2205 for instance definitions for functions; various improvements.
  2206 INCOMPATIBILITY.
  2207 
  2208 * Theory Library/Multiset provides stable quicksort implementation of
  2209 sort_key.
  2210 
  2211 * Theory Library/Multiset: renamed empty_idemp ~> empty_neutral.
  2212 INCOMPATIBILITY.
  2213 
  2214 * Session Multivariate_Analysis: introduced a type class for euclidean
  2215 space.  Most theorems are now stated in terms of euclidean spaces
  2216 instead of finite cartesian products.
  2217 
  2218   types
  2219     real ^ 'n ~>  'a::real_vector
  2220               ~>  'a::euclidean_space
  2221               ~>  'a::ordered_euclidean_space
  2222         (depends on your needs)
  2223 
  2224   constants
  2225      _ $ _        ~> _ $$ _
  2226      \<chi> x. _  ~> \<chi>\<chi> x. _
  2227      CARD('n)     ~> DIM('a)
  2228 
  2229 Also note that the indices are now natural numbers and not from some
  2230 finite type. Finite cartesian products of euclidean spaces, products
  2231 of euclidean spaces the real and complex numbers are instantiated to
  2232 be euclidean_spaces.  INCOMPATIBILITY.
  2233 
  2234 * Session Probability: introduced pextreal as positive extended real
  2235 numbers.  Use pextreal as value for measures.  Introduce the
  2236 Radon-Nikodym derivative, product spaces and Fubini's theorem for
  2237 arbitrary sigma finite measures.  Introduces Lebesgue measure based on
  2238 the integral in Multivariate Analysis.  INCOMPATIBILITY.
  2239 
  2240 * Session Imperative_HOL: revamped, corrected dozens of inadequacies.
  2241 INCOMPATIBILITY.
  2242 
  2243 * Session SPARK (with image HOL-SPARK) provides commands to load and
  2244 prove verification conditions generated by the SPARK Ada program
  2245 verifier.  See also src/HOL/SPARK and src/HOL/SPARK/Examples.
  2246 
  2247 
  2248 *** HOL-Algebra ***
  2249 
  2250 * Theorems for additive ring operations (locale abelian_monoid and
  2251 descendants) are generated by interpretation from their multiplicative
  2252 counterparts.  Names (in particular theorem names) have the mandatory
  2253 qualifier 'add'.  Previous theorem names are redeclared for
  2254 compatibility.
  2255 
  2256 * Structure "int_ring" is now an abbreviation (previously a
  2257 definition).  This fits more natural with advanced interpretations.
  2258 
  2259 
  2260 *** HOLCF ***
  2261 
  2262 * The domain package now runs in definitional mode by default: The
  2263 former command 'new_domain' is now called 'domain'.  To use the domain
  2264 package in its original axiomatic mode, use 'domain (unsafe)'.
  2265 INCOMPATIBILITY.
  2266 
  2267 * The new class "domain" is now the default sort.  Class "predomain"
  2268 is an unpointed version of "domain". Theories can be updated by
  2269 replacing sort annotations as shown below.  INCOMPATIBILITY.
  2270 
  2271   'a::type ~> 'a::countable
  2272   'a::cpo  ~> 'a::predomain
  2273   'a::pcpo ~> 'a::domain
  2274 
  2275 * The old type class "rep" has been superseded by class "domain".
  2276 Accordingly, users of the definitional package must remove any
  2277 "default_sort rep" declarations.  INCOMPATIBILITY.
  2278 
  2279 * The domain package (definitional mode) now supports unpointed
  2280 predomain argument types, as long as they are marked 'lazy'. (Strict
  2281 arguments must be in class "domain".) For example, the following
  2282 domain definition now works:
  2283 
  2284   domain natlist = nil | cons (lazy "nat discr") (lazy "natlist")
  2285 
  2286 * Theory HOLCF/Library/HOL_Cpo provides cpo and predomain class
  2287 instances for types from main HOL: bool, nat, int, char, 'a + 'b,
  2288 'a option, and 'a list.  Additionally, it configures fixrec and the
  2289 domain package to work with these types.  For example:
  2290 
  2291   fixrec isInl :: "('a + 'b) u -> tr"
  2292     where "isInl$(up$(Inl x)) = TT" | "isInl$(up$(Inr y)) = FF"
  2293 
  2294   domain V = VFun (lazy "V -> V") | VCon (lazy "nat") (lazy "V list")
  2295 
  2296 * The "(permissive)" option of fixrec has been replaced with a
  2297 per-equation "(unchecked)" option. See
  2298 src/HOL/HOLCF/Tutorial/Fixrec_ex.thy for examples. INCOMPATIBILITY.
  2299 
  2300 * The "bifinite" class no longer fixes a constant "approx"; the class
  2301 now just asserts that such a function exists.  INCOMPATIBILITY.
  2302 
  2303 * Former type "alg_defl" has been renamed to "defl".  HOLCF no longer
  2304 defines an embedding of type 'a defl into udom by default; instances
  2305 of "bifinite" and "domain" classes are available in
  2306 src/HOL/HOLCF/Library/Defl_Bifinite.thy.
  2307 
  2308 * The syntax "REP('a)" has been replaced with "DEFL('a)".
  2309 
  2310 * The predicate "directed" has been removed.  INCOMPATIBILITY.
  2311 
  2312 * The type class "finite_po" has been removed.  INCOMPATIBILITY.
  2313 
  2314 * The function "cprod_map" has been renamed to "prod_map".
  2315 INCOMPATIBILITY.
  2316 
  2317 * The monadic bind operator on each powerdomain has new binder syntax
  2318 similar to sets, e.g. "\<Union>\<sharp>x\<in>xs. t" represents
  2319 "upper_bind\<cdot>xs\<cdot>(\<Lambda> x. t)".
  2320 
  2321 * The infix syntax for binary union on each powerdomain has changed
  2322 from e.g. "+\<sharp>" to "\<union>\<sharp>", for consistency with set
  2323 syntax.  INCOMPATIBILITY.
  2324 
  2325 * The constant "UU" has been renamed to "bottom".  The syntax "UU" is
  2326 still supported as an input translation.
  2327 
  2328 * Renamed some theorems (the original names are also still available).
  2329 
  2330   expand_fun_below   ~> fun_below_iff
  2331   below_fun_ext      ~> fun_belowI
  2332   expand_cfun_eq     ~> cfun_eq_iff
  2333   ext_cfun           ~> cfun_eqI
  2334   expand_cfun_below  ~> cfun_below_iff
  2335   below_cfun_ext     ~> cfun_belowI
  2336   cont2cont_Rep_CFun ~> cont2cont_APP
  2337 
  2338 * The Abs and Rep functions for various types have changed names.
  2339 Related theorem names have also changed to match. INCOMPATIBILITY.
  2340 
  2341   Rep_CFun  ~> Rep_cfun
  2342   Abs_CFun  ~> Abs_cfun
  2343   Rep_Sprod ~> Rep_sprod
  2344   Abs_Sprod ~> Abs_sprod
  2345   Rep_Ssum  ~> Rep_ssum
  2346   Abs_Ssum  ~> Abs_ssum
  2347 
  2348 * Lemmas with names of the form *_defined_iff or *_strict_iff have
  2349 been renamed to *_bottom_iff.  INCOMPATIBILITY.
  2350 
  2351 * Various changes to bisimulation/coinduction with domain package:
  2352 
  2353   - Definitions of "bisim" constants no longer mention definedness.
  2354   - With mutual recursion, "bisim" predicate is now curried.
  2355   - With mutual recursion, each type gets a separate coind theorem.
  2356   - Variable names in bisim_def and coinduct rules have changed.
  2357 
  2358 INCOMPATIBILITY.
  2359 
  2360 * Case combinators generated by the domain package for type "foo" are
  2361 now named "foo_case" instead of "foo_when".  INCOMPATIBILITY.
  2362 
  2363 * Several theorems have been renamed to more accurately reflect the
  2364 names of constants and types involved.  INCOMPATIBILITY.
  2365 
  2366   thelub_const    ~> lub_const
  2367   lub_const       ~> is_lub_const
  2368   thelubI         ~> lub_eqI
  2369   is_lub_lub      ~> is_lubD2
  2370   lubI            ~> is_lub_lub
  2371   unique_lub      ~> is_lub_unique
  2372   is_ub_lub       ~> is_lub_rangeD1
  2373   lub_bin_chain   ~> is_lub_bin_chain
  2374   lub_fun         ~> is_lub_fun
  2375   thelub_fun      ~> lub_fun
  2376   thelub_cfun     ~> lub_cfun
  2377   thelub_Pair     ~> lub_Pair
  2378   lub_cprod       ~> is_lub_prod
  2379   thelub_cprod    ~> lub_prod
  2380   minimal_cprod   ~> minimal_prod
  2381   inst_cprod_pcpo ~> inst_prod_pcpo
  2382   UU_I            ~> bottomI
  2383   compact_UU      ~> compact_bottom
  2384   deflation_UU    ~> deflation_bottom
  2385   finite_deflation_UU ~> finite_deflation_bottom
  2386 
  2387 * Many legacy theorem names have been discontinued.  INCOMPATIBILITY.
  2388 
  2389   sq_ord_less_eq_trans ~> below_eq_trans
  2390   sq_ord_eq_less_trans ~> eq_below_trans
  2391   refl_less            ~> below_refl
  2392   trans_less           ~> below_trans
  2393   antisym_less         ~> below_antisym
  2394   antisym_less_inverse ~> po_eq_conv [THEN iffD1]
  2395   box_less             ~> box_below
  2396   rev_trans_less       ~> rev_below_trans
  2397   not_less2not_eq      ~> not_below2not_eq
  2398   less_UU_iff          ~> below_UU_iff
  2399   flat_less_iff        ~> flat_below_iff
  2400   adm_less             ~> adm_below
  2401   adm_not_less         ~> adm_not_below
  2402   adm_compact_not_less ~> adm_compact_not_below
  2403   less_fun_def         ~> below_fun_def
  2404   expand_fun_less      ~> fun_below_iff
  2405   less_fun_ext         ~> fun_belowI
  2406   less_discr_def       ~> below_discr_def
  2407   discr_less_eq        ~> discr_below_eq
  2408   less_unit_def        ~> below_unit_def
  2409   less_cprod_def       ~> below_prod_def
  2410   prod_lessI           ~> prod_belowI
  2411   Pair_less_iff        ~> Pair_below_iff
  2412   fst_less_iff         ~> fst_below_iff
  2413   snd_less_iff         ~> snd_below_iff
  2414   expand_cfun_less     ~> cfun_below_iff
  2415   less_cfun_ext        ~> cfun_belowI
  2416   injection_less       ~> injection_below
  2417   less_up_def          ~> below_up_def
  2418   not_Iup_less         ~> not_Iup_below
  2419   Iup_less             ~> Iup_below
  2420   up_less              ~> up_below
  2421   Def_inject_less_eq   ~> Def_below_Def
  2422   Def_less_is_eq       ~> Def_below_iff
  2423   spair_less_iff       ~> spair_below_iff
  2424   less_sprod           ~> below_sprod
  2425   spair_less           ~> spair_below
  2426   sfst_less_iff        ~> sfst_below_iff
  2427   ssnd_less_iff        ~> ssnd_below_iff
  2428   fix_least_less       ~> fix_least_below
  2429   dist_less_one        ~> dist_below_one
  2430   less_ONE             ~> below_ONE
  2431   ONE_less_iff         ~> ONE_below_iff
  2432   less_sinlD           ~> below_sinlD
  2433   less_sinrD           ~> below_sinrD
  2434 
  2435 
  2436 *** FOL and ZF ***
  2437 
  2438 * All constant names are now qualified internally and use proper
  2439 identifiers, e.g. "IFOL.eq" instead of "op =".  INCOMPATIBILITY.
  2440 
  2441 
  2442 *** ML ***
  2443 
  2444 * Antiquotation @{assert} inlines a function bool -> unit that raises
  2445 Fail if the argument is false.  Due to inlining the source position of
  2446 failed assertions is included in the error output.
  2447 
  2448 * Discontinued antiquotation @{theory_ref}, which is obsolete since ML
  2449 text is in practice always evaluated with a stable theory checkpoint.
  2450 Minor INCOMPATIBILITY, use (Theory.check_thy @{theory}) instead.
  2451 
  2452 * Antiquotation @{theory A} refers to theory A from the ancestry of
  2453 the current context, not any accidental theory loader state as before.
  2454 Potential INCOMPATIBILITY, subtle change in semantics.
  2455 
  2456 * Syntax.pretty_priority (default 0) configures the required priority
  2457 of pretty-printed output and thus affects insertion of parentheses.
  2458 
  2459 * Syntax.default_root (default "any") configures the inner syntax
  2460 category (nonterminal symbol) for parsing of terms.
  2461 
  2462 * Former exception Library.UnequalLengths now coincides with
  2463 ListPair.UnequalLengths.
  2464 
  2465 * Renamed structure MetaSimplifier to Raw_Simplifier.  Note that the
  2466 main functionality is provided by structure Simplifier.
  2467 
  2468 * Renamed raw "explode" function to "raw_explode" to emphasize its
  2469 meaning.  Note that internally to Isabelle, Symbol.explode is used in
  2470 almost all situations.
  2471 
  2472 * Discontinued obsolete function sys_error and exception SYS_ERROR.
  2473 See implementation manual for further details on exceptions in
  2474 Isabelle/ML.
  2475 
  2476 * Renamed setmp_noncritical to Unsynchronized.setmp to emphasize its
  2477 meaning.
  2478 
  2479 * Renamed structure PureThy to Pure_Thy and moved most of its
  2480 operations to structure Global_Theory, to emphasize that this is
  2481 rarely-used global-only stuff.
  2482 
  2483 * Discontinued Output.debug.  Minor INCOMPATIBILITY, use plain writeln
  2484 instead (or tracing for high-volume output).
  2485 
  2486 * Configuration option show_question_marks only affects regular pretty
  2487 printing of types and terms, not raw Term.string_of_vname.
  2488 
  2489 * ML_Context.thm and ML_Context.thms are no longer pervasive.  Rare
  2490 INCOMPATIBILITY, superseded by static antiquotations @{thm} and
  2491 @{thms} for most purposes.
  2492 
  2493 * ML structure Unsynchronized is never opened, not even in Isar
  2494 interaction mode as before.  Old Unsynchronized.set etc. have been
  2495 discontinued -- use plain := instead.  This should be *rare* anyway,
  2496 since modern tools always work via official context data, notably
  2497 configuration options.
  2498 
  2499 * Parallel and asynchronous execution requires special care concerning
  2500 interrupts.  Structure Exn provides some convenience functions that
  2501 avoid working directly with raw Interrupt.  User code must not absorb
  2502 interrupts -- intermediate handling (for cleanup etc.) needs to be
  2503 followed by re-raising of the original exception.  Another common
  2504 source of mistakes are "handle _" patterns, which make the meaning of
  2505 the program subject to physical effects of the environment.
  2506 
  2507 
  2508 
  2509 New in Isabelle2009-2 (June 2010)
  2510 ---------------------------------
  2511 
  2512 *** General ***
  2513 
  2514 * Authentic syntax for *all* logical entities (type classes, type
  2515 constructors, term constants): provides simple and robust
  2516 correspondence between formal entities and concrete syntax.  Within
  2517 the parse tree / AST representations, "constants" are decorated by
  2518 their category (class, type, const) and spelled out explicitly with
  2519 their full internal name.
  2520 
  2521 Substantial INCOMPATIBILITY concerning low-level syntax declarations
  2522 and translations (translation rules and translation functions in ML).
  2523 Some hints on upgrading:
  2524 
  2525   - Many existing uses of 'syntax' and 'translations' can be replaced
  2526     by more modern 'type_notation', 'notation' and 'abbreviation',
  2527     which are independent of this issue.
  2528 
  2529   - 'translations' require markup within the AST; the term syntax
  2530     provides the following special forms:
  2531 
  2532       CONST c   -- produces syntax version of constant c from context
  2533       XCONST c  -- literally c, checked as constant from context
  2534       c         -- literally c, if declared by 'syntax'
  2535 
  2536     Plain identifiers are treated as AST variables -- occasionally the
  2537     system indicates accidental variables via the error "rhs contains
  2538     extra variables".
  2539 
  2540     Type classes and type constructors are marked according to their
  2541     concrete syntax.  Some old translations rules need to be written
  2542     for the "type" category, using type constructor application
  2543     instead of pseudo-term application of the default category
  2544     "logic".
  2545 
  2546   - 'parse_translation' etc. in ML may use the following
  2547     antiquotations:
  2548 
  2549       @{class_syntax c}   -- type class c within parse tree / AST
  2550       @{term_syntax c}    -- type constructor c within parse tree / AST
  2551       @{const_syntax c}   -- ML version of "CONST c" above
  2552       @{syntax_const c}   -- literally c (checked wrt. 'syntax' declarations)
  2553 
  2554   - Literal types within 'typed_print_translations', i.e. those *not*
  2555     represented as pseudo-terms are represented verbatim.  Use @{class
  2556     c} or @{type_name c} here instead of the above syntax
  2557     antiquotations.
  2558 
  2559 Note that old non-authentic syntax was based on unqualified base
  2560 names, so all of the above "constant" names would coincide.  Recall
  2561 that 'print_syntax' and ML_command "set Syntax.trace_ast" help to
  2562 diagnose syntax problems.
  2563 
  2564 * Type constructors admit general mixfix syntax, not just infix.
  2565 
  2566 * Concrete syntax may be attached to local entities without a proof
  2567 body, too.  This works via regular mixfix annotations for 'fix',
  2568 'def', 'obtain' etc. or via the explicit 'write' command, which is
  2569 similar to the 'notation' command in theory specifications.
  2570 
  2571 * Discontinued unnamed infix syntax (legacy feature for many years) --
  2572 need to specify constant name and syntax separately.  Internal ML
  2573 datatype constructors have been renamed from InfixName to Infix etc.
  2574 Minor INCOMPATIBILITY.
  2575 
  2576 * Schematic theorem statements need to be explicitly markup as such,
  2577 via commands 'schematic_lemma', 'schematic_theorem',
  2578 'schematic_corollary'.  Thus the relevance of the proof is made
  2579 syntactically clear, which impacts performance in a parallel or
  2580 asynchronous interactive environment.  Minor INCOMPATIBILITY.
  2581 
  2582 * Use of cumulative prems via "!" in some proof methods has been
  2583 discontinued (old legacy feature).
  2584 
  2585 * References 'trace_simp' and 'debug_simp' have been replaced by
  2586 configuration options stored in the context. Enabling tracing (the
  2587 case of debugging is similar) in proofs works via
  2588 
  2589   using [[trace_simp = true]]
  2590 
  2591 Tracing is then active for all invocations of the simplifier in
  2592 subsequent goal refinement steps. Tracing may also still be enabled or
  2593 disabled via the ProofGeneral settings menu.
  2594 
  2595 * Separate commands 'hide_class', 'hide_type', 'hide_const',
  2596 'hide_fact' replace the former 'hide' KIND command.  Minor
  2597 INCOMPATIBILITY.
  2598 
  2599 * Improved parallelism of proof term normalization: usedir -p2 -q0 is
  2600 more efficient than combinations with -q1 or -q2.
  2601 
  2602 
  2603 *** Pure ***
  2604 
  2605 * Proofterms record type-class reasoning explicitly, using the
  2606 "unconstrain" operation internally.  This eliminates all sort
  2607 constraints from a theorem and proof, introducing explicit
  2608 OFCLASS-premises.  On the proof term level, this operation is
  2609 automatically applied at theorem boundaries, such that closed proofs
  2610 are always free of sort constraints.  INCOMPATIBILITY for tools that
  2611 inspect proof terms.
  2612 
  2613 * Local theory specifications may depend on extra type variables that
  2614 are not present in the result type -- arguments TYPE('a) :: 'a itself
  2615 are added internally.  For example:
  2616 
  2617   definition unitary :: bool where "unitary = (ALL (x::'a) y. x = y)"
  2618 
  2619 * Predicates of locales introduced by classes carry a mandatory
  2620 "class" prefix.  INCOMPATIBILITY.
  2621 
  2622 * Vacuous class specifications observe default sort.  INCOMPATIBILITY.
  2623 
  2624 * Old 'axclass' command has been discontinued.  INCOMPATIBILITY, use
  2625 'class' instead.
  2626 
  2627 * Command 'code_reflect' allows to incorporate generated ML code into
  2628 runtime environment; replaces immature code_datatype antiquotation.
  2629 INCOMPATIBILITY.
  2630 
  2631 * Code generator: simple concept for abstract datatypes obeying
  2632 invariants.
  2633 
  2634 * Code generator: details of internal data cache have no impact on the
  2635 user space functionality any longer.
  2636 
  2637 * Methods "unfold_locales" and "intro_locales" ignore non-locale
  2638 subgoals.  This is more appropriate for interpretations with 'where'.
  2639 INCOMPATIBILITY.
  2640 
  2641 * Command 'example_proof' opens an empty proof body.  This allows to
  2642 experiment with Isar, without producing any persistent result.
  2643 
  2644 * Commands 'type_notation' and 'no_type_notation' declare type syntax
  2645 within a local theory context, with explicit checking of the
  2646 constructors involved (in contrast to the raw 'syntax' versions).
  2647 
  2648 * Commands 'types' and 'typedecl' now work within a local theory
  2649 context -- without introducing dependencies on parameters or
  2650 assumptions, which is not possible in Isabelle/Pure.
  2651 
  2652 * Command 'defaultsort' has been renamed to 'default_sort', it works
  2653 within a local theory context.  Minor INCOMPATIBILITY.
  2654 
  2655 
  2656 *** HOL ***
  2657 
  2658 * Command 'typedef' now works within a local theory context -- without
  2659 introducing dependencies on parameters or assumptions, which is not
  2660 possible in Isabelle/Pure/HOL.  Note that the logical environment may
  2661 contain multiple interpretations of local typedefs (with different
  2662 non-emptiness proofs), even in a global theory context.
  2663 
  2664 * New package for quotient types.  Commands 'quotient_type' and
  2665 'quotient_definition' may be used for defining types and constants by
  2666 quotient constructions.  An example is the type of integers created by
  2667 quotienting pairs of natural numbers:
  2668 
  2669   fun
  2670     intrel :: "(nat * nat) => (nat * nat) => bool"
  2671   where
  2672     "intrel (x, y) (u, v) = (x + v = u + y)"
  2673 
  2674   quotient_type int = "nat * nat" / intrel
  2675     by (auto simp add: equivp_def expand_fun_eq)
  2676 
  2677   quotient_definition
  2678     "0::int" is "(0::nat, 0::nat)"
  2679 
  2680 The method "lifting" can be used to lift of theorems from the
  2681 underlying "raw" type to the quotient type.  The example
  2682 src/HOL/Quotient_Examples/FSet.thy includes such a quotient
  2683 construction and provides a reasoning infrastructure for finite sets.
  2684 
  2685 * Renamed Library/Quotient.thy to Library/Quotient_Type.thy to avoid
  2686 clash with new theory Quotient in Main HOL.
  2687 
  2688 * Moved the SMT binding into the main HOL session, eliminating
  2689 separate HOL-SMT session.
  2690 
  2691 * List membership infix mem operation is only an input abbreviation.
  2692 INCOMPATIBILITY.
  2693 
  2694 * Theory Library/Word.thy has been removed.  Use library Word/Word.thy
  2695 for future developements; former Library/Word.thy is still present in
  2696 the AFP entry RSAPPS.
  2697 
  2698 * Theorem Int.int_induct renamed to Int.int_of_nat_induct and is no
  2699 longer shadowed.  INCOMPATIBILITY.
  2700 
  2701 * Dropped theorem duplicate comp_arith; use semiring_norm instead.
  2702 INCOMPATIBILITY.
  2703 
  2704 * Dropped theorem RealPow.real_sq_order; use power2_le_imp_le instead.
  2705 INCOMPATIBILITY.
  2706 
  2707 * Dropped normalizing_semiring etc; use the facts in semiring classes
  2708 instead.  INCOMPATIBILITY.
  2709 
  2710 * Dropped several real-specific versions of lemmas about floor and
  2711 ceiling; use the generic lemmas from theory "Archimedean_Field"
  2712 instead.  INCOMPATIBILITY.
  2713 
  2714   floor_number_of_eq         ~> floor_number_of
  2715   le_floor_eq_number_of      ~> number_of_le_floor
  2716   le_floor_eq_zero           ~> zero_le_floor
  2717   le_floor_eq_one            ~> one_le_floor
  2718   floor_less_eq_number_of    ~> floor_less_number_of
  2719   floor_less_eq_zero         ~> floor_less_zero
  2720   floor_less_eq_one          ~> floor_less_one
  2721   less_floor_eq_number_of    ~> number_of_less_floor
  2722   less_floor_eq_zero         ~> zero_less_floor
  2723   less_floor_eq_one          ~> one_less_floor
  2724   floor_le_eq_number_of      ~> floor_le_number_of
  2725   floor_le_eq_zero           ~> floor_le_zero
  2726   floor_le_eq_one            ~> floor_le_one
  2727   floor_subtract_number_of   ~> floor_diff_number_of
  2728   floor_subtract_one         ~> floor_diff_one
  2729   ceiling_number_of_eq       ~> ceiling_number_of
  2730   ceiling_le_eq_number_of    ~> ceiling_le_number_of
  2731   ceiling_le_zero_eq         ~> ceiling_le_zero
  2732   ceiling_le_eq_one          ~> ceiling_le_one
  2733   less_ceiling_eq_number_of  ~> number_of_less_ceiling
  2734   less_ceiling_eq_zero       ~> zero_less_ceiling
  2735   less_ceiling_eq_one        ~> one_less_ceiling
  2736   ceiling_less_eq_number_of  ~> ceiling_less_number_of
  2737   ceiling_less_eq_zero       ~> ceiling_less_zero
  2738   ceiling_less_eq_one        ~> ceiling_less_one
  2739   le_ceiling_eq_number_of    ~> number_of_le_ceiling
  2740   le_ceiling_eq_zero         ~> zero_le_ceiling
  2741   le_ceiling_eq_one          ~> one_le_ceiling
  2742   ceiling_subtract_number_of ~> ceiling_diff_number_of
  2743   ceiling_subtract_one       ~> ceiling_diff_one
  2744 
  2745 * Theory "Finite_Set": various folding_XXX locales facilitate the
  2746 application of the various fold combinators on finite sets.
  2747 
  2748 * Library theory "RBT" renamed to "RBT_Impl"; new library theory "RBT"
  2749 provides abstract red-black tree type which is backed by "RBT_Impl" as
  2750 implementation.  INCOMPATIBILTY.
  2751 
  2752 * Theory Library/Coinductive_List has been removed -- superseded by
  2753 AFP/thys/Coinductive.
  2754 
  2755 * Theory PReal, including the type "preal" and related operations, has
  2756 been removed.  INCOMPATIBILITY.
  2757 
  2758 * Real: new development using Cauchy Sequences.
  2759 
  2760 * Split off theory "Big_Operators" containing setsum, setprod,
  2761 Inf_fin, Sup_fin, Min, Max from theory Finite_Set.  INCOMPATIBILITY.
  2762 
  2763 * Theory "Rational" renamed to "Rat", for consistency with "Nat",
  2764 "Int" etc.  INCOMPATIBILITY.
  2765 
  2766 * Constant Rat.normalize needs to be qualified.  INCOMPATIBILITY.
  2767 
  2768 * New set of rules "ac_simps" provides combined assoc / commute
  2769 rewrites for all interpretations of the appropriate generic locales.
  2770 
  2771 * Renamed theory "OrderedGroup" to "Groups" and split theory
  2772 "Ring_and_Field" into theories "Rings" and "Fields"; for more
  2773 appropriate and more consistent names suitable for name prefixes
  2774 within the HOL theories.  INCOMPATIBILITY.
  2775 
  2776 * Some generic constants have been put to appropriate theories:
  2777   - less_eq, less: Orderings
  2778   - zero, one, plus, minus, uminus, times, abs, sgn: Groups
  2779   - inverse, divide: Rings
  2780 INCOMPATIBILITY.
  2781 
  2782 * More consistent naming of type classes involving orderings (and
  2783 lattices):
  2784 
  2785     lower_semilattice                   ~> semilattice_inf
  2786     upper_semilattice                   ~> semilattice_sup
  2787 
  2788     dense_linear_order                  ~> dense_linorder
  2789 
  2790     pordered_ab_group_add               ~> ordered_ab_group_add
  2791     pordered_ab_group_add_abs           ~> ordered_ab_group_add_abs
  2792     pordered_ab_semigroup_add           ~> ordered_ab_semigroup_add
  2793     pordered_ab_semigroup_add_imp_le    ~> ordered_ab_semigroup_add_imp_le
  2794     pordered_cancel_ab_semigroup_add    ~> ordered_cancel_ab_semigroup_add
  2795     pordered_cancel_comm_semiring       ~> ordered_cancel_comm_semiring
  2796     pordered_cancel_semiring            ~> ordered_cancel_semiring
  2797     pordered_comm_monoid_add            ~> ordered_comm_monoid_add
  2798     pordered_comm_ring                  ~> ordered_comm_ring
  2799     pordered_comm_semiring              ~> ordered_comm_semiring
  2800     pordered_ring                       ~> ordered_ring
  2801     pordered_ring_abs                   ~> ordered_ring_abs
  2802     pordered_semiring                   ~> ordered_semiring
  2803 
  2804     ordered_ab_group_add                ~> linordered_ab_group_add
  2805     ordered_ab_semigroup_add            ~> linordered_ab_semigroup_add
  2806     ordered_cancel_ab_semigroup_add     ~> linordered_cancel_ab_semigroup_add
  2807     ordered_comm_semiring_strict        ~> linordered_comm_semiring_strict
  2808     ordered_field                       ~> linordered_field
  2809     ordered_field_no_lb                 ~> linordered_field_no_lb
  2810     ordered_field_no_ub                 ~> linordered_field_no_ub
  2811     ordered_field_dense_linear_order    ~> dense_linordered_field
  2812     ordered_idom                        ~> linordered_idom
  2813     ordered_ring                        ~> linordered_ring
  2814     ordered_ring_le_cancel_factor       ~> linordered_ring_le_cancel_factor
  2815     ordered_ring_less_cancel_factor     ~> linordered_ring_less_cancel_factor
  2816     ordered_ring_strict                 ~> linordered_ring_strict
  2817     ordered_semidom                     ~> linordered_semidom
  2818     ordered_semiring                    ~> linordered_semiring
  2819     ordered_semiring_1                  ~> linordered_semiring_1
  2820     ordered_semiring_1_strict           ~> linordered_semiring_1_strict
  2821     ordered_semiring_strict             ~> linordered_semiring_strict
  2822 
  2823   The following slightly odd type classes have been moved to a
  2824   separate theory Library/Lattice_Algebras:
  2825 
  2826     lordered_ab_group_add               ~> lattice_ab_group_add
  2827     lordered_ab_group_add_abs           ~> lattice_ab_group_add_abs
  2828     lordered_ab_group_add_meet          ~> semilattice_inf_ab_group_add
  2829     lordered_ab_group_add_join          ~> semilattice_sup_ab_group_add
  2830     lordered_ring                       ~> lattice_ring
  2831 
  2832 INCOMPATIBILITY.
  2833 
  2834 * Refined field classes:
  2835   - classes division_ring_inverse_zero, field_inverse_zero,
  2836     linordered_field_inverse_zero include rule inverse 0 = 0 --
  2837     subsumes former division_by_zero class;
  2838   - numerous lemmas have been ported from field to division_ring.
  2839 INCOMPATIBILITY.
  2840 
  2841 * Refined algebra theorem collections:
  2842   - dropped theorem group group_simps, use algebra_simps instead;
  2843   - dropped theorem group ring_simps, use field_simps instead;
  2844   - proper theorem collection field_simps subsumes former theorem
  2845     groups field_eq_simps and field_simps;
  2846   - dropped lemma eq_minus_self_iff which is a duplicate for
  2847     equal_neg_zero.
  2848 INCOMPATIBILITY.
  2849 
  2850 * Theory Finite_Set and List: some lemmas have been generalized from
  2851 sets to lattices:
  2852 
  2853   fun_left_comm_idem_inter      ~> fun_left_comm_idem_inf
  2854   fun_left_comm_idem_union      ~> fun_left_comm_idem_sup
  2855   inter_Inter_fold_inter        ~> inf_Inf_fold_inf
  2856   union_Union_fold_union        ~> sup_Sup_fold_sup
  2857   Inter_fold_inter              ~> Inf_fold_inf
  2858   Union_fold_union              ~> Sup_fold_sup
  2859   inter_INTER_fold_inter        ~> inf_INFI_fold_inf
  2860   union_UNION_fold_union        ~> sup_SUPR_fold_sup
  2861   INTER_fold_inter              ~> INFI_fold_inf
  2862   UNION_fold_union              ~> SUPR_fold_sup
  2863 
  2864 * Theory "Complete_Lattice": lemmas top_def and bot_def have been
  2865 replaced by the more convenient lemmas Inf_empty and Sup_empty.
  2866 Dropped lemmas Inf_insert_simp and Sup_insert_simp, which are subsumed
  2867 by Inf_insert and Sup_insert.  Lemmas Inf_UNIV and Sup_UNIV replace
  2868 former Inf_Univ and Sup_Univ.  Lemmas inf_top_right and sup_bot_right
  2869 subsume inf_top and sup_bot respectively.  INCOMPATIBILITY.
  2870 
  2871 * Reorganized theory Multiset: swapped notation of pointwise and
  2872 multiset order:
  2873 
  2874   - pointwise ordering is instance of class order with standard syntax
  2875     <= and <;
  2876   - multiset ordering has syntax <=# and <#; partial order properties
  2877     are provided by means of interpretation with prefix
  2878     multiset_order;
  2879   - less duplication, less historical organization of sections,
  2880     conversion from associations lists to multisets, rudimentary code
  2881     generation;
  2882   - use insert_DiffM2 [symmetric] instead of elem_imp_eq_diff_union,
  2883     if needed.
  2884 
  2885 Renamed:
  2886 
  2887   multiset_eq_conv_count_eq  ~>  multiset_ext_iff
  2888   multi_count_ext  ~>  multiset_ext
  2889   diff_union_inverse2  ~>  diff_union_cancelR
  2890 
  2891 INCOMPATIBILITY.
  2892 
  2893 * Theory Permutation: replaced local "remove" by List.remove1.
  2894 
  2895 * Code generation: ML and OCaml code is decorated with signatures.
  2896 
  2897 * Theory List: added transpose.
  2898 
  2899 * Library/Nat_Bijection.thy is a collection of bijective functions
  2900 between nat and other types, which supersedes the older libraries
  2901 Library/Nat_Int_Bij.thy and HOLCF/NatIso.thy.  INCOMPATIBILITY.
  2902 
  2903   Constants:
  2904   Nat_Int_Bij.nat2_to_nat         ~> prod_encode
  2905   Nat_Int_Bij.nat_to_nat2         ~> prod_decode
  2906   Nat_Int_Bij.int_to_nat_bij      ~> int_encode
  2907   Nat_Int_Bij.nat_to_int_bij      ~> int_decode
  2908   Countable.pair_encode           ~> prod_encode
  2909   NatIso.prod2nat                 ~> prod_encode
  2910   NatIso.nat2prod                 ~> prod_decode
  2911   NatIso.sum2nat                  ~> sum_encode
  2912   NatIso.nat2sum                  ~> sum_decode
  2913   NatIso.list2nat                 ~> list_encode
  2914   NatIso.nat2list                 ~> list_decode
  2915   NatIso.set2nat                  ~> set_encode
  2916   NatIso.nat2set                  ~> set_decode
  2917 
  2918   Lemmas:
  2919   Nat_Int_Bij.bij_nat_to_int_bij  ~> bij_int_decode
  2920   Nat_Int_Bij.nat2_to_nat_inj     ~> inj_prod_encode
  2921   Nat_Int_Bij.nat2_to_nat_surj    ~> surj_prod_encode
  2922   Nat_Int_Bij.nat_to_nat2_inj     ~> inj_prod_decode
  2923   Nat_Int_Bij.nat_to_nat2_surj    ~> surj_prod_decode
  2924   Nat_Int_Bij.i2n_n2i_id          ~> int_encode_inverse
  2925   Nat_Int_Bij.n2i_i2n_id          ~> int_decode_inverse
  2926   Nat_Int_Bij.surj_nat_to_int_bij ~> surj_int_encode
  2927   Nat_Int_Bij.surj_int_to_nat_bij ~> surj_int_decode
  2928   Nat_Int_Bij.inj_nat_to_int_bij  ~> inj_int_encode
  2929   Nat_Int_Bij.inj_int_to_nat_bij  ~> inj_int_decode
  2930   Nat_Int_Bij.bij_nat_to_int_bij  ~> bij_int_encode
  2931   Nat_Int_Bij.bij_int_to_nat_bij  ~> bij_int_decode
  2932 
  2933 * Sledgehammer:
  2934   - Renamed ATP commands:
  2935     atp_info     ~> sledgehammer running_atps
  2936     atp_kill     ~> sledgehammer kill_atps
  2937     atp_messages ~> sledgehammer messages
  2938     atp_minimize ~> sledgehammer minimize
  2939     print_atps   ~> sledgehammer available_atps
  2940     INCOMPATIBILITY.
  2941   - Added user's manual ("isabelle doc sledgehammer").
  2942   - Added option syntax and "sledgehammer_params" to customize
  2943     Sledgehammer's behavior.  See the manual for details.
  2944   - Modified the Isar proof reconstruction code so that it produces
  2945     direct proofs rather than proofs by contradiction.  (This feature
  2946     is still experimental.)
  2947   - Made Isar proof reconstruction work for SPASS, remote ATPs, and in
  2948     full-typed mode.
  2949   - Added support for TPTP syntax for SPASS via the "spass_tptp" ATP.
  2950 
  2951 * Nitpick:
  2952   - Added and implemented "binary_ints" and "bits" options.
  2953   - Added "std" option and implemented support for nonstandard models.
  2954   - Added and implemented "finitize" option to improve the precision
  2955     of infinite datatypes based on a monotonicity analysis.
  2956   - Added support for quotient types.
  2957   - Added support for "specification" and "ax_specification"
  2958     constructs.
  2959   - Added support for local definitions (for "function" and
  2960     "termination" proofs).
  2961   - Added support for term postprocessors.
  2962   - Optimized "Multiset.multiset" and "FinFun.finfun".
  2963   - Improved efficiency of "destroy_constrs" optimization.
  2964   - Fixed soundness bugs related to "destroy_constrs" optimization and
  2965     record getters.
  2966   - Fixed soundness bug related to higher-order constructors.
  2967   - Fixed soundness bug when "full_descrs" is enabled.
  2968   - Improved precision of set constructs.
  2969   - Added "atoms" option.
  2970   - Added cache to speed up repeated Kodkod invocations on the same
  2971     problems.
  2972   - Renamed "MiniSatJNI", "zChaffJNI", "BerkMinAlloy", and
  2973     "SAT4JLight" to "MiniSat_JNI", "zChaff_JNI", "BerkMin_Alloy", and
  2974     "SAT4J_Light".  INCOMPATIBILITY.
  2975   - Removed "skolemize", "uncurry", "sym_break", "flatten_prop",
  2976     "sharing_depth", and "show_skolems" options.  INCOMPATIBILITY.
  2977   - Removed "nitpick_intro" attribute.  INCOMPATIBILITY.
  2978 
  2979 * Method "induct" now takes instantiations of the form t, where t is not
  2980   a variable, as a shorthand for "x == t", where x is a fresh variable.
  2981   If this is not intended, t has to be enclosed in parentheses.
  2982   By default, the equalities generated by definitional instantiations
  2983   are pre-simplified, which may cause parameters of inductive cases
  2984   to disappear, or may even delete some of the inductive cases.
  2985   Use "induct (no_simp)" instead of "induct" to restore the old
  2986   behaviour. The (no_simp) option is also understood by the "cases"
  2987   and "nominal_induct" methods, which now perform pre-simplification, too.
  2988   INCOMPATIBILITY.
  2989 
  2990 
  2991 *** HOLCF ***
  2992 
  2993 * Variable names in lemmas generated by the domain package have
  2994 changed; the naming scheme is now consistent with the HOL datatype
  2995 package.  Some proof scripts may be affected, INCOMPATIBILITY.
  2996 
  2997 * The domain package no longer defines the function "foo_copy" for
  2998 recursive domain "foo".  The reach lemma is now stated directly in
  2999 terms of "foo_take".  Lemmas and proofs that mention "foo_copy" must
  3000 be reformulated in terms of "foo_take", INCOMPATIBILITY.
  3001 
  3002 * Most definedness lemmas generated by the domain package (previously
  3003 of the form "x ~= UU ==> foo$x ~= UU") now have an if-and-only-if form
  3004 like "foo$x = UU <-> x = UU", which works better as a simp rule.
  3005 Proofs that used definedness lemmas as intro rules may break,
  3006 potential INCOMPATIBILITY.
  3007 
  3008 * Induction and casedist rules generated by the domain package now
  3009 declare proper case_names (one called "bottom", and one named for each
  3010 constructor).  INCOMPATIBILITY.
  3011 
  3012 * For mutually-recursive domains, separate "reach" and "take_lemma"
  3013 rules are generated for each domain, INCOMPATIBILITY.
  3014 
  3015   foo_bar.reach       ~> foo.reach  bar.reach
  3016   foo_bar.take_lemmas ~> foo.take_lemma  bar.take_lemma
  3017 
  3018 * Some lemmas generated by the domain package have been renamed for
  3019 consistency with the datatype package, INCOMPATIBILITY.
  3020 
  3021   foo.ind        ~> foo.induct
  3022   foo.finite_ind ~> foo.finite_induct
  3023   foo.coind      ~> foo.coinduct
  3024   foo.casedist   ~> foo.exhaust
  3025   foo.exhaust    ~> foo.nchotomy
  3026 
  3027 * For consistency with other definition packages, the fixrec package
  3028 now generates qualified theorem names, INCOMPATIBILITY.
  3029 
  3030   foo_simps  ~> foo.simps
  3031   foo_unfold ~> foo.unfold
  3032   foo_induct ~> foo.induct
  3033 
  3034 * The "fixrec_simp" attribute has been removed.  The "fixrec_simp"
  3035 method and internal fixrec proofs now use the default simpset instead.
  3036 INCOMPATIBILITY.
  3037 
  3038 * The "contlub" predicate has been removed.  Proof scripts should use
  3039 lemma contI2 in place of monocontlub2cont, INCOMPATIBILITY.
  3040 
  3041 * The "admw" predicate has been removed, INCOMPATIBILITY.
  3042 
  3043 * The constants cpair, cfst, and csnd have been removed in favor of
  3044 Pair, fst, and snd from Isabelle/HOL, INCOMPATIBILITY.
  3045 
  3046 
  3047 *** ML ***
  3048 
  3049 * Antiquotations for basic formal entities:
  3050 
  3051     @{class NAME}         -- type class
  3052     @{class_syntax NAME}  -- syntax representation of the above
  3053 
  3054     @{type_name NAME}     -- logical type
  3055     @{type_abbrev NAME}   -- type abbreviation
  3056     @{nonterminal NAME}   -- type of concrete syntactic category
  3057     @{type_syntax NAME}   -- syntax representation of any of the above
  3058 
  3059     @{const_name NAME}    -- logical constant (INCOMPATIBILITY)
  3060     @{const_abbrev NAME}  -- abbreviated constant
  3061     @{const_syntax NAME}  -- syntax representation of any of the above
  3062 
  3063 * Antiquotation @{syntax_const NAME} ensures that NAME refers to a raw
  3064 syntax constant (cf. 'syntax' command).
  3065 
  3066 * Antiquotation @{make_string} inlines a function to print arbitrary
  3067 values similar to the ML toplevel.  The result is compiler dependent
  3068 and may fall back on "?" in certain situations.
  3069 
  3070 * Diagnostic commands 'ML_val' and 'ML_command' may refer to
  3071 antiquotations @{Isar.state} and @{Isar.goal}.  This replaces impure
  3072 Isar.state() and Isar.goal(), which belong to the old TTY loop and do
  3073 not work with the asynchronous Isar document model.
  3074 
  3075 * Configuration options now admit dynamic default values, depending on
  3076 the context or even global references.
  3077 
  3078 * SHA1.digest digests strings according to SHA-1 (see RFC 3174).  It
  3079 uses an efficient external library if available (for Poly/ML).
  3080 
  3081 * Renamed some important ML structures, while keeping the old names
  3082 for some time as aliases within the structure Legacy:
  3083 
  3084   OuterKeyword  ~>  Keyword
  3085   OuterLex      ~>  Token
  3086   OuterParse    ~>  Parse
  3087   OuterSyntax   ~>  Outer_Syntax
  3088   PrintMode     ~>  Print_Mode
  3089   SpecParse     ~>  Parse_Spec
  3090   ThyInfo       ~>  Thy_Info
  3091   ThyLoad       ~>  Thy_Load
  3092   ThyOutput     ~>  Thy_Output
  3093   TypeInfer     ~>  Type_Infer
  3094 
  3095 Note that "open Legacy" simplifies porting of sources, but forgetting
  3096 to remove it again will complicate porting again in the future.
  3097 
  3098 * Most operations that refer to a global context are named
  3099 accordingly, e.g. Simplifier.global_context or
  3100 ProofContext.init_global.  There are some situations where a global
  3101 context actually works, but under normal circumstances one needs to
  3102 pass the proper local context through the code!
  3103 
  3104 * Discontinued old TheoryDataFun with its copy/init operation -- data
  3105 needs to be pure.  Functor Theory_Data_PP retains the traditional
  3106 Pretty.pp argument to merge, which is absent in the standard
  3107 Theory_Data version.
  3108 
  3109 * Sorts.certify_sort and derived "cert" operations for types and terms
  3110 no longer minimize sorts.  Thus certification at the boundary of the
  3111 inference kernel becomes invariant under addition of class relations,
  3112 which is an important monotonicity principle.  Sorts are now minimized
  3113 in the syntax layer only, at the boundary between the end-user and the
  3114 system.  Subtle INCOMPATIBILITY, may have to use Sign.minimize_sort
  3115 explicitly in rare situations.
  3116 
  3117 * Renamed old-style Drule.standard to Drule.export_without_context, to
  3118 emphasize that this is in no way a standard operation.
  3119 INCOMPATIBILITY.
  3120 
  3121 * Subgoal.FOCUS (and variants): resulting goal state is normalized as
  3122 usual for resolution.  Rare INCOMPATIBILITY.
  3123 
  3124 * Renamed varify/unvarify operations to varify_global/unvarify_global
  3125 to emphasize that these only work in a global situation (which is
  3126 quite rare).
  3127 
  3128 * Curried take and drop in library.ML; negative length is interpreted
  3129 as infinity (as in chop).  Subtle INCOMPATIBILITY.
  3130 
  3131 * Proof terms: type substitutions on proof constants now use canonical
  3132 order of type variables.  INCOMPATIBILITY for tools working with proof
  3133 terms.
  3134 
  3135 * Raw axioms/defs may no longer carry sort constraints, and raw defs
  3136 may no longer carry premises.  User-level specifications are
  3137 transformed accordingly by Thm.add_axiom/add_def.
  3138 
  3139 
  3140 *** System ***
  3141 
  3142 * Discontinued special HOL_USEDIR_OPTIONS for the main HOL image;
  3143 ISABELLE_USEDIR_OPTIONS applies uniformly to all sessions.  Note that
  3144 proof terms are enabled unconditionally in the new HOL-Proofs image.
  3145 
  3146 * Discontinued old ISABELLE and ISATOOL environment settings (legacy
  3147 feature since Isabelle2009).  Use ISABELLE_PROCESS and ISABELLE_TOOL,
  3148 respectively.
  3149 
  3150 * Old lib/scripts/polyml-platform is superseded by the
  3151 ISABELLE_PLATFORM setting variable, which defaults to the 32 bit
  3152 variant, even on a 64 bit machine.  The following example setting
  3153 prefers 64 bit if available:
  3154 
  3155   ML_PLATFORM="${ISABELLE_PLATFORM64:-$ISABELLE_PLATFORM}"
  3156 
  3157 * The preliminary Isabelle/jEdit application demonstrates the emerging
  3158 Isabelle/Scala layer for advanced prover interaction and integration.
  3159 See src/Tools/jEdit or "isabelle jedit" provided by the properly built
  3160 component.
  3161 
  3162 * "IsabelleText" is a Unicode font derived from Bitstream Vera Mono
  3163 and Bluesky TeX fonts.  It provides the usual Isabelle symbols,
  3164 similar to the default assignment of the document preparation system
  3165 (cf. isabellesym.sty).  The Isabelle/Scala class Isabelle_System
  3166 provides some operations for direct access to the font without asking
  3167 the user for manual installation.
  3168 
  3169 
  3170 
  3171 New in Isabelle2009-1 (December 2009)
  3172 -------------------------------------
  3173 
  3174 *** General ***
  3175 
  3176 * Discontinued old form of "escaped symbols" such as \\<forall>.  Only
  3177 one backslash should be used, even in ML sources.
  3178 
  3179 
  3180 *** Pure ***
  3181 
  3182 * Locale interpretation propagates mixins along the locale hierarchy.
  3183 The currently only available mixins are the equations used to map
  3184 local definitions to terms of the target domain of an interpretation.
  3185 
  3186 * Reactivated diagnostic command 'print_interps'.  Use "print_interps
  3187 loc" to print all interpretations of locale "loc" in the theory.
  3188 Interpretations in proofs are not shown.
  3189 
  3190 * Thoroughly revised locales tutorial.  New section on conditional
  3191 interpretation.
  3192 
  3193 * On instantiation of classes, remaining undefined class parameters
  3194 are formally declared.  INCOMPATIBILITY.
  3195 
  3196 
  3197 *** Document preparation ***
  3198 
  3199 * New generalized style concept for printing terms: @{foo (style) ...}
  3200 instead of @{foo_style style ...}  (old form is still retained for
  3201 backward compatibility).  Styles can be also applied for
  3202 antiquotations prop, term_type and typeof.
  3203 
  3204 
  3205 *** HOL ***
  3206 
  3207 * New proof method "smt" for a combination of first-order logic with
  3208 equality, linear and nonlinear (natural/integer/real) arithmetic, and
  3209 fixed-size bitvectors; there is also basic support for higher-order
  3210 features (esp. lambda abstractions).  It is an incomplete decision
  3211 procedure based on external SMT solvers using the oracle mechanism;
  3212 for the SMT solver Z3, this method is proof-producing.  Certificates
  3213 are provided to avoid calling the external solvers solely for
  3214 re-checking proofs.  Due to a remote SMT service there is no need for
  3215 installing SMT solvers locally.  See src/HOL/SMT.
  3216 
  3217 * New commands to load and prove verification conditions generated by
  3218 the Boogie program verifier or derived systems (e.g. the Verifying C
  3219 Compiler (VCC) or Spec#).  See src/HOL/Boogie.
  3220 
  3221 * New counterexample generator tool 'nitpick' based on the Kodkod
  3222 relational model finder.  See src/HOL/Tools/Nitpick and
  3223 src/HOL/Nitpick_Examples.
  3224 
  3225 * New commands 'code_pred' and 'values' to invoke the predicate
  3226 compiler and to enumerate values of inductive predicates.
  3227 
  3228 * A tabled implementation of the reflexive transitive closure.
  3229 
  3230 * New implementation of quickcheck uses generic code generator;
  3231 default generators are provided for all suitable HOL types, records
  3232 and datatypes.  Old quickcheck can be re-activated importing theory
  3233 Library/SML_Quickcheck.
  3234 
  3235 * New testing tool Mirabelle for automated proof tools.  Applies
  3236 several tools and tactics like sledgehammer, metis, or quickcheck, to
  3237 every proof step in a theory.  To be used in batch mode via the
  3238 "mirabelle" utility.
  3239 
  3240 * New proof method "sos" (sum of squares) for nonlinear real
  3241 arithmetic (originally due to John Harison). It requires theory
  3242 Library/Sum_Of_Squares.  It is not a complete decision procedure but
  3243 works well in practice on quantifier-free real arithmetic with +, -,
  3244 *, ^, =, <= and <, i.e. boolean combinations of equalities and
  3245 inequalities between polynomials.  It makes use of external
  3246 semidefinite programming solvers.  Method "sos" generates a
  3247 certificate that can be pasted into the proof thus avoiding the need
  3248 to call an external tool every time the proof is checked.  See
  3249 src/HOL/Library/Sum_Of_Squares.
  3250 
  3251 * New method "linarith" invokes existing linear arithmetic decision
  3252 procedure only.
  3253 
  3254 * New command 'atp_minimal' reduces result produced by Sledgehammer.
  3255 
  3256 * New Sledgehammer option "Full Types" in Proof General settings menu.
  3257 Causes full type information to be output to the ATPs.  This slows
  3258 ATPs down considerably but eliminates a source of unsound "proofs"
  3259 that fail later.
  3260 
  3261 * New method "metisFT": A version of metis that uses full type
  3262 information in order to avoid failures of proof reconstruction.
  3263 
  3264 * New evaluator "approximate" approximates an real valued term using
  3265 the same method as the approximation method.
  3266 
  3267 * Method "approximate" now supports arithmetic expressions as
  3268 boundaries of intervals and implements interval splitting and Taylor
  3269 series expansion.
  3270 
  3271 * ML antiquotation @{code_datatype} inserts definition of a datatype
  3272 generated by the code generator; e.g. see src/HOL/Predicate.thy.
  3273 
  3274 * New theory SupInf of the supremum and infimum operators for sets of
  3275 reals.
  3276 
  3277 * New theory Probability, which contains a development of measure
  3278 theory, eventually leading to Lebesgue integration and probability.
  3279 
  3280 * Extended Multivariate Analysis to include derivation and Brouwer's
  3281 fixpoint theorem.
  3282 
  3283 * Reorganization of number theory, INCOMPATIBILITY:
  3284   - new number theory development for nat and int, in theories Divides
  3285     and GCD as well as in new session Number_Theory
  3286   - some constants and facts now suffixed with _nat and _int
  3287     accordingly
  3288   - former session NumberTheory now named Old_Number_Theory, including
  3289     theories Legacy_GCD and Primes (prefer Number_Theory if possible)
  3290   - moved theory Pocklington from src/HOL/Library to
  3291     src/HOL/Old_Number_Theory
  3292 
  3293 * Theory GCD includes functions Gcd/GCD and Lcm/LCM for the gcd and
  3294 lcm of finite and infinite sets. It is shown that they form a complete
  3295 lattice.
  3296 
  3297 * Class semiring_div requires superclass no_zero_divisors and proof of
  3298 div_mult_mult1; theorems div_mult_mult1, div_mult_mult2,
  3299 div_mult_mult1_if, div_mult_mult1 and div_mult_mult2 have been
  3300 generalized to class semiring_div, subsuming former theorems
  3301 zdiv_zmult_zmult1, zdiv_zmult_zmult1_if, zdiv_zmult_zmult1 and
  3302 zdiv_zmult_zmult2.  div_mult_mult1 is now [simp] by default.
  3303 INCOMPATIBILITY.
  3304 
  3305 * Refinements to lattice classes and sets:
  3306   - less default intro/elim rules in locale variant, more default
  3307     intro/elim rules in class variant: more uniformity
  3308   - lemma ge_sup_conv renamed to le_sup_iff, in accordance with
  3309     le_inf_iff
  3310   - dropped lemma alias inf_ACI for inf_aci (same for sup_ACI and
  3311     sup_aci)
  3312   - renamed ACI to inf_sup_aci
  3313   - new class "boolean_algebra"
  3314   - class "complete_lattice" moved to separate theory
  3315     "Complete_Lattice"; corresponding constants (and abbreviations)
  3316     renamed and with authentic syntax:
  3317     Set.Inf ~>    Complete_Lattice.Inf
  3318     Set.Sup ~>    Complete_Lattice.Sup
  3319     Set.INFI ~>   Complete_Lattice.INFI
  3320     Set.SUPR ~>   Complete_Lattice.SUPR
  3321     Set.Inter ~>  Complete_Lattice.Inter
  3322     Set.Union ~>  Complete_Lattice.Union
  3323     Set.INTER ~>  Complete_Lattice.INTER
  3324     Set.UNION ~>  Complete_Lattice.UNION
  3325   - authentic syntax for
  3326     Set.Pow
  3327     Set.image
  3328   - mere abbreviations:
  3329     Set.empty               (for bot)
  3330     Set.UNIV                (for top)
  3331     Set.inter               (for inf, formerly Set.Int)
  3332     Set.union               (for sup, formerly Set.Un)
  3333     Complete_Lattice.Inter  (for Inf)
  3334     Complete_Lattice.Union  (for Sup)
  3335     Complete_Lattice.INTER  (for INFI)
  3336     Complete_Lattice.UNION  (for SUPR)
  3337   - object-logic definitions as far as appropriate
  3338 
  3339 INCOMPATIBILITY.  Care is required when theorems Int_subset_iff or
  3340 Un_subset_iff are explicitly deleted as default simp rules; then also
  3341 their lattice counterparts le_inf_iff and le_sup_iff have to be
  3342 deleted to achieve the desired effect.
  3343 
  3344 * Rules inf_absorb1, inf_absorb2, sup_absorb1, sup_absorb2 are no simp
  3345 rules by default any longer; the same applies to min_max.inf_absorb1
  3346 etc.  INCOMPATIBILITY.
  3347 
  3348 * Rules sup_Int_eq and sup_Un_eq are no longer declared as
  3349 pred_set_conv by default.  INCOMPATIBILITY.
  3350 
  3351 * Power operations on relations and functions are now one dedicated
  3352 constant "compow" with infix syntax "^^".  Power operation on
  3353 multiplicative monoids retains syntax "^" and is now defined generic
  3354 in class power.  INCOMPATIBILITY.
  3355 
  3356 * Relation composition "R O S" now has a more standard argument order:
  3357 "R O S = {(x, z). EX y. (x, y) : R & (y, z) : S}".  INCOMPATIBILITY,
  3358 rewrite propositions with "S O R" --> "R O S". Proofs may occasionally
  3359 break, since the O_assoc rule was not rewritten like this.  Fix using
  3360 O_assoc[symmetric].  The same applies to the curried version "R OO S".
  3361 
  3362 * Function "Inv" is renamed to "inv_into" and function "inv" is now an
  3363 abbreviation for "inv_into UNIV".  Lemmas are renamed accordingly.
  3364 INCOMPATIBILITY.
  3365 
  3366 * Most rules produced by inductive and datatype package have mandatory
  3367 prefixes.  INCOMPATIBILITY.
  3368 
  3369 * Changed "DERIV_intros" to a dynamic fact, which can be augmented by
  3370 the attribute of the same name.  Each of the theorems in the list
  3371 DERIV_intros assumes composition with an additional function and
  3372 matches a variable to the derivative, which has to be solved by the
  3373 Simplifier.  Hence (auto intro!: DERIV_intros) computes the derivative
  3374 of most elementary terms.  Former Maclauren.DERIV_tac and
  3375 Maclauren.deriv_tac should be replaced by (auto intro!: DERIV_intros).
  3376 INCOMPATIBILITY.
  3377 
  3378 * Code generator attributes follow the usual underscore convention:
  3379     code_unfold     replaces    code unfold
  3380     code_post       replaces    code post
  3381     etc.
  3382   INCOMPATIBILITY.
  3383 
  3384 * Renamed methods:
  3385     sizechange -> size_change
  3386     induct_scheme -> induction_schema
  3387   INCOMPATIBILITY.
  3388 
  3389 * Discontinued abbreviation "arbitrary" of constant "undefined".
  3390 INCOMPATIBILITY, use "undefined" directly.
  3391 
  3392 * Renamed theorems:
  3393     Suc_eq_add_numeral_1 -> Suc_eq_plus1
  3394     Suc_eq_add_numeral_1_left -> Suc_eq_plus1_left
  3395     Suc_plus1 -> Suc_eq_plus1
  3396     *anti_sym -> *antisym*
  3397     vector_less_eq_def -> vector_le_def
  3398   INCOMPATIBILITY.
  3399 
  3400 * Added theorem List.map_map as [simp].  Removed List.map_compose.
  3401 INCOMPATIBILITY.
  3402 
  3403 * Removed predicate "M hassize n" (<--> card M = n & finite M).
  3404 INCOMPATIBILITY.
  3405 
  3406 
  3407 *** HOLCF ***
  3408 
  3409 * Theory Representable defines a class "rep" of domains that are
  3410 representable (via an ep-pair) in the universal domain type "udom".
  3411 Instances are provided for all type constructors defined in HOLCF.
  3412 
  3413 * The 'new_domain' command is a purely definitional version of the
  3414 domain package, for representable domains.  Syntax is identical to the
  3415 old domain package.  The 'new_domain' package also supports indirect
  3416 recursion using previously-defined type constructors.  See
  3417 src/HOLCF/ex/New_Domain.thy for examples.
  3418 
  3419 * Method "fixrec_simp" unfolds one step of a fixrec-defined constant
  3420 on the left-hand side of an equation, and then performs
  3421 simplification.  Rewriting is done using rules declared with the
  3422 "fixrec_simp" attribute.  The "fixrec_simp" method is intended as a
  3423 replacement for "fixpat"; see src/HOLCF/ex/Fixrec_ex.thy for examples.
  3424 
  3425 * The pattern-match compiler in 'fixrec' can now handle constructors
  3426 with HOL function types.  Pattern-match combinators for the Pair
  3427 constructor are pre-configured.
  3428 
  3429 * The 'fixrec' package now produces better fixed-point induction rules
  3430 for mutually-recursive definitions:  Induction rules have conclusions
  3431 of the form "P foo bar" instead of "P <foo, bar>".
  3432 
  3433 * The constant "sq_le" (with infix syntax "<<" or "\<sqsubseteq>") has
  3434 been renamed to "below".  The name "below" now replaces "less" in many
  3435 theorem names.  (Legacy theorem names using "less" are still supported
  3436 as well.)
  3437 
  3438 * The 'fixrec' package now supports "bottom patterns".  Bottom
  3439 patterns can be used to generate strictness rules, or to make
  3440 functions more strict (much like the bang-patterns supported by the
  3441 Glasgow Haskell Compiler).  See src/HOLCF/ex/Fixrec_ex.thy for
  3442 examples.
  3443 
  3444 
  3445 *** ML ***
  3446 
  3447 * Support for Poly/ML 5.3.0, with improved reporting of compiler
  3448 errors and run-time exceptions, including detailed source positions.
  3449 
  3450 * Structure Name_Space (formerly NameSpace) now manages uniquely
  3451 identified entries, with some additional information such as source
  3452 position, logical grouping etc.
  3453 
  3454 * Theory and context data is now introduced by the simplified and
  3455 modernized functors Theory_Data, Proof_Data, Generic_Data.  Data needs
  3456 to be pure, but the old TheoryDataFun for mutable data (with explicit
  3457 copy operation) is still available for some time.
  3458 
  3459 * Structure Synchronized (cf. src/Pure/Concurrent/synchronized.ML)
  3460 provides a high-level programming interface to synchronized state
  3461 variables with atomic update.  This works via pure function
  3462 application within a critical section -- its runtime should be as
  3463 short as possible; beware of deadlocks if critical code is nested,
  3464 either directly or indirectly via other synchronized variables!
  3465 
  3466 * Structure Unsynchronized (cf. src/Pure/ML-Systems/unsynchronized.ML)
  3467 wraps raw ML references, explicitly indicating their non-thread-safe
  3468 behaviour.  The Isar toplevel keeps this structure open, to
  3469 accommodate Proof General as well as quick and dirty interactive
  3470 experiments with references.
  3471 
  3472 * PARALLEL_CHOICE and PARALLEL_GOALS provide basic support for
  3473 parallel tactical reasoning.
  3474 
  3475 * Tacticals Subgoal.FOCUS, Subgoal.FOCUS_PREMS, Subgoal.FOCUS_PARAMS
  3476 are similar to SUBPROOF, but are slightly more flexible: only the
  3477 specified parts of the subgoal are imported into the context, and the
  3478 body tactic may introduce new subgoals and schematic variables.
  3479 
  3480 * Old tactical METAHYPS, which does not observe the proof context, has
  3481 been renamed to Old_Goals.METAHYPS and awaits deletion.  Use SUBPROOF
  3482 or Subgoal.FOCUS etc.
  3483 
  3484 * Renamed functor TableFun to Table, and GraphFun to Graph.  (Since
  3485 functors have their own ML name space there is no point to mark them
  3486 separately.)  Minor INCOMPATIBILITY.
  3487 
  3488 * Renamed NamedThmsFun to Named_Thms.  INCOMPATIBILITY.
  3489 
  3490 * Renamed several structures FooBar to Foo_Bar.  Occasional,
  3491 INCOMPATIBILITY.
  3492 
  3493 * Operations of structure Skip_Proof no longer require quick_and_dirty
  3494 mode, which avoids critical setmp.
  3495 
  3496 * Eliminated old Attrib.add_attributes, Method.add_methods and related
  3497 combinators for "args".  INCOMPATIBILITY, need to use simplified
  3498 Attrib/Method.setup introduced in Isabelle2009.
  3499 
  3500 * Proper context for simpset_of, claset_of, clasimpset_of.  May fall
  3501 back on global_simpset_of, global_claset_of, global_clasimpset_of as
  3502 last resort.  INCOMPATIBILITY.
  3503 
  3504 * Display.pretty_thm now requires a proper context (cf. former
  3505 ProofContext.pretty_thm).  May fall back on Display.pretty_thm_global
  3506 or even Display.pretty_thm_without_context as last resort.
  3507 INCOMPATIBILITY.
  3508 
  3509 * Discontinued Display.pretty_ctyp/cterm etc.  INCOMPATIBILITY, use
  3510 Syntax.pretty_typ/term directly, preferably with proper context
  3511 instead of global theory.
  3512 
  3513 
  3514 *** System ***
  3515 
  3516 * Further fine tuning of parallel proof checking, scales up to 8 cores
  3517 (max. speedup factor 5.0).  See also Goal.parallel_proofs in ML and
  3518 usedir option -q.
  3519 
  3520 * Support for additional "Isabelle components" via etc/components, see
  3521 also the system manual.
  3522 
  3523 * The isabelle makeall tool now operates on all components with
  3524 IsaMakefile, not just hardwired "logics".
  3525 
  3526 * Removed "compress" option from isabelle-process and isabelle usedir;
  3527 this is always enabled.
  3528 
  3529 * Discontinued support for Poly/ML 4.x versions.
  3530 
  3531 * Isabelle tool "wwwfind" provides web interface for 'find_theorems'
  3532 on a given logic image.  This requires the lighttpd webserver and is
  3533 currently supported on Linux only.
  3534 
  3535 
  3536 
  3537 New in Isabelle2009 (April 2009)
  3538 --------------------------------
  3539 
  3540 *** General ***
  3541 
  3542 * Simplified main Isabelle executables, with less surprises on
  3543 case-insensitive file-systems (such as Mac OS).
  3544 
  3545   - The main Isabelle tool wrapper is now called "isabelle" instead of
  3546     "isatool."
  3547 
  3548   - The former "isabelle" alias for "isabelle-process" has been
  3549     removed (should rarely occur to regular users).
  3550 
  3551   - The former "isabelle-interface" and its alias "Isabelle" have been
  3552     removed (interfaces are now regular Isabelle tools).
  3553 
  3554 Within scripts and make files, the Isabelle environment variables
  3555 ISABELLE_TOOL and ISABELLE_PROCESS replace old ISATOOL and ISABELLE,
  3556 respectively.  (The latter are still available as legacy feature.)
  3557 
  3558 The old isabelle-interface wrapper could react in confusing ways if
  3559 the interface was uninstalled or changed otherwise.  Individual
  3560 interface tool configuration is now more explicit, see also the
  3561 Isabelle system manual.  In particular, Proof General is now available
  3562 via "isabelle emacs".
  3563 
  3564 INCOMPATIBILITY, need to adapt derivative scripts.  Users may need to
  3565 purge installed copies of Isabelle executables and re-run "isabelle
  3566 install -p ...", or use symlinks.
  3567 
  3568 * The default for ISABELLE_HOME_USER is now ~/.isabelle instead of the
  3569 old ~/isabelle, which was slightly non-standard and apt to cause
  3570 surprises on case-insensitive file-systems (such as Mac OS).
  3571 
  3572 INCOMPATIBILITY, need to move existing ~/isabelle/etc,
  3573 ~/isabelle/heaps, ~/isabelle/browser_info to the new place.  Special
  3574 care is required when using older releases of Isabelle.  Note that
  3575 ISABELLE_HOME_USER can be changed in Isabelle/etc/settings of any
  3576 Isabelle distribution, in order to use the new ~/.isabelle uniformly.
  3577 
  3578 * Proofs of fully specified statements are run in parallel on
  3579 multi-core systems.  A speedup factor of 2.5 to 3.2 can be expected on
  3580 a regular 4-core machine, if the initial heap space is made reasonably
  3581 large (cf. Poly/ML option -H).  (Requires Poly/ML 5.2.1 or later.)
  3582 
  3583 * The main reference manuals ("isar-ref", "implementation", and
  3584 "system") have been updated and extended.  Formally checked references
  3585 as hyperlinks are now available uniformly.
  3586 
  3587 
  3588 *** Pure ***
  3589 
  3590 * Complete re-implementation of locales.  INCOMPATIBILITY in several
  3591 respects.  The most important changes are listed below.  See the
  3592 Tutorial on Locales ("locales" manual) for details.
  3593 
  3594 - In locale expressions, instantiation replaces renaming.  Parameters
  3595 must be declared in a for clause.  To aid compatibility with previous
  3596 parameter inheritance, in locale declarations, parameters that are not
  3597 'touched' (instantiation position "_" or omitted) are implicitly added
  3598 with their syntax at the beginning of the for clause.
  3599 
  3600 - Syntax from abbreviations and definitions in locales is available in
  3601 locale expressions and context elements.  The latter is particularly
  3602 useful in locale declarations.
  3603 
  3604 - More flexible mechanisms to qualify names generated by locale
  3605 expressions.  Qualifiers (prefixes) may be specified in locale
  3606 expressions, and can be marked as mandatory (syntax: "name!:") or
  3607 optional (syntax "name?:").  The default depends for plain "name:"
  3608 depends on the situation where a locale expression is used: in
  3609 commands 'locale' and 'sublocale' prefixes are optional, in
  3610 'interpretation' and 'interpret' prefixes are mandatory.  The old
  3611 implicit qualifiers derived from the parameter names of a locale are
  3612 no longer generated.
  3613 
  3614 - Command "sublocale l < e" replaces "interpretation l < e".  The
  3615 instantiation clause in "interpretation" and "interpret" (square
  3616 brackets) is no longer available.  Use locale expressions.
  3617 
  3618 - When converting proof scripts, mandatory qualifiers in
  3619 'interpretation' and 'interpret' should be retained by default, even
  3620 if this is an INCOMPATIBILITY compared to former behavior.  In the
  3621 worst case, use the "name?:" form for non-mandatory ones.  Qualifiers
  3622 in locale expressions range over a single locale instance only.
  3623 
  3624 - Dropped locale element "includes".  This is a major INCOMPATIBILITY.
  3625 In existing theorem specifications replace the includes element by the
  3626 respective context elements of the included locale, omitting those
  3627 that are already present in the theorem specification.  Multiple
  3628 assume elements of a locale should be replaced by a single one
  3629 involving the locale predicate.  In the proof body, declarations (most
  3630 notably theorems) may be regained by interpreting the respective
  3631 locales in the proof context as required (command "interpret").
  3632 
  3633 If using "includes" in replacement of a target solely because the
  3634 parameter types in the theorem are not as general as in the target,
  3635 consider declaring a new locale with additional type constraints on
  3636 the parameters (context element "constrains").
  3637 
  3638 - Discontinued "locale (open)".  INCOMPATIBILITY.
  3639 
  3640 - Locale interpretation commands no longer attempt to simplify goal.
  3641 INCOMPATIBILITY: in rare situations the generated goal differs.  Use
  3642 methods intro_locales and unfold_locales to clarify.
  3643 
  3644 - Locale interpretation commands no longer accept interpretation
  3645 attributes.  INCOMPATIBILITY.
  3646 
  3647 * Class declaration: so-called "base sort" must not be given in import
  3648 list any longer, but is inferred from the specification.  Particularly
  3649 in HOL, write
  3650 
  3651     class foo = ...
  3652 
  3653 instead of
  3654 
  3655     class foo = type + ...
  3656 
  3657 * Class target: global versions of theorems stemming do not carry a
  3658 parameter prefix any longer.  INCOMPATIBILITY.
  3659 
  3660 * Class 'instance' command no longer accepts attached definitions.
  3661 INCOMPATIBILITY, use proper 'instantiation' target instead.
  3662 
  3663 * Recovered hiding of consts, which was accidentally broken in
  3664 Isabelle2007.  Potential INCOMPATIBILITY, ``hide const c'' really
  3665 makes c inaccessible; consider using ``hide (open) const c'' instead.
  3666 
  3667 * Slightly more coherent Pure syntax, with updated documentation in
  3668 isar-ref manual.  Removed locales meta_term_syntax and
  3669 meta_conjunction_syntax: TERM and &&& (formerly &&) are now permanent,
  3670 INCOMPATIBILITY in rare situations.  Note that &&& should not be used
  3671 directly in regular applications.
  3672 
  3673 * There is a new syntactic category "float_const" for signed decimal
  3674 fractions (e.g. 123.45 or -123.45).
  3675 
  3676 * Removed exotic 'token_translation' command.  INCOMPATIBILITY, use ML
  3677 interface with 'setup' command instead.
  3678 
  3679 * Command 'local_setup' is similar to 'setup', but operates on a local
  3680 theory context.
  3681 
  3682 * The 'axiomatization' command now only works within a global theory
  3683 context.  INCOMPATIBILITY.
  3684 
  3685 * Goal-directed proof now enforces strict proof irrelevance wrt. sort
  3686 hypotheses.  Sorts required in the course of reasoning need to be
  3687 covered by the constraints in the initial statement, completed by the
  3688 type instance information of the background theory.  Non-trivial sort
  3689 hypotheses, which rarely occur in practice, may be specified via
  3690 vacuous propositions of the form SORT_CONSTRAINT('a::c).  For example:
  3691 
  3692   lemma assumes "SORT_CONSTRAINT('a::empty)" shows False ...
  3693 
  3694 The result contains an implicit sort hypotheses as before --
  3695 SORT_CONSTRAINT premises are eliminated as part of the canonical rule
  3696 normalization.
  3697 
  3698 * Generalized Isar history, with support for linear undo, direct state
  3699 addressing etc.
  3700 
  3701 * Changed defaults for unify configuration options:
  3702 
  3703   unify_trace_bound = 50 (formerly 25)
  3704   unify_search_bound = 60 (formerly 30)
  3705 
  3706 * Different bookkeeping for code equations (INCOMPATIBILITY):
  3707 
  3708   a) On theory merge, the last set of code equations for a particular
  3709      constant is taken (in accordance with the policy applied by other
  3710      parts of the code generator framework).
  3711 
  3712   b) Code equations stemming from explicit declarations (e.g. code
  3713      attribute) gain priority over default code equations stemming
  3714      from definition, primrec, fun etc.
  3715 
  3716 * Keyword 'code_exception' now named 'code_abort'.  INCOMPATIBILITY.
  3717 
  3718 * Unified theorem tables for both code generators.  Thus [code
  3719 func] has disappeared and only [code] remains.  INCOMPATIBILITY.
  3720 
  3721 * Command 'find_consts' searches for constants based on type and name
  3722 patterns, e.g.
  3723 
  3724     find_consts "_ => bool"
  3725 
  3726 By default, matching is against subtypes, but it may be restricted to
  3727 the whole type.  Searching by name is possible.  Multiple queries are
  3728 conjunctive and queries may be negated by prefixing them with a
  3729 hyphen:
  3730 
  3731     find_consts strict: "_ => bool" name: "Int" -"int => int"
  3732 
  3733 * New 'find_theorems' criterion "solves" matches theorems that
  3734 directly solve the current goal (modulo higher-order unification).
  3735 
  3736 * Auto solve feature for main theorem statements: whenever a new goal
  3737 is stated, "find_theorems solves" is called; any theorems that could
  3738 solve the lemma directly are listed as part of the goal state.
  3739 Cf. associated options in Proof General Isabelle settings menu,
  3740 enabled by default, with reasonable timeout for pathological cases of
  3741 higher-order unification.
  3742 
  3743 
  3744 *** Document preparation ***
  3745 
  3746 * Antiquotation @{lemma} now imitates a regular terminal proof,
  3747 demanding keyword 'by' and supporting the full method expression
  3748 syntax just like the Isar command 'by'.
  3749 
  3750 
  3751 *** HOL ***
  3752 
  3753 * Integrated main parts of former image HOL-Complex with HOL.  Entry
  3754 points Main and Complex_Main remain as before.
  3755 
  3756 * Logic image HOL-Plain provides a minimal HOL with the most important
  3757 tools available (inductive, datatype, primrec, ...).  This facilitates
  3758 experimentation and tool development.  Note that user applications
  3759 (and library theories) should never refer to anything below theory
  3760 Main, as before.
  3761 
  3762 * Logic image HOL-Main stops at theory Main, and thus facilitates
  3763 experimentation due to shorter build times.
  3764 
  3765 * Logic image HOL-NSA contains theories of nonstandard analysis which
  3766 were previously part of former HOL-Complex.  Entry point Hyperreal
  3767 remains valid, but theories formerly using Complex_Main should now use
  3768 new entry point Hypercomplex.
  3769 
  3770 * Generic ATP manager for Sledgehammer, based on ML threads instead of
  3771 Posix processes.  Avoids potentially expensive forking of the ML
  3772 process.  New thread-based implementation also works on non-Unix
  3773 platforms (Cygwin).  Provers are no longer hardwired, but defined
  3774 within the theory via plain ML wrapper functions.  Basic Sledgehammer
  3775 commands are covered in the isar-ref manual.
  3776 
  3777 * Wrapper scripts for remote SystemOnTPTP service allows to use
  3778 sledgehammer without local ATP installation (Vampire etc.). Other
  3779 provers may be included via suitable ML wrappers, see also
  3780 src/HOL/ATP_Linkup.thy.
  3781 
  3782 * ATP selection (E/Vampire/Spass) is now via Proof General's settings
  3783 menu.
  3784 
  3785 * The metis method no longer fails because the theorem is too trivial
  3786 (contains the empty clause).
  3787 
  3788 * The metis method now fails in the usual manner, rather than raising
  3789 an exception, if it determines that it cannot prove the theorem.
  3790 
  3791 * Method "coherent" implements a prover for coherent logic (see also
  3792 src/Tools/coherent.ML).
  3793 
  3794 * Constants "undefined" and "default" replace "arbitrary".  Usually
  3795 "undefined" is the right choice to replace "arbitrary", though
  3796 logically there is no difference.  INCOMPATIBILITY.
  3797 
  3798 * Command "value" now integrates different evaluation mechanisms.  The
  3799 result of the first successful evaluation mechanism is printed.  In
  3800 square brackets a particular named evaluation mechanisms may be
  3801 specified (currently, [SML], [code] or [nbe]).  See further
  3802 src/HOL/ex/Eval_Examples.thy.
  3803 
  3804 * Normalization by evaluation now allows non-leftlinear equations.
  3805 Declare with attribute [code nbe].
  3806 
  3807 * Methods "case_tac" and "induct_tac" now refer to the very same rules
  3808 as the structured Isar versions "cases" and "induct", cf. the
  3809 corresponding "cases" and "induct" attributes.  Mutual induction rules
  3810 are now presented as a list of individual projections
  3811 (e.g. foo_bar.inducts for types foo and bar); the old format with
  3812 explicit HOL conjunction is no longer supported.  INCOMPATIBILITY, in
  3813 rare situations a different rule is selected --- notably nested tuple
  3814 elimination instead of former prod.exhaust: use explicit (case_tac t
  3815 rule: prod.exhaust) here.
  3816 
  3817 * Attributes "cases", "induct", "coinduct" support "del" option.
  3818 
  3819 * Removed fact "case_split_thm", which duplicates "case_split".
  3820 
  3821 * The option datatype has been moved to a new theory Option.  Renamed
  3822 option_map to Option.map, and o2s to Option.set, INCOMPATIBILITY.
  3823 
  3824 * New predicate "strict_mono" classifies strict functions on partial
  3825 orders.  With strict functions on linear orders, reasoning about
  3826 (in)equalities is facilitated by theorems "strict_mono_eq",
  3827 "strict_mono_less_eq" and "strict_mono_less".
  3828 
  3829 * Some set operations are now proper qualified constants with
  3830 authentic syntax.  INCOMPATIBILITY:
  3831 
  3832     op Int ~>   Set.Int
  3833     op Un ~>    Set.Un
  3834     INTER ~>    Set.INTER
  3835     UNION ~>    Set.UNION
  3836     Inter ~>    Set.Inter
  3837     Union ~>    Set.Union
  3838     {} ~>       Set.empty
  3839     UNIV ~>     Set.UNIV
  3840 
  3841 * Class complete_lattice with operations Inf, Sup, INFI, SUPR now in
  3842 theory Set.
  3843 
  3844 * Auxiliary class "itself" has disappeared -- classes without any
  3845 parameter are treated as expected by the 'class' command.
  3846 
  3847 * Leibnitz's Series for Pi and the arcus tangens and logarithm series.
  3848 
  3849 * Common decision procedures (Cooper, MIR, Ferrack, Approximation,
  3850 Dense_Linear_Order) are now in directory HOL/Decision_Procs.
  3851 
  3852 * Theory src/HOL/Decision_Procs/Approximation provides the new proof
  3853 method "approximation".  It proves formulas on real values by using
  3854 interval arithmetic.  In the formulas are also the transcendental
  3855 functions sin, cos, tan, atan, ln, exp and the constant pi are
  3856 allowed. For examples see
  3857 src/HOL/Descision_Procs/ex/Approximation_Ex.thy.
  3858 
  3859 * Theory "Reflection" now resides in HOL/Library.
  3860 
  3861 * Entry point to Word library now simply named "Word".
  3862 INCOMPATIBILITY.
  3863 
  3864 * Made source layout more coherent with logical distribution
  3865 structure:
  3866 
  3867     src/HOL/Library/RType.thy ~> src/HOL/Typerep.thy
  3868     src/HOL/Library/Code_Message.thy ~> src/HOL/
  3869     src/HOL/Library/GCD.thy ~> src/HOL/
  3870     src/HOL/Library/Order_Relation.thy ~> src/HOL/
  3871     src/HOL/Library/Parity.thy ~> src/HOL/
  3872     src/HOL/Library/Univ_Poly.thy ~> src/HOL/
  3873     src/HOL/Real/ContNotDenum.thy ~> src/HOL/Library/
  3874     src/HOL/Real/Lubs.thy ~> src/HOL/
  3875     src/HOL/Real/PReal.thy ~> src/HOL/
  3876     src/HOL/Real/Rational.thy ~> src/HOL/
  3877     src/HOL/Real/RComplete.thy ~> src/HOL/
  3878     src/HOL/Real/RealDef.thy ~> src/HOL/
  3879     src/HOL/Real/RealPow.thy ~> src/HOL/
  3880     src/HOL/Real/Real.thy ~> src/HOL/
  3881     src/HOL/Complex/Complex_Main.thy ~> src/HOL/
  3882     src/HOL/Complex/Complex.thy ~> src/HOL/
  3883     src/HOL/Complex/FrechetDeriv.thy ~> src/HOL/Library/
  3884     src/HOL/Complex/Fundamental_Theorem_Algebra.thy ~> src/HOL/Library/
  3885     src/HOL/Hyperreal/Deriv.thy ~> src/HOL/
  3886     src/HOL/Hyperreal/Fact.thy ~> src/HOL/
  3887     src/HOL/Hyperreal/Integration.thy ~> src/HOL/
  3888     src/HOL/Hyperreal/Lim.thy ~> src/HOL/
  3889     src/HOL/Hyperreal/Ln.thy ~> src/HOL/
  3890     src/HOL/Hyperreal/Log.thy ~> src/HOL/
  3891     src/HOL/Hyperreal/MacLaurin.thy ~> src/HOL/
  3892     src/HOL/Hyperreal/NthRoot.thy ~> src/HOL/
  3893     src/HOL/Hyperreal/Series.thy ~> src/HOL/
  3894     src/HOL/Hyperreal/SEQ.thy ~> src/HOL/
  3895     src/HOL/Hyperreal/Taylor.thy ~> src/HOL/
  3896     src/HOL/Hyperreal/Transcendental.thy ~> src/HOL/
  3897     src/HOL/Real/Float ~> src/HOL/Library/
  3898     src/HOL/Real/HahnBanach ~> src/HOL/HahnBanach
  3899     src/HOL/Real/RealVector.thy ~> src/HOL/
  3900 
  3901     src/HOL/arith_data.ML ~> src/HOL/Tools
  3902     src/HOL/hologic.ML ~> src/HOL/Tools
  3903     src/HOL/simpdata.ML ~> src/HOL/Tools
  3904     src/HOL/int_arith1.ML ~> src/HOL/Tools/int_arith.ML
  3905     src/HOL/int_factor_simprocs.ML ~> src/HOL/Tools
  3906     src/HOL/nat_simprocs.ML ~> src/HOL/Tools
  3907     src/HOL/Real/float_arith.ML ~> src/HOL/Tools
  3908     src/HOL/Real/float_syntax.ML ~> src/HOL/Tools
  3909     src/HOL/Real/rat_arith.ML ~> src/HOL/Tools
  3910     src/HOL/Real/real_arith.ML ~> src/HOL/Tools
  3911 
  3912     src/HOL/Library/Array.thy ~> src/HOL/Imperative_HOL
  3913     src/HOL/Library/Heap_Monad.thy ~> src/HOL/Imperative_HOL
  3914     src/HOL/Library/Heap.thy ~> src/HOL/Imperative_HOL
  3915     src/HOL/Library/Imperative_HOL.thy ~> src/HOL/Imperative_HOL
  3916     src/HOL/Library/Ref.thy ~> src/HOL/Imperative_HOL
  3917     src/HOL/Library/Relational.thy ~> src/HOL/Imperative_HOL
  3918 
  3919 * If methods "eval" and "evaluation" encounter a structured proof
  3920 state with !!/==>, only the conclusion is evaluated to True (if
  3921 possible), avoiding strange error messages.
  3922 
  3923 * Method "sizechange" automates termination proofs using (a
  3924 modification of) the size-change principle.  Requires SAT solver.  See
  3925 src/HOL/ex/Termination.thy for examples.
  3926 
  3927 * Simplifier: simproc for let expressions now unfolds if bound
  3928 variable occurs at most once in let expression body.  INCOMPATIBILITY.
  3929 
  3930 * Method "arith": Linear arithmetic now ignores all inequalities when
  3931 fast_arith_neq_limit is exceeded, instead of giving up entirely.
  3932 
  3933 * New attribute "arith" for facts that should always be used
  3934 automatically by arithmetic. It is intended to be used locally in
  3935 proofs, e.g.
  3936 
  3937   assumes [arith]: "x > 0"
  3938 
  3939 Global usage is discouraged because of possible performance impact.
  3940 
  3941 * New classes "top" and "bot" with corresponding operations "top" and
  3942 "bot" in theory Orderings; instantiation of class "complete_lattice"
  3943 requires instantiation of classes "top" and "bot".  INCOMPATIBILITY.
  3944 
  3945 * Changed definition lemma "less_fun_def" in order to provide an
  3946 instance for preorders on functions; use lemma "less_le" instead.
  3947 INCOMPATIBILITY.
  3948 
  3949 * Theory Orderings: class "wellorder" moved here, with explicit
  3950 induction rule "less_induct" as assumption.  For instantiation of
  3951 "wellorder" by means of predicate "wf", use rule wf_wellorderI.
  3952 INCOMPATIBILITY.
  3953 
  3954 * Theory Orderings: added class "preorder" as superclass of "order".
  3955 INCOMPATIBILITY: Instantiation proofs for order, linorder
  3956 etc. slightly changed.  Some theorems named order_class.* now named
  3957 preorder_class.*.
  3958 
  3959 * Theory Relation: renamed "refl" to "refl_on", "reflexive" to "refl,
  3960 "diag" to "Id_on".
  3961 
  3962 * Theory Finite_Set: added a new fold combinator of type
  3963 
  3964   ('a => 'b => 'b) => 'b => 'a set => 'b
  3965 
  3966 Occasionally this is more convenient than the old fold combinator
  3967 which is now defined in terms of the new one and renamed to
  3968 fold_image.
  3969 
  3970 * Theories Ring_and_Field and OrderedGroup: The lemmas "group_simps"
  3971 and "ring_simps" have been replaced by "algebra_simps" (which can be
  3972 extended with further lemmas!).  At the moment both still exist but
  3973 the former will disappear at some point.
  3974 
  3975 * Theory Power: Lemma power_Suc is now declared as a simp rule in
  3976 class recpower.  Type-specific simp rules for various recpower types
  3977 have been removed.  INCOMPATIBILITY, rename old lemmas as follows:
  3978 
  3979 rat_power_0    -> power_0
  3980 rat_power_Suc  -> power_Suc
  3981 realpow_0      -> power_0
  3982 realpow_Suc    -> power_Suc
  3983 complexpow_0   -> power_0
  3984 complexpow_Suc -> power_Suc
  3985 power_poly_0   -> power_0
  3986 power_poly_Suc -> power_Suc
  3987 
  3988 * Theories Ring_and_Field and Divides: Definition of "op dvd" has been
  3989 moved to separate class dvd in Ring_and_Field; a couple of lemmas on
  3990 dvd has been generalized to class comm_semiring_1.  Likewise a bunch
  3991 of lemmas from Divides has been generalized from nat to class
  3992 semiring_div.  INCOMPATIBILITY.  This involves the following theorem
  3993 renames resulting from duplicate elimination:
  3994 
  3995     dvd_def_mod ~>          dvd_eq_mod_eq_0
  3996     zero_dvd_iff ~>         dvd_0_left_iff
  3997     dvd_0 ~>                dvd_0_right
  3998     DIVISION_BY_ZERO_DIV ~> div_by_0
  3999     DIVISION_BY_ZERO_MOD ~> mod_by_0
  4000     mult_div ~>             div_mult_self2_is_id
  4001     mult_mod ~>             mod_mult_self2_is_0
  4002 
  4003 * Theory IntDiv: removed many lemmas that are instances of class-based
  4004 generalizations (from Divides and Ring_and_Field).  INCOMPATIBILITY,
  4005 rename old lemmas as follows:
  4006 
  4007 dvd_diff               -> nat_dvd_diff
  4008 dvd_zminus_iff         -> dvd_minus_iff
  4009 mod_add1_eq            -> mod_add_eq
  4010 mod_mult1_eq           -> mod_mult_right_eq
  4011 mod_mult1_eq'          -> mod_mult_left_eq
  4012 mod_mult_distrib_mod   -> mod_mult_eq
  4013 nat_mod_add_left_eq    -> mod_add_left_eq
  4014 nat_mod_add_right_eq   -> mod_add_right_eq
  4015 nat_mod_div_trivial    -> mod_div_trivial
  4016 nat_mod_mod_trivial    -> mod_mod_trivial
  4017 zdiv_zadd_self1        -> div_add_self1
  4018 zdiv_zadd_self2        -> div_add_self2
  4019 zdiv_zmult_self1       -> div_mult_self2_is_id
  4020 zdiv_zmult_self2       -> div_mult_self1_is_id
  4021 zdvd_triv_left         -> dvd_triv_left
  4022 zdvd_triv_right        -> dvd_triv_right
  4023 zdvd_zmult_cancel_disj -> dvd_mult_cancel_left
  4024 zmod_eq0_zdvd_iff      -> dvd_eq_mod_eq_0[symmetric]
  4025 zmod_zadd_left_eq      -> mod_add_left_eq
  4026 zmod_zadd_right_eq     -> mod_add_right_eq
  4027 zmod_zadd_self1        -> mod_add_self1
  4028 zmod_zadd_self2        -> mod_add_self2
  4029 zmod_zadd1_eq          -> mod_add_eq
  4030 zmod_zdiff1_eq         -> mod_diff_eq
  4031 zmod_zdvd_zmod         -> mod_mod_cancel
  4032 zmod_zmod_cancel       -> mod_mod_cancel
  4033 zmod_zmult_self1       -> mod_mult_self2_is_0
  4034 zmod_zmult_self2       -> mod_mult_self1_is_0
  4035 zmod_1                 -> mod_by_1
  4036 zdiv_1                 -> div_by_1
  4037 zdvd_abs1              -> abs_dvd_iff
  4038 zdvd_abs2              -> dvd_abs_iff
  4039 zdvd_refl              -> dvd_refl
  4040 zdvd_trans             -> dvd_trans
  4041 zdvd_zadd              -> dvd_add
  4042 zdvd_zdiff             -> dvd_diff
  4043 zdvd_zminus_iff        -> dvd_minus_iff
  4044 zdvd_zminus2_iff       -> minus_dvd_iff
  4045 zdvd_zmultD            -> dvd_mult_right
  4046 zdvd_zmultD2           -> dvd_mult_left
  4047 zdvd_zmult_mono        -> mult_dvd_mono
  4048 zdvd_0_right           -> dvd_0_right
  4049 zdvd_0_left            -> dvd_0_left_iff
  4050 zdvd_1_left            -> one_dvd
  4051 zminus_dvd_iff         -> minus_dvd_iff
  4052 
  4053 * Theory Rational: 'Fract k 0' now equals '0'.  INCOMPATIBILITY.
  4054 
  4055 * The real numbers offer decimal input syntax: 12.34 is translated
  4056 into 1234/10^2. This translation is not reversed upon output.
  4057 
  4058 * Theory Library/Polynomial defines an abstract type 'a poly of
  4059 univariate polynomials with coefficients of type 'a.  In addition to
  4060 the standard ring operations, it also supports div and mod.  Code
  4061 generation is also supported, using list-style constructors.
  4062 
  4063 * Theory Library/Inner_Product defines a class of real_inner for real
  4064 inner product spaces, with an overloaded operation inner :: 'a => 'a
  4065 => real.  Class real_inner is a subclass of real_normed_vector from
  4066 theory RealVector.
  4067 
  4068 * Theory Library/Product_Vector provides instances for the product
  4069 type 'a * 'b of several classes from RealVector and Inner_Product.
  4070 Definitions of addition, subtraction, scalar multiplication, norms,
  4071 and inner products are included.
  4072 
  4073 * Theory Library/Bit defines the field "bit" of integers modulo 2.  In
  4074 addition to the field operations, numerals and case syntax are also
  4075 supported.
  4076 
  4077 * Theory Library/Diagonalize provides constructive version of Cantor's
  4078 first diagonalization argument.
  4079 
  4080 * Theory Library/GCD: Curried operations gcd, lcm (for nat) and zgcd,
  4081 zlcm (for int); carried together from various gcd/lcm developements in
  4082 the HOL Distribution.  Constants zgcd and zlcm replace former igcd and
  4083 ilcm; corresponding theorems renamed accordingly.  INCOMPATIBILITY,
  4084 may recover tupled syntax as follows:
  4085 
  4086     hide (open) const gcd
  4087     abbreviation gcd where
  4088       "gcd == (%(a, b). GCD.gcd a b)"
  4089     notation (output)
  4090       GCD.gcd ("gcd '(_, _')")
  4091 
  4092 The same works for lcm, zgcd, zlcm.
  4093 
  4094 * Theory Library/Nat_Infinity: added addition, numeral syntax and more
  4095 instantiations for algebraic structures.  Removed some duplicate
  4096 theorems.  Changes in simp rules.  INCOMPATIBILITY.
  4097 
  4098 * ML antiquotation @{code} takes a constant as argument and generates
  4099 corresponding code in background and inserts name of the corresponding
  4100 resulting ML value/function/datatype constructor binding in place.
  4101 All occurrences of @{code} with a single ML block are generated
  4102 simultaneously.  Provides a generic and safe interface for
  4103 instrumentalizing code generation.  See
  4104 src/HOL/Decision_Procs/Ferrack.thy for a more ambitious application.
  4105 In future you ought to refrain from ad-hoc compiling generated SML
  4106 code on the ML toplevel.  Note that (for technical reasons) @{code}
  4107 cannot refer to constants for which user-defined serializations are
  4108 set.  Refer to the corresponding ML counterpart directly in that
  4109 cases.
  4110 
  4111 * Command 'rep_datatype': instead of theorem names the command now
  4112 takes a list of terms denoting the constructors of the type to be
  4113 represented as datatype.  The characteristic theorems have to be
  4114 proven.  INCOMPATIBILITY.  Also observe that the following theorems
  4115 have disappeared in favour of existing ones:
  4116 
  4117     unit_induct                 ~> unit.induct
  4118     prod_induct                 ~> prod.induct
  4119     sum_induct                  ~> sum.induct
  4120     Suc_Suc_eq                  ~> nat.inject
  4121     Suc_not_Zero Zero_not_Suc   ~> nat.distinct
  4122 
  4123 
  4124 *** HOL-Algebra ***
  4125 
  4126 * New locales for orders and lattices where the equivalence relation
  4127 is not restricted to equality.  INCOMPATIBILITY: all order and lattice
  4128 locales use a record structure with field eq for the equivalence.
  4129 
  4130 * New theory of factorial domains.
  4131 
  4132 * Units_l_inv and Units_r_inv are now simp rules by default.
  4133 INCOMPATIBILITY.  Simplifier proof that require deletion of l_inv
  4134 and/or r_inv will now also require deletion of these lemmas.
  4135 
  4136 * Renamed the following theorems, INCOMPATIBILITY:
  4137 
  4138 UpperD ~> Upper_memD
  4139 LowerD ~> Lower_memD
  4140 least_carrier ~> least_closed
  4141 greatest_carrier ~> greatest_closed
  4142 greatest_Lower_above ~> greatest_Lower_below
  4143 one_zero ~> carrier_one_zero
  4144 one_not_zero ~> carrier_one_not_zero  (collision with assumption)
  4145 
  4146 
  4147 *** HOL-Nominal ***
  4148 
  4149 * Nominal datatypes can now contain type-variables.
  4150 
  4151 * Commands 'nominal_inductive' and 'equivariance' work with local
  4152 theory targets.
  4153 
  4154 * Nominal primrec can now works with local theory targets and its
  4155 specification syntax now conforms to the general format as seen in
  4156 'inductive' etc.
  4157 
  4158 * Method "perm_simp" honours the standard simplifier attributes
  4159 (no_asm), (no_asm_use) etc.
  4160 
  4161 * The new predicate #* is defined like freshness, except that on the
  4162 left hand side can be a set or list of atoms.
  4163 
  4164 * Experimental command 'nominal_inductive2' derives strong induction
  4165 principles for inductive definitions.  In contrast to
  4166 'nominal_inductive', which can only deal with a fixed number of
  4167 binders, it can deal with arbitrary expressions standing for sets of
  4168 atoms to be avoided.  The only inductive definition we have at the
  4169 moment that needs this generalisation is the typing rule for Lets in
  4170 the algorithm W:
  4171 
  4172  Gamma |- t1 : T1   (x,close Gamma T1)::Gamma |- t2 : T2   x#Gamma
  4173  -----------------------------------------------------------------
  4174          Gamma |- Let x be t1 in t2 : T2
  4175 
  4176 In this rule one wants to avoid all the binders that are introduced by
  4177 "close Gamma T1".  We are looking for other examples where this
  4178 feature might be useful.  Please let us know.
  4179 
  4180 
  4181 *** HOLCF ***
  4182 
  4183 * Reimplemented the simplification procedure for proving continuity
  4184 subgoals.  The new simproc is extensible; users can declare additional
  4185 continuity introduction rules with the attribute [cont2cont].
  4186 
  4187 * The continuity simproc now uses a different introduction rule for
  4188 solving continuity subgoals on terms with lambda abstractions.  In
  4189 some rare cases the new simproc may fail to solve subgoals that the
  4190 old one could solve, and "simp add: cont2cont_LAM" may be necessary.
  4191 Potential INCOMPATIBILITY.
  4192 
  4193 * Command 'fixrec': specification syntax now conforms to the general
  4194 format as seen in 'inductive' etc.  See src/HOLCF/ex/Fixrec_ex.thy for
  4195 examples.  INCOMPATIBILITY.
  4196 
  4197 
  4198 *** ZF ***
  4199 
  4200 * Proof of Zorn's Lemma for partial orders.
  4201 
  4202 
  4203 *** ML ***
  4204 
  4205 * Multithreading for Poly/ML 5.1/5.2 is no longer supported, only for
  4206 Poly/ML 5.2.1 or later.  Important note: the TimeLimit facility
  4207 depends on multithreading, so timouts will not work before Poly/ML
  4208 5.2.1!
  4209 
  4210 * High-level support for concurrent ML programming, see
  4211 src/Pure/Cuncurrent.  The data-oriented model of "future values" is
  4212 particularly convenient to organize independent functional
  4213 computations.  The concept of "synchronized variables" provides a
  4214 higher-order interface for components with shared state, avoiding the
  4215 delicate details of mutexes and condition variables.  (Requires
  4216 Poly/ML 5.2.1 or later.)
  4217 
  4218 * ML bindings produced via Isar commands are stored within the Isar
  4219 context (theory or proof).  Consequently, commands like 'use' and 'ML'
  4220 become thread-safe and work with undo as expected (concerning
  4221 top-level bindings, not side-effects on global references).
  4222 INCOMPATIBILITY, need to provide proper Isar context when invoking the
  4223 compiler at runtime; really global bindings need to be given outside a
  4224 theory.  (Requires Poly/ML 5.2 or later.)
  4225 
  4226 * Command 'ML_prf' is analogous to 'ML' but works within a proof
  4227 context.  Top-level ML bindings are stored within the proof context in
  4228 a purely sequential fashion, disregarding the nested proof structure.
  4229 ML bindings introduced by 'ML_prf' are discarded at the end of the
  4230 proof.  (Requires Poly/ML 5.2 or later.)
  4231 
  4232 * Simplified ML attribute and method setup, cf. functions Attrib.setup
  4233 and Method.setup, as well as Isar commands 'attribute_setup' and
  4234 'method_setup'.  INCOMPATIBILITY for 'method_setup', need to simplify
  4235 existing code accordingly, or use plain 'setup' together with old
  4236 Method.add_method.
  4237 
  4238 * Simplified ML oracle interface Thm.add_oracle promotes 'a -> cterm
  4239 to 'a -> thm, while results are always tagged with an authentic oracle
  4240 name.  The Isar command 'oracle' is now polymorphic, no argument type
  4241 is specified.  INCOMPATIBILITY, need to simplify existing oracle code
  4242 accordingly.  Note that extra performance may be gained by producing
  4243 the cterm carefully, avoiding slow Thm.cterm_of.
  4244 
  4245 * Simplified interface for defining document antiquotations via
  4246 ThyOutput.antiquotation, ThyOutput.output, and optionally
  4247 ThyOutput.maybe_pretty_source.  INCOMPATIBILITY, need to simplify user
  4248 antiquotations accordingly, see src/Pure/Thy/thy_output.ML for common
  4249 examples.
  4250 
  4251 * More systematic treatment of long names, abstract name bindings, and
  4252 name space operations.  Basic operations on qualified names have been
  4253 move from structure NameSpace to Long_Name, e.g. Long_Name.base_name,
  4254 Long_Name.append.  Old type bstring has been mostly replaced by
  4255 abstract type binding (see structure Binding), which supports precise
  4256 qualification by packages and local theory targets, as well as proper
  4257 tracking of source positions.  INCOMPATIBILITY, need to wrap old
  4258 bstring values into Binding.name, or better pass through abstract
  4259 bindings everywhere.  See further src/Pure/General/long_name.ML,
  4260 src/Pure/General/binding.ML and src/Pure/General/name_space.ML
  4261 
  4262 * Result facts (from PureThy.note_thms, ProofContext.note_thms,
  4263 LocalTheory.note etc.) now refer to the *full* internal name, not the
  4264 bstring as before.  INCOMPATIBILITY, not detected by ML type-checking!
  4265 
  4266 * Disposed old type and term read functions (Sign.read_def_typ,
  4267 Sign.read_typ, Sign.read_def_terms, Sign.read_term,
  4268 Thm.read_def_cterms, Thm.read_cterm etc.).  INCOMPATIBILITY, should
  4269 use regular Syntax.read_typ, Syntax.read_term, Syntax.read_typ_global,
  4270 Syntax.read_term_global etc.; see also OldGoals.read_term as last
  4271 resort for legacy applications.
  4272 
  4273 * Disposed old declarations, tactics, tactic combinators that refer to
  4274 the simpset or claset of an implicit theory (such as Addsimps,
  4275 Simp_tac, SIMPSET).  INCOMPATIBILITY, should use @{simpset} etc. in
  4276 embedded ML text, or local_simpset_of with a proper context passed as
  4277 explicit runtime argument.
  4278 
  4279 * Rules and tactics that read instantiations (read_instantiate,
  4280 res_inst_tac, thin_tac, subgoal_tac etc.) now demand a proper proof
  4281 context, which is required for parsing and type-checking.  Moreover,
  4282 the variables are specified as plain indexnames, not string encodings
  4283 thereof.  INCOMPATIBILITY.
  4284 
  4285 * Generic Toplevel.add_hook interface allows to analyze the result of
  4286 transactions.  E.g. see src/Pure/ProofGeneral/proof_general_pgip.ML
  4287 for theorem dependency output of transactions resulting in a new
  4288 theory state.
  4289 
  4290 * ML antiquotations: block-structured compilation context indicated by
  4291 \<lbrace> ... \<rbrace>; additional antiquotation forms:
  4292 
  4293   @{binding name}                         - basic name binding
  4294   @{let ?pat = term}                      - term abbreviation (HO matching)
  4295   @{note name = fact}                     - fact abbreviation
  4296   @{thm fact}                             - singleton fact (with attributes)
  4297   @{thms fact}                            - general fact (with attributes)
  4298   @{lemma prop by method}                 - singleton goal
  4299   @{lemma prop by meth1 meth2}            - singleton goal
  4300   @{lemma prop1 ... propN by method}      - general goal
  4301   @{lemma prop1 ... propN by meth1 meth2} - general goal
  4302   @{lemma (open) ...}                     - open derivation
  4303 
  4304 
  4305 *** System ***
  4306 
  4307 * The Isabelle "emacs" tool provides a specific interface to invoke
  4308 Proof General / Emacs, with more explicit failure if that is not
  4309 installed (the old isabelle-interface script silently falls back on
  4310 isabelle-process).  The PROOFGENERAL_HOME setting determines the
  4311 installation location of the Proof General distribution.
  4312 
  4313 * Isabelle/lib/classes/Pure.jar provides basic support to integrate
  4314 the Isabelle process into a JVM/Scala application.  See
  4315 Isabelle/lib/jedit/plugin for a minimal example.  (The obsolete Java
  4316 process wrapper has been discontinued.)
  4317 
  4318 * Added homegrown Isabelle font with unicode layout, see lib/fonts.
  4319 
  4320 * Various status messages (with exact source position information) are
  4321 emitted, if proper markup print mode is enabled.  This allows
  4322 user-interface components to provide detailed feedback on internal
  4323 prover operations.
  4324 
  4325 
  4326 
  4327 New in Isabelle2008 (June 2008)
  4328 -------------------------------
  4329 
  4330 *** General ***
  4331 
  4332 * The Isabelle/Isar Reference Manual (isar-ref) has been reorganized
  4333 and updated, with formally checked references as hyperlinks.
  4334 
  4335 * Theory loader: use_thy (and similar operations) no longer set the
  4336 implicit ML context, which was occasionally hard to predict and in
  4337 conflict with concurrency.  INCOMPATIBILITY, use ML within Isar which
  4338 provides a proper context already.
  4339 
  4340 * Theory loader: old-style ML proof scripts being *attached* to a thy
  4341 file are no longer supported.  INCOMPATIBILITY, regular 'uses' and
  4342 'use' within a theory file will do the job.
  4343 
  4344 * Name space merge now observes canonical order, i.e. the second space
  4345 is inserted into the first one, while existing entries in the first
  4346 space take precedence.  INCOMPATIBILITY in rare situations, may try to
  4347 swap theory imports.
  4348 
  4349 * Syntax: symbol \<chi> is now considered a letter.  Potential
  4350 INCOMPATIBILITY in identifier syntax etc.
  4351 
  4352 * Outer syntax: string tokens no longer admit escaped white space,
  4353 which was an accidental (undocumented) feature.  INCOMPATIBILITY, use
  4354 white space without escapes.
  4355 
  4356 * Outer syntax: string tokens may contain arbitrary character codes
  4357 specified via 3 decimal digits (as in SML).  E.g. "foo\095bar" for
  4358 "foo_bar".
  4359 
  4360 
  4361 *** Pure ***
  4362 
  4363 * Context-dependent token translations.  Default setup reverts locally
  4364 fixed variables, and adds hilite markup for undeclared frees.
  4365 
  4366 * Unused theorems can be found using the new command 'unused_thms'.
  4367 There are three ways of invoking it:
  4368 
  4369 (1) unused_thms
  4370      Only finds unused theorems in the current theory.
  4371 
  4372 (2) unused_thms thy_1 ... thy_n -
  4373      Finds unused theorems in the current theory and all of its ancestors,
  4374      excluding the theories thy_1 ... thy_n and all of their ancestors.
  4375 
  4376 (3) unused_thms thy_1 ... thy_n - thy'_1 ... thy'_m
  4377      Finds unused theorems in the theories thy'_1 ... thy'_m and all of
  4378      their ancestors, excluding the theories thy_1 ... thy_n and all of
  4379      their ancestors.
  4380 
  4381 In order to increase the readability of the list produced by
  4382 unused_thms, theorems that have been created by a particular instance
  4383 of a theory command such as 'inductive' or 'function' are considered
  4384 to belong to the same "group", meaning that if at least one theorem in
  4385 this group is used, the other theorems in the same group are no longer
  4386 reported as unused.  Moreover, if all theorems in the group are
  4387 unused, only one theorem in the group is displayed.
  4388 
  4389 Note that proof objects have to be switched on in order for
  4390 unused_thms to work properly (i.e. !proofs must be >= 1, which is
  4391 usually the case when using Proof General with the default settings).
  4392 
  4393 * Authentic naming of facts disallows ad-hoc overwriting of previous
  4394 theorems within the same name space.  INCOMPATIBILITY, need to remove
  4395 duplicate fact bindings, or even accidental fact duplications.  Note
  4396 that tools may maintain dynamically scoped facts systematically, using
  4397 PureThy.add_thms_dynamic.
  4398 
  4399 * Command 'hide' now allows to hide from "fact" name space as well.
  4400 
  4401 * Eliminated destructive theorem database, simpset, claset, and
  4402 clasimpset.  Potential INCOMPATIBILITY, really need to observe linear
  4403 update of theories within ML code.
  4404 
  4405 * Eliminated theory ProtoPure and CPure, leaving just one Pure theory.
  4406 INCOMPATIBILITY, object-logics depending on former Pure require
  4407 additional setup PureThy.old_appl_syntax_setup; object-logics
  4408 depending on former CPure need to refer to Pure.
  4409 
  4410 * Commands 'use' and 'ML' are now purely functional, operating on
  4411 theory/local_theory.  Removed former 'ML_setup' (on theory), use 'ML'
  4412 instead.  Added 'ML_val' as mere diagnostic replacement for 'ML'.
  4413 INCOMPATIBILITY.
  4414 
  4415 * Command 'setup': discontinued implicit version with ML reference.
  4416 
  4417 * Instantiation target allows for simultaneous specification of class
  4418 instance operations together with an instantiation proof.
  4419 Type-checking phase allows to refer to class operations uniformly.
  4420 See src/HOL/Complex/Complex.thy for an Isar example and
  4421 src/HOL/Library/Eval.thy for an ML example.
  4422 
  4423 * Indexing of literal facts: be more serious about including only
  4424 facts from the visible specification/proof context, but not the
  4425 background context (locale etc.).  Affects `prop` notation and method
  4426 "fact".  INCOMPATIBILITY: need to name facts explicitly in rare
  4427 situations.
  4428 
  4429 * Method "cases", "induct", "coinduct": removed obsolete/undocumented
  4430 "(open)" option, which used to expose internal bound variables to the
  4431 proof text.
  4432 
  4433 * Isar statements: removed obsolete case "rule_context".
  4434 INCOMPATIBILITY, better use explicit fixes/assumes.
  4435 
  4436 * Locale proofs: default proof step now includes 'unfold_locales';
  4437 hence 'proof' without argument may be used to unfold locale
  4438 predicates.
  4439 
  4440 
  4441 *** Document preparation ***
  4442 
  4443 * Simplified pdfsetup.sty: color/hyperref is used unconditionally for
  4444 both pdf and dvi (hyperlinks usually work in xdvi as well); removed
  4445 obsolete thumbpdf setup (contemporary PDF viewers do this on the
  4446 spot); renamed link color from "darkblue" to "linkcolor" (default
  4447 value unchanged, can be redefined via \definecolor); no longer sets
  4448 "a4paper" option (unnecessary or even intrusive).
  4449 
  4450 * Antiquotation @{lemma A method} proves proposition A by the given
  4451 method (either a method name or a method name plus (optional) method
  4452 arguments in parentheses) and prints A just like @{prop A}.
  4453 
  4454 
  4455 *** HOL ***
  4456 
  4457 * New primrec package.  Specification syntax conforms in style to
  4458 definition/function/....  No separate induction rule is provided.  The
  4459 "primrec" command distinguishes old-style and new-style specifications
  4460 by syntax.  The former primrec package is now named OldPrimrecPackage.
  4461 When adjusting theories, beware: constants stemming from new-style
  4462 primrec specifications have authentic syntax.
  4463 
  4464 * Metis prover is now an order of magnitude faster, and also works
  4465 with multithreading.
  4466 
  4467 * Metis: the maximum number of clauses that can be produced from a
  4468 theorem is now given by the attribute max_clauses.  Theorems that
  4469 exceed this number are ignored, with a warning printed.
  4470 
  4471 * Sledgehammer no longer produces structured proofs by default. To
  4472 enable, declare [[sledgehammer_full = true]].  Attributes
  4473 reconstruction_modulus, reconstruction_sorts renamed
  4474 sledgehammer_modulus, sledgehammer_sorts.  INCOMPATIBILITY.
  4475 
  4476 * Method "induct_scheme" derives user-specified induction rules
  4477 from well-founded induction and completeness of patterns. This factors
  4478 out some operations that are done internally by the function package
  4479 and makes them available separately.  See
  4480 src/HOL/ex/Induction_Scheme.thy for examples.
  4481 
  4482 * More flexible generation of measure functions for termination
  4483 proofs: Measure functions can be declared by proving a rule of the
  4484 form "is_measure f" and giving it the [measure_function] attribute.
  4485 The "is_measure" predicate is logically meaningless (always true), and
  4486 just guides the heuristic.  To find suitable measure functions, the
  4487 termination prover sets up the goal "is_measure ?f" of the appropriate
  4488 type and generates all solutions by prolog-style backwards proof using
  4489 the declared rules.
  4490 
  4491 This setup also deals with rules like 
  4492 
  4493   "is_measure f ==> is_measure (list_size f)"
  4494 
  4495 which accommodates nested datatypes that recurse through lists.
  4496 Similar rules are predeclared for products and option types.
  4497 
  4498 * Turned the type of sets "'a set" into an abbreviation for "'a => bool"
  4499 
  4500   INCOMPATIBILITIES:
  4501 
  4502   - Definitions of overloaded constants on sets have to be replaced by
  4503     definitions on => and bool.
  4504 
  4505   - Some definitions of overloaded operators on sets can now be proved
  4506     using the definitions of the operators on => and bool.  Therefore,
  4507     the following theorems have been renamed:
  4508 
  4509       subset_def   -> subset_eq
  4510       psubset_def  -> psubset_eq
  4511       set_diff_def -> set_diff_eq
  4512       Compl_def    -> Compl_eq
  4513       Sup_set_def  -> Sup_set_eq
  4514       Inf_set_def  -> Inf_set_eq
  4515       sup_set_def  -> sup_set_eq
  4516       inf_set_def  -> inf_set_eq
  4517 
  4518   - Due to the incompleteness of the HO unification algorithm, some
  4519     rules such as subst may require manual instantiation, if some of
  4520     the unknowns in the rule is a set.
  4521 
  4522   - Higher order unification and forward proofs:
  4523     The proof pattern
  4524 
  4525       have "P (S::'a set)" <...>
  4526       then have "EX S. P S" ..
  4527 
  4528     no longer works (due to the incompleteness of the HO unification
  4529     algorithm) and must be replaced by the pattern
  4530 
  4531       have "EX S. P S"
  4532       proof
  4533         show "P S" <...>
  4534       qed
  4535 
  4536   - Calculational reasoning with subst (or similar rules):
  4537     The proof pattern
  4538 
  4539       have "P (S::'a set)" <...>
  4540       also have "S = T" <...>
  4541       finally have "P T" .
  4542 
  4543     no longer works (for similar reasons as the previous example) and
  4544     must be replaced by something like
  4545 
  4546       have "P (S::'a set)" <...>
  4547       moreover have "S = T" <...>
  4548       ultimately have "P T" by simp
  4549 
  4550   - Tactics or packages written in ML code:
  4551     Code performing pattern matching on types via
  4552 
  4553       Type ("set", [T]) => ...
  4554 
  4555     must be rewritten. Moreover, functions like strip_type or
  4556     binder_types no longer return the right value when applied to a
  4557     type of the form
  4558 
  4559       T1 => ... => Tn => U => bool
  4560 
  4561     rather than
  4562 
  4563       T1 => ... => Tn => U set
  4564 
  4565 * Merged theories Wellfounded_Recursion, Accessible_Part and
  4566 Wellfounded_Relations to theory Wellfounded.
  4567 
  4568 * Explicit class "eq" for executable equality.  INCOMPATIBILITY.
  4569 
  4570 * Class finite no longer treats UNIV as class parameter.  Use class
  4571 enum from theory Library/Enum instead to achieve a similar effect.
  4572 INCOMPATIBILITY.
  4573 
  4574 * Theory List: rule list_induct2 now has explicitly named cases "Nil"
  4575 and "Cons".  INCOMPATIBILITY.
  4576 
  4577 * HOL (and FOL): renamed variables in rules imp_elim and swap.
  4578 Potential INCOMPATIBILITY.
  4579 
  4580 * Theory Product_Type: duplicated lemmas split_Pair_apply and
  4581 injective_fst_snd removed, use split_eta and prod_eqI instead.
  4582 Renamed upd_fst to apfst and upd_snd to apsnd.  INCOMPATIBILITY.
  4583 
  4584 * Theory Nat: removed redundant lemmas that merely duplicate lemmas of
  4585 the same name in theory Orderings:
  4586 
  4587   less_trans
  4588   less_linear
  4589   le_imp_less_or_eq
  4590   le_less_trans
  4591   less_le_trans
  4592   less_not_sym
  4593   less_asym
  4594 
  4595 Renamed less_imp_le to less_imp_le_nat, and less_irrefl to
  4596 less_irrefl_nat.  Potential INCOMPATIBILITY due to more general types
  4597 and different variable names.
  4598 
  4599 * Library/Option_ord.thy: Canonical order on option type.
  4600 
  4601 * Library/RBT.thy: Red-black trees, an efficient implementation of
  4602 finite maps.
  4603 
  4604 * Library/Countable.thy: Type class for countable types.
  4605 
  4606 * Theory Int: The representation of numerals has changed.  The infix
  4607 operator BIT and the bit datatype with constructors B0 and B1 have
  4608 disappeared.  INCOMPATIBILITY, use "Int.Bit0 x" and "Int.Bit1 y" in
  4609 place of "x BIT bit.B0" and "y BIT bit.B1", respectively.  Theorems
  4610 involving BIT, B0, or B1 have been renamed with "Bit0" or "Bit1"
  4611 accordingly.
  4612 
  4613 * Theory Nat: definition of <= and < on natural numbers no longer
  4614 depend on well-founded relations.  INCOMPATIBILITY.  Definitions
  4615 le_def and less_def have disappeared.  Consider lemmas not_less
  4616 [symmetric, where ?'a = nat] and less_eq [symmetric] instead.
  4617 
  4618 * Theory Finite_Set: locales ACf, ACe, ACIf, ACIfSL and ACIfSLlin
  4619 (whose purpose mainly is for various fold_set functionals) have been
  4620 abandoned in favor of the existing algebraic classes
  4621 ab_semigroup_mult, comm_monoid_mult, ab_semigroup_idem_mult,
  4622 lower_semilattice (resp. upper_semilattice) and linorder.
  4623 INCOMPATIBILITY.
  4624 
  4625 * Theory Transitive_Closure: induct and cases rules now declare proper
  4626 case_names ("base" and "step").  INCOMPATIBILITY.
  4627 
  4628 * Theorem Inductive.lfp_ordinal_induct generalized to complete
  4629 lattices.  The form set-specific version is available as
  4630 Inductive.lfp_ordinal_induct_set.
  4631 
  4632 * Renamed theorems "power.simps" to "power_int.simps".
  4633 INCOMPATIBILITY.
  4634 
  4635 * Class semiring_div provides basic abstract properties of semirings
  4636 with division and modulo operations.  Subsumes former class dvd_mod.
  4637 
  4638 * Merged theories IntDef, Numeral and IntArith into unified theory
  4639 Int.  INCOMPATIBILITY.
  4640 
  4641 * Theory Library/Code_Index: type "index" now represents natural
  4642 numbers rather than integers.  INCOMPATIBILITY.
  4643 
  4644 * New class "uminus" with operation "uminus" (split of from class
  4645 "minus" which now only has operation "minus", binary).
  4646 INCOMPATIBILITY.
  4647 
  4648 * Constants "card", "internal_split", "option_map" now with authentic
  4649 syntax.  INCOMPATIBILITY.
  4650 
  4651 * Definitions subset_def, psubset_def, set_diff_def, Compl_def,
  4652 le_bool_def, less_bool_def, le_fun_def, less_fun_def, inf_bool_def,
  4653 sup_bool_def, Inf_bool_def, Sup_bool_def, inf_fun_def, sup_fun_def,
  4654 Inf_fun_def, Sup_fun_def, inf_set_def, sup_set_def, Inf_set_def,
  4655 Sup_set_def, le_def, less_def, option_map_def now with object
  4656 equality.  INCOMPATIBILITY.
  4657 
  4658 * Records. Removed K_record, and replaced it by pure lambda term
  4659 %x. c. The simplifier setup is now more robust against eta expansion.
  4660 INCOMPATIBILITY: in cases explicitly referring to K_record.
  4661 
  4662 * Library/Multiset: {#a, b, c#} abbreviates {#a#} + {#b#} + {#c#}.
  4663 
  4664 * Library/ListVector: new theory of arithmetic vector operations.
  4665 
  4666 * Library/Order_Relation: new theory of various orderings as sets of
  4667 pairs.  Defines preorders, partial orders, linear orders and
  4668 well-orders on sets and on types.
  4669 
  4670 
  4671 *** ZF ***
  4672 
  4673 * Renamed some theories to allow to loading both ZF and HOL in the
  4674 same session:
  4675 
  4676   Datatype  -> Datatype_ZF
  4677   Inductive -> Inductive_ZF
  4678   Int       -> Int_ZF
  4679   IntDiv    -> IntDiv_ZF
  4680   Nat       -> Nat_ZF
  4681   List      -> List_ZF
  4682   Main      -> Main_ZF
  4683 
  4684 INCOMPATIBILITY: ZF theories that import individual theories below
  4685 Main might need to be adapted.  Regular theory Main is still
  4686 available, as trivial extension of Main_ZF.
  4687 
  4688 
  4689 *** ML ***
  4690 
  4691 * ML within Isar: antiquotation @{const name} or @{const
  4692 name(typargs)} produces statically-checked Const term.
  4693 
  4694 * Functor NamedThmsFun: data is available to the user as dynamic fact
  4695 (of the same name).  Removed obsolete print command.
  4696 
  4697 * Removed obsolete "use_legacy_bindings" function.
  4698 
  4699 * The ``print mode'' is now a thread-local value derived from a global
  4700 template (the former print_mode reference), thus access becomes
  4701 non-critical.  The global print_mode reference is for session
  4702 management only; user-code should use print_mode_value,
  4703 print_mode_active, PrintMode.setmp etc.  INCOMPATIBILITY.
  4704 
  4705 * Functions system/system_out provide a robust way to invoke external
  4706 shell commands, with propagation of interrupts (requires Poly/ML
  4707 5.2.1).  Do not use OS.Process.system etc. from the basis library!
  4708 
  4709 
  4710 *** System ***
  4711 
  4712 * Default settings: PROOFGENERAL_OPTIONS no longer impose xemacs ---
  4713 in accordance with Proof General 3.7, which prefers GNU emacs.
  4714 
  4715 * isatool tty runs Isabelle process with plain tty interaction;
  4716 optional line editor may be specified via ISABELLE_LINE_EDITOR
  4717 setting, the default settings attempt to locate "ledit" and "rlwrap".
  4718 
  4719 * isatool browser now works with Cygwin as well, using general
  4720 "javapath" function defined in Isabelle process environment.
  4721 
  4722 * YXML notation provides a simple and efficient alternative to
  4723 standard XML transfer syntax.  See src/Pure/General/yxml.ML and
  4724 isatool yxml as described in the Isabelle system manual.
  4725 
  4726 * JVM class isabelle.IsabelleProcess (located in Isabelle/lib/classes)
  4727 provides general wrapper for managing an Isabelle process in a robust
  4728 fashion, with ``cooked'' output from stdin/stderr.
  4729 
  4730 * Rudimentary Isabelle plugin for jEdit (see Isabelle/lib/jedit),
  4731 based on Isabelle/JVM process wrapper (see Isabelle/lib/classes).
  4732 
  4733 * Removed obsolete THIS_IS_ISABELLE_BUILD feature.  NB: the documented
  4734 way of changing the user's settings is via
  4735 ISABELLE_HOME_USER/etc/settings, which is a fully featured bash
  4736 script.
  4737 
  4738 * Multithreading.max_threads := 0 refers to the number of actual CPU
  4739 cores of the underlying machine, which is a good starting point for
  4740 optimal performance tuning.  The corresponding usedir option -M allows
  4741 "max" as an alias for "0".  WARNING: does not work on certain versions
  4742 of Mac OS (with Poly/ML 5.1).
  4743 
  4744 * isabelle-process: non-ML sessions are run with "nice", to reduce the
  4745 adverse effect of Isabelle flooding interactive front-ends (notably
  4746 ProofGeneral / XEmacs).
  4747 
  4748 
  4749 
  4750 New in Isabelle2007 (November 2007)
  4751 -----------------------------------
  4752 
  4753 *** General ***
  4754 
  4755 * More uniform information about legacy features, notably a
  4756 warning/error of "Legacy feature: ...", depending on the state of the
  4757 tolerate_legacy_features flag (default true). FUTURE INCOMPATIBILITY:
  4758 legacy features will disappear eventually.
  4759 
  4760 * Theory syntax: the header format ``theory A = B + C:'' has been
  4761 discontinued in favour of ``theory A imports B C begin''.  Use isatool
  4762 fixheaders to convert existing theory files.  INCOMPATIBILITY.
  4763 
  4764 * Theory syntax: the old non-Isar theory file format has been
  4765 discontinued altogether.  Note that ML proof scripts may still be used
  4766 with Isar theories; migration is usually quite simple with the ML
  4767 function use_legacy_bindings.  INCOMPATIBILITY.
  4768 
  4769 * Theory syntax: some popular names (e.g. 'class', 'declaration',
  4770 'fun', 'help', 'if') are now keywords.  INCOMPATIBILITY, use double
  4771 quotes.
  4772 
  4773 * Theory loader: be more serious about observing the static theory
  4774 header specifications (including optional directories), but not the
  4775 accidental file locations of previously successful loads.  The strict
  4776 update policy of former update_thy is now already performed by
  4777 use_thy, so the former has been removed; use_thys updates several
  4778 theories simultaneously, just as 'imports' within a theory header
  4779 specification, but without merging the results.  Potential
  4780 INCOMPATIBILITY: may need to refine theory headers and commands
  4781 ROOT.ML which depend on load order.
  4782 
  4783 * Theory loader: optional support for content-based file
  4784 identification, instead of the traditional scheme of full physical
  4785 path plus date stamp; configured by the ISABELLE_FILE_IDENT setting
  4786 (cf. the system manual).  The new scheme allows to work with
  4787 non-finished theories in persistent session images, such that source
  4788 files may be moved later on without requiring reloads.
  4789 
  4790 * Theory loader: old-style ML proof scripts being *attached* to a thy
  4791 file (with the same base name as the theory) are considered a legacy
  4792 feature, which will disappear eventually. Even now, the theory loader
  4793 no longer maintains dependencies on such files.
  4794 
  4795 * Syntax: the scope for resolving ambiguities via type-inference is
  4796 now limited to individual terms, instead of whole simultaneous
  4797 specifications as before. This greatly reduces the complexity of the
  4798 syntax module and improves flexibility by separating parsing and
  4799 type-checking. INCOMPATIBILITY: additional type-constraints (explicit
  4800 'fixes' etc.) are required in rare situations.
  4801 
  4802 * Syntax: constants introduced by new-style packages ('definition',
  4803 'abbreviation' etc.) are passed through the syntax module in
  4804 ``authentic mode''. This means that associated mixfix annotations
  4805 really stick to such constants, independently of potential name space
  4806 ambiguities introduced later on. INCOMPATIBILITY: constants in parse
  4807 trees are represented slightly differently, may need to adapt syntax
  4808 translations accordingly. Use CONST marker in 'translations' and
  4809 @{const_syntax} antiquotation in 'parse_translation' etc.
  4810 
  4811 * Legacy goal package: reduced interface to the bare minimum required
  4812 to keep existing proof scripts running.  Most other user-level
  4813 functions are now part of the OldGoals structure, which is *not* open
  4814 by default (consider isatool expandshort before open OldGoals).
  4815 Removed top_sg, prin, printyp, pprint_term/typ altogether, because
  4816 these tend to cause confusion about the actual goal (!) context being
  4817 used here, which is not necessarily the same as the_context().
  4818 
  4819 * Command 'find_theorems': supports "*" wild-card in "name:"
  4820 criterion; "with_dups" option.  Certain ProofGeneral versions might
  4821 support a specific search form (see ProofGeneral/CHANGES).
  4822 
  4823 * The ``prems limit'' option (cf. ProofContext.prems_limit) is now -1
  4824 by default, which means that "prems" (and also "fixed variables") are
  4825 suppressed from proof state output.  Note that the ProofGeneral
  4826 settings mechanism allows to change and save options persistently, but
  4827 older versions of Isabelle will fail to start up if a negative prems
  4828 limit is imposed.
  4829 
  4830 * Local theory targets may be specified by non-nested blocks of
  4831 ``context/locale/class ... begin'' followed by ``end''.  The body may
  4832 contain definitions, theorems etc., including any derived mechanism
  4833 that has been implemented on top of these primitives.  This concept
  4834 generalizes the existing ``theorem (in ...)'' towards more versatility
  4835 and scalability.
  4836 
  4837 * Proof General interface: proper undo of final 'end' command;
  4838 discontinued Isabelle/classic mode (ML proof scripts).
  4839 
  4840 
  4841 *** Document preparation ***
  4842 
  4843 * Added antiquotation @{theory name} which prints the given name,
  4844 after checking that it refers to a valid ancestor theory in the
  4845 current context.
  4846 
  4847 * Added antiquotations @{ML_type text} and @{ML_struct text} which
  4848 check the given source text as ML type/structure, printing verbatim.
  4849 
  4850 * Added antiquotation @{abbrev "c args"} which prints the abbreviation
  4851 "c args == rhs" given in the current context.  (Any number of
  4852 arguments may be given on the LHS.)
  4853 
  4854 
  4855 *** Pure ***
  4856 
  4857 * The 'class' package offers a combination of axclass and locale to
  4858 achieve Haskell-like type classes in Isabelle.  Definitions and
  4859 theorems within a class context produce both relative results (with
  4860 implicit parameters according to the locale context), and polymorphic
  4861 constants with qualified polymorphism (according to the class
  4862 context).  Within the body context of a 'class' target, a separate
  4863 syntax layer ("user space type system") takes care of converting
  4864 between global polymorphic consts and internal locale representation.
  4865 See src/HOL/ex/Classpackage.thy for examples (as well as main HOL).
  4866 "isatool doc classes" provides a tutorial.
  4867 
  4868 * Generic code generator framework allows to generate executable
  4869 code for ML and Haskell (including Isabelle classes).  A short usage
  4870 sketch:
  4871 
  4872     internal compilation:
  4873         export_code <list of constants (term syntax)> in SML
  4874     writing SML code to a file:
  4875         export_code <list of constants (term syntax)> in SML <filename>
  4876     writing OCaml code to a file:
  4877         export_code <list of constants (term syntax)> in OCaml <filename>
  4878     writing Haskell code to a bunch of files:
  4879         export_code <list of constants (term syntax)> in Haskell <filename>
  4880 
  4881     evaluating closed propositions to True/False using code generation:
  4882         method ``eval''
  4883 
  4884 Reasonable default setup of framework in HOL.
  4885 
  4886 Theorem attributs for selecting and transforming function equations theorems:
  4887 
  4888     [code fun]:        select a theorem as function equation for a specific constant
  4889     [code fun del]:    deselect a theorem as function equation for a specific constant
  4890     [code inline]:     select an equation theorem for unfolding (inlining) in place
  4891     [code inline del]: deselect an equation theorem for unfolding (inlining) in place
  4892 
  4893 User-defined serializations (target in {SML, OCaml, Haskell}):
  4894 
  4895     code_const <and-list of constants (term syntax)>
  4896       {(target) <and-list of const target syntax>}+
  4897 
  4898     code_type <and-list of type constructors>
  4899       {(target) <and-list of type target syntax>}+
  4900 
  4901     code_instance <and-list of instances>
  4902       {(target)}+
  4903         where instance ::= <type constructor> :: <class>
  4904 
  4905     code_class <and_list of classes>
  4906       {(target) <and-list of class target syntax>}+
  4907         where class target syntax ::= <class name> {where {<classop> == <target syntax>}+}?
  4908 
  4909 code_instance and code_class only are effective to target Haskell.
  4910 
  4911 For example usage see src/HOL/ex/Codegenerator.thy and
  4912 src/HOL/ex/Codegenerator_Pretty.thy.  A separate tutorial on code
  4913 generation from Isabelle/HOL theories is available via "isatool doc
  4914 codegen".
  4915 
  4916 * Code generator: consts in 'consts_code' Isar commands are now
  4917 referred to by usual term syntax (including optional type
  4918 annotations).
  4919 
  4920 * Command 'no_translations' removes translation rules from theory
  4921 syntax.
  4922 
  4923 * Overloaded definitions are now actually checked for acyclic
  4924 dependencies.  The overloading scheme is slightly more general than
  4925 that of Haskell98, although Isabelle does not demand an exact
  4926 correspondence to type class and instance declarations.
  4927 INCOMPATIBILITY, use ``defs (unchecked overloaded)'' to admit more
  4928 exotic versions of overloading -- at the discretion of the user!
  4929 
  4930 Polymorphic constants are represented via type arguments, i.e. the
  4931 instantiation that matches an instance against the most general
  4932 declaration given in the signature.  For example, with the declaration
  4933 c :: 'a => 'a => 'a, an instance c :: nat => nat => nat is represented
  4934 as c(nat).  Overloading is essentially simultaneous structural
  4935 recursion over such type arguments.  Incomplete specification patterns
  4936 impose global constraints on all occurrences, e.g. c('a * 'a) on the
  4937 LHS means that more general c('a * 'b) will be disallowed on any RHS.
  4938 Command 'print_theory' outputs the normalized system of recursive
  4939 equations, see section "definitions".
  4940 
  4941 * Configuration options are maintained within the theory or proof
  4942 context (with name and type bool/int/string), providing a very simple
  4943 interface to a poor-man's version of general context data.  Tools may
  4944 declare options in ML (e.g. using Attrib.config_int) and then refer to
  4945 these values using Config.get etc.  Users may change options via an
  4946 associated attribute of the same name.  This form of context
  4947 declaration works particularly well with commands 'declare' or
  4948 'using', for example ``declare [[foo = 42]]''.  Thus it has become
  4949 very easy to avoid global references, which would not observe Isar
  4950 toplevel undo/redo and fail to work with multithreading.
  4951 
  4952 Various global ML references of Pure and HOL have been turned into
  4953 configuration options:
  4954 
  4955   Unify.search_bound		unify_search_bound
  4956   Unify.trace_bound		unify_trace_bound
  4957   Unify.trace_simp		unify_trace_simp
  4958   Unify.trace_types		unify_trace_types
  4959   Simplifier.simp_depth_limit	simp_depth_limit
  4960   Blast.depth_limit		blast_depth_limit
  4961   DatatypeProp.dtK		datatype_distinctness_limit
  4962   fast_arith_neq_limit  	fast_arith_neq_limit
  4963   fast_arith_split_limit	fast_arith_split_limit
  4964 
  4965 * Named collections of theorems may be easily installed as context
  4966 data using the functor NamedThmsFun (see also
  4967 src/Pure/Tools/named_thms.ML).  The user may add or delete facts via
  4968 attributes; there is also a toplevel print command.  This facility is
  4969 just a common case of general context data, which is the preferred way
  4970 for anything more complex than just a list of facts in canonical
  4971 order.
  4972 
  4973 * Isar: command 'declaration' augments a local theory by generic
  4974 declaration functions written in ML.  This enables arbitrary content
  4975 being added to the context, depending on a morphism that tells the
  4976 difference of the original declaration context wrt. the application
  4977 context encountered later on.
  4978 
  4979 * Isar: proper interfaces for simplification procedures.  Command
  4980 'simproc_setup' declares named simprocs (with match patterns, and body
  4981 text in ML).  Attribute "simproc" adds/deletes simprocs in the current
  4982 context.  ML antiquotation @{simproc name} retrieves named simprocs.
  4983 
  4984 * Isar: an extra pair of brackets around attribute declarations
  4985 abbreviates a theorem reference involving an internal dummy fact,
  4986 which will be ignored later --- only the effect of the attribute on
  4987 the background context will persist.  This form of in-place
  4988 declarations is particularly useful with commands like 'declare' and
  4989 'using', for example ``have A using [[simproc a]] by simp''.
  4990 
  4991 * Isar: method "assumption" (and implicit closing of subproofs) now
  4992 takes simple non-atomic goal assumptions into account: after applying
  4993 an assumption as a rule the resulting subgoals are solved by atomic
  4994 assumption steps.  This is particularly useful to finish 'obtain'
  4995 goals, such as "!!x. (!!x. P x ==> thesis) ==> P x ==> thesis",
  4996 without referring to the original premise "!!x. P x ==> thesis" in the
  4997 Isar proof context.  POTENTIAL INCOMPATIBILITY: method "assumption" is
  4998 more permissive.
  4999 
  5000 * Isar: implicit use of prems from the Isar proof context is
  5001 considered a legacy feature.  Common applications like ``have A .''
  5002 may be replaced by ``have A by fact'' or ``note `A`''.  In general,
  5003 referencing facts explicitly here improves readability and
  5004 maintainability of proof texts.
  5005 
  5006 * Isar: improper proof element 'guess' is like 'obtain', but derives
  5007 the obtained context from the course of reasoning!  For example:
  5008 
  5009   assume "EX x y. A x & B y"   -- "any previous fact"
  5010   then guess x and y by clarify
  5011 
  5012 This technique is potentially adventurous, depending on the facts and
  5013 proof tools being involved here.
  5014 
  5015 * Isar: known facts from the proof context may be specified as literal
  5016 propositions, using ASCII back-quote syntax.  This works wherever
  5017 named facts used to be allowed so far, in proof commands, proof
  5018 methods, attributes etc.  Literal facts are retrieved from the context
  5019 according to unification of type and term parameters.  For example,
  5020 provided that "A" and "A ==> B" and "!!x. P x ==> Q x" are known
  5021 theorems in the current context, then these are valid literal facts:
  5022 `A` and `A ==> B` and `!!x. P x ==> Q x" as well as `P a ==> Q a` etc.
  5023 
  5024 There is also a proof method "fact" which does the same composition
  5025 for explicit goal states, e.g. the following proof texts coincide with
  5026 certain special cases of literal facts:
  5027 
  5028   have "A" by fact                 ==  note `A`
  5029   have "A ==> B" by fact           ==  note `A ==> B`
  5030   have "!!x. P x ==> Q x" by fact  ==  note `!!x. P x ==> Q x`
  5031   have "P a ==> Q a" by fact       ==  note `P a ==> Q a`
  5032 
  5033 * Isar: ":" (colon) is no longer a symbolic identifier character in
  5034 outer syntax.  Thus symbolic identifiers may be used without
  5035 additional white space in declarations like this: ``assume *: A''.
  5036 
  5037 * Isar: 'print_facts' prints all local facts of the current context,
  5038 both named and unnamed ones.
  5039 
  5040 * Isar: 'def' now admits simultaneous definitions, e.g.:
  5041 
  5042   def x == "t" and y == "u"
  5043 
  5044 * Isar: added command 'unfolding', which is structurally similar to
  5045 'using', but affects both the goal state and facts by unfolding given
  5046 rewrite rules.  Thus many occurrences of the 'unfold' method or
  5047 'unfolded' attribute may be replaced by first-class proof text.
  5048 
  5049 * Isar: methods 'unfold' / 'fold', attributes 'unfolded' / 'folded',
  5050 and command 'unfolding' now all support object-level equalities
  5051 (potentially conditional).  The underlying notion of rewrite rule is
  5052 analogous to the 'rule_format' attribute, but *not* that of the
  5053 Simplifier (which is usually more generous).
  5054 
  5055 * Isar: the new attribute [rotated n] (default n = 1) rotates the
  5056 premises of a theorem by n. Useful in conjunction with drule.
  5057 
  5058 * Isar: the goal restriction operator [N] (default N = 1) evaluates a
  5059 method expression within a sandbox consisting of the first N
  5060 sub-goals, which need to exist.  For example, ``simp_all [3]''
  5061 simplifies the first three sub-goals, while (rule foo, simp_all)[]
  5062 simplifies all new goals that emerge from applying rule foo to the
  5063 originally first one.
  5064 
  5065 * Isar: schematic goals are no longer restricted to higher-order
  5066 patterns; e.g. ``lemma "?P(?x)" by (rule TrueI)'' now works as
  5067 expected.
  5068 
  5069 * Isar: the conclusion of a long theorem statement is now either
  5070 'shows' (a simultaneous conjunction, as before), or 'obtains'
  5071 (essentially a disjunction of cases with local parameters and
  5072 assumptions).  The latter allows to express general elimination rules
  5073 adequately; in this notation common elimination rules look like this:
  5074 
  5075   lemma exE:    -- "EX x. P x ==> (!!x. P x ==> thesis) ==> thesis"
  5076     assumes "EX x. P x"
  5077     obtains x where "P x"
  5078 
  5079   lemma conjE:  -- "A & B ==> (A ==> B ==> thesis) ==> thesis"
  5080     assumes "A & B"
  5081     obtains A and B
  5082 
  5083   lemma disjE:  -- "A | B ==> (A ==> thesis) ==> (B ==> thesis) ==> thesis"
  5084     assumes "A | B"
  5085     obtains
  5086       A
  5087     | B
  5088 
  5089 The subsequent classical rules even refer to the formal "thesis"
  5090 explicitly:
  5091 
  5092   lemma classical:     -- "(~ thesis ==> thesis) ==> thesis"
  5093     obtains "~ thesis"
  5094 
  5095   lemma Peirce's_Law:  -- "((thesis ==> something) ==> thesis) ==> thesis"
  5096     obtains "thesis ==> something"
  5097 
  5098 The actual proof of an 'obtains' statement is analogous to that of the
  5099 Isar proof element 'obtain', only that there may be several cases.
  5100 Optional case names may be specified in parentheses; these will be
  5101 available both in the present proof and as annotations in the
  5102 resulting rule, for later use with the 'cases' method (cf. attribute
  5103 case_names).
  5104 
  5105 * Isar: the assumptions of a long theorem statement are available as
  5106 "assms" fact in the proof context.  This is more appropriate than the
  5107 (historical) "prems", which refers to all assumptions of the current
  5108 context, including those from the target locale, proof body etc.
  5109 
  5110 * Isar: 'print_statement' prints theorems from the current theory or
  5111 proof context in long statement form, according to the syntax of a
  5112 top-level lemma.
  5113 
  5114 * Isar: 'obtain' takes an optional case name for the local context
  5115 introduction rule (default "that").
  5116 
  5117 * Isar: removed obsolete 'concl is' patterns.  INCOMPATIBILITY, use
  5118 explicit (is "_ ==> ?foo") in the rare cases where this still happens
  5119 to occur.
  5120 
  5121 * Pure: syntax "CONST name" produces a fully internalized constant
  5122 according to the current context.  This is particularly useful for
  5123 syntax translations that should refer to internal constant
  5124 representations independently of name spaces.
  5125 
  5126 * Pure: syntax constant for foo (binder "FOO ") is called "foo_binder"
  5127 instead of "FOO ". This allows multiple binder declarations to coexist
  5128 in the same context.  INCOMPATIBILITY.
  5129 
  5130 * Isar/locales: 'notation' provides a robust interface to the 'syntax'
  5131 primitive that also works in a locale context (both for constants and
  5132 fixed variables). Type declaration and internal syntactic representation
  5133 of given constants retrieved from the context. Likewise, the
  5134 'no_notation' command allows to remove given syntax annotations from the
  5135 current context.
  5136 
  5137 * Isar/locales: new derived specification elements 'axiomatization',
  5138 'definition', 'abbreviation', which support type-inference, admit
  5139 object-level specifications (equality, equivalence).  See also the
  5140 isar-ref manual.  Examples:
  5141 
  5142   axiomatization
  5143     eq  (infix "===" 50) where
  5144     eq_refl: "x === x" and eq_subst: "x === y ==> P x ==> P y"
  5145 
  5146   definition "f x y = x + y + 1"
  5147   definition g where "g x = f x x"
  5148 
  5149   abbreviation
  5150     neq  (infix "=!=" 50) where
  5151     "x =!= y == ~ (x === y)"
  5152 
  5153 These specifications may be also used in a locale context.  Then the
  5154 constants being introduced depend on certain fixed parameters, and the
  5155 constant name is qualified by the locale base name.  An internal
  5156 abbreviation takes care for convenient input and output, making the
  5157 parameters implicit and using the original short name.  See also
  5158 src/HOL/ex/Abstract_NAT.thy for an example of deriving polymorphic
  5159 entities from a monomorphic theory.
  5160 
  5161 Presently, abbreviations are only available 'in' a target locale, but
  5162 not inherited by general import expressions.  Also note that
  5163 'abbreviation' may be used as a type-safe replacement for 'syntax' +
  5164 'translations' in common applications.  The "no_abbrevs" print mode
  5165 prevents folding of abbreviations in term output.
  5166 
  5167 Concrete syntax is attached to specified constants in internal form,
  5168 independently of name spaces.  The parse tree representation is
  5169 slightly different -- use 'notation' instead of raw 'syntax', and
  5170 'translations' with explicit "CONST" markup to accommodate this.
  5171 
  5172 * Pure/Isar: unified syntax for new-style specification mechanisms
  5173 (e.g.  'definition', 'abbreviation', or 'inductive' in HOL) admits
  5174 full type inference and dummy patterns ("_").  For example:
  5175 
  5176   definition "K x _ = x"
  5177 
  5178   inductive conj for A B
  5179   where "A ==> B ==> conj A B"
  5180 
  5181 * Pure: command 'print_abbrevs' prints all constant abbreviations of
  5182 the current context.  Print mode "no_abbrevs" prevents inversion of
  5183 abbreviations on output.
  5184 
  5185 * Isar/locales: improved parameter handling: use of locales "var" and
  5186 "struct" no longer necessary; - parameter renamings are no longer
  5187 required to be injective.  For example, this allows to define
  5188 endomorphisms as locale endom = homom mult mult h.
  5189 
  5190 * Isar/locales: changed the way locales with predicates are defined.
  5191 Instead of accumulating the specification, the imported expression is
  5192 now an interpretation.  INCOMPATIBILITY: different normal form of
  5193 locale expressions.  In particular, in interpretations of locales with
  5194 predicates, goals repesenting already interpreted fragments are not
  5195 removed automatically.  Use methods `intro_locales' and
  5196 `unfold_locales'; see below.
  5197 
  5198 * Isar/locales: new methods `intro_locales' and `unfold_locales'
  5199 provide backward reasoning on locales predicates.  The methods are
  5200 aware of interpretations and discharge corresponding goals.
  5201 `intro_locales' is less aggressive then `unfold_locales' and does not
  5202 unfold predicates to assumptions.
  5203 
  5204 * Isar/locales: the order in which locale fragments are accumulated
  5205 has changed.  This enables to override declarations from fragments due
  5206 to interpretations -- for example, unwanted simp rules.
  5207 
  5208 * Isar/locales: interpretation in theories and proof contexts has been
  5209 extended.  One may now specify (and prove) equations, which are
  5210 unfolded in interpreted theorems.  This is useful for replacing
  5211 defined concepts (constants depending on locale parameters) by
  5212 concepts already existing in the target context.  Example:
  5213 
  5214   interpretation partial_order ["op <= :: [int, int] => bool"]
  5215     where "partial_order.less (op <=) (x::int) y = (x < y)"
  5216 
  5217 Typically, the constant `partial_order.less' is created by a
  5218 definition specification element in the context of locale
  5219 partial_order.
  5220 
  5221 * Method "induct": improved internal context management to support
  5222 local fixes and defines on-the-fly. Thus explicit meta-level
  5223 connectives !!  and ==> are rarely required anymore in inductive goals
  5224 (using object-logic connectives for this purpose has been long
  5225 obsolete anyway). Common proof patterns are explained in
  5226 src/HOL/Induct/Common_Patterns.thy, see also
  5227 src/HOL/Isar_examples/Puzzle.thy and src/HOL/Lambda for realistic
  5228 examples.
  5229 
  5230 * Method "induct": improved handling of simultaneous goals. Instead of
  5231 introducing object-level conjunction, the statement is now split into
  5232 several conclusions, while the corresponding symbolic cases are nested
  5233 accordingly. INCOMPATIBILITY, proofs need to be structured explicitly,
  5234 see src/HOL/Induct/Common_Patterns.thy, for example.
  5235 
  5236 * Method "induct": mutual induction rules are now specified as a list
  5237 of rule sharing the same induction cases. HOL packages usually provide
  5238 foo_bar.inducts for mutually defined items foo and bar (e.g. inductive
  5239 predicates/sets or datatypes). INCOMPATIBILITY, users need to specify
  5240 mutual induction rules differently, i.e. like this:
  5241 
  5242   (induct rule: foo_bar.inducts)
  5243   (induct set: foo bar)
  5244   (induct pred: foo bar)
  5245   (induct type: foo bar)
  5246 
  5247 The ML function ProjectRule.projections turns old-style rules into the
  5248 new format.
  5249 
  5250 * Method "coinduct": dual of induction, see
  5251 src/HOL/Library/Coinductive_List.thy for various examples.
  5252 
  5253 * Method "cases", "induct", "coinduct": the ``(open)'' option is
  5254 considered a legacy feature.
  5255 
  5256 * Attribute "symmetric" produces result with standardized schematic
  5257 variables (index 0).  Potential INCOMPATIBILITY.
  5258 
  5259 * Simplifier: by default the simplifier trace only shows top level
  5260 rewrites now. That is, trace_simp_depth_limit is set to 1 by
  5261 default. Thus there is less danger of being flooded by the trace. The
  5262 trace indicates where parts have been suppressed.
  5263   
  5264 * Provers/classical: removed obsolete classical version of elim_format
  5265 attribute; classical elim/dest rules are now treated uniformly when
  5266 manipulating the claset.
  5267 
  5268 * Provers/classical: stricter checks to ensure that supplied intro,
  5269 dest and elim rules are well-formed; dest and elim rules must have at
  5270 least one premise.
  5271 
  5272 * Provers/classical: attributes dest/elim/intro take an optional
  5273 weight argument for the rule (just as the Pure versions).  Weights are
  5274 ignored by automated tools, but determine the search order of single
  5275 rule steps.
  5276 
  5277 * Syntax: input syntax now supports dummy variable binding "%_. b",
  5278 where the body does not mention the bound variable.  Note that dummy
  5279 patterns implicitly depend on their context of bounds, which makes
  5280 "{_. _}" match any set comprehension as expected.  Potential
  5281 INCOMPATIBILITY -- parse translations need to cope with syntactic
  5282 constant "_idtdummy" in the binding position.
  5283 
  5284 * Syntax: removed obsolete syntactic constant "_K" and its associated
  5285 parse translation.  INCOMPATIBILITY -- use dummy abstraction instead,
  5286 for example "A -> B" => "Pi A (%_. B)".
  5287 
  5288 * Pure: 'class_deps' command visualizes the subclass relation, using
  5289 the graph browser tool.
  5290 
  5291 * Pure: 'print_theory' now suppresses certain internal declarations by
  5292 default; use '!' option for full details.
  5293 
  5294 
  5295 *** HOL ***
  5296 
  5297 * Method "metis" proves goals by applying the Metis general-purpose
  5298 resolution prover (see also http://gilith.com/software/metis/).
  5299 Examples are in the directory MetisExamples.  WARNING: the
  5300 Isabelle/HOL-Metis integration does not yet work properly with
  5301 multi-threading.
  5302   
  5303 * Command 'sledgehammer' invokes external automatic theorem provers as
  5304 background processes.  It generates calls to the "metis" method if
  5305 successful. These can be pasted into the proof.  Users do not have to
  5306 wait for the automatic provers to return.  WARNING: does not really
  5307 work with multi-threading.
  5308 
  5309 * New "auto_quickcheck" feature tests outermost goal statements for
  5310 potential counter-examples.  Controlled by ML references
  5311 auto_quickcheck (default true) and auto_quickcheck_time_limit (default
  5312 5000 milliseconds).  Fails silently if statements is outside of
  5313 executable fragment, or any other codgenerator problem occurs.
  5314 
  5315 * New constant "undefined" with axiom "undefined x = undefined".
  5316 
  5317 * Added class "HOL.eq", allowing for code generation with polymorphic
  5318 equality.
  5319 
  5320 * Some renaming of class constants due to canonical name prefixing in
  5321 the new 'class' package:
  5322 
  5323     HOL.abs ~> HOL.abs_class.abs
  5324     HOL.divide ~> HOL.divide_class.divide
  5325     0 ~> HOL.zero_class.zero
  5326     1 ~> HOL.one_class.one
  5327     op + ~> HOL.plus_class.plus
  5328     op - ~> HOL.minus_class.minus
  5329     uminus ~> HOL.minus_class.uminus
  5330     op * ~> HOL.times_class.times
  5331     op < ~> HOL.ord_class.less
  5332     op <= > HOL.ord_class.less_eq
  5333     Nat.power ~> Power.power_class.power
  5334     Nat.size ~> Nat.size_class.size
  5335     Numeral.number_of ~> Numeral.number_class.number_of
  5336     FixedPoint.Inf ~> Lattices.complete_lattice_class.Inf
  5337     FixedPoint.Sup ~> Lattices.complete_lattice_class.Sup
  5338     Orderings.min ~> Orderings.ord_class.min
  5339     Orderings.max ~> Orderings.ord_class.max
  5340     Divides.op div ~> Divides.div_class.div
  5341     Divides.op mod ~> Divides.div_class.mod
  5342     Divides.op dvd ~> Divides.div_class.dvd
  5343 
  5344 INCOMPATIBILITY.  Adaptions may be required in the following cases:
  5345 
  5346 a) User-defined constants using any of the names "plus", "minus",
  5347 "times", "less" or "less_eq". The standard syntax translations for
  5348 "+", "-" and "*" may go wrong.  INCOMPATIBILITY: use more specific
  5349 names.
  5350 
  5351 b) Variables named "plus", "minus", "times", "less", "less_eq"
  5352 INCOMPATIBILITY: use more specific names.
  5353 
  5354 c) Permutative equations (e.g. "a + b = b + a")
  5355 Since the change of names also changes the order of terms, permutative
  5356 rewrite rules may get applied in a different order. Experience shows
  5357 that this is rarely the case (only two adaptions in the whole Isabelle
  5358 distribution).  INCOMPATIBILITY: rewrite proofs
  5359 
  5360 d) ML code directly refering to constant names
  5361 This in general only affects hand-written proof tactics, simprocs and
  5362 so on.  INCOMPATIBILITY: grep your sourcecode and replace names.
  5363 Consider using @{const_name} antiquotation.
  5364 
  5365 * New class "default" with associated constant "default".
  5366 
  5367 * Function "sgn" is now overloaded and available on int, real, complex
  5368 (and other numeric types), using class "sgn".  Two possible defs of
  5369 sgn are given as equational assumptions in the classes sgn_if and
  5370 sgn_div_norm; ordered_idom now also inherits from sgn_if.
  5371 INCOMPATIBILITY.
  5372 
  5373 * Locale "partial_order" now unified with class "order" (cf. theory
  5374 Orderings), added parameter "less".  INCOMPATIBILITY.
  5375 
  5376 * Renamings in classes "order" and "linorder": facts "refl", "trans" and
  5377 "cases" to "order_refl", "order_trans" and "linorder_cases", to avoid
  5378 clashes with HOL "refl" and "trans".  INCOMPATIBILITY.
  5379 
  5380 * Classes "order" and "linorder": potential INCOMPATIBILITY due to
  5381 changed order of proof goals in instance proofs.
  5382 
  5383 * The transitivity reasoner for partial and linear orders is set up
  5384 for classes "order" and "linorder".  Instances of the reasoner are available
  5385 in all contexts importing or interpreting the corresponding locales.
  5386 Method "order" invokes the reasoner separately; the reasoner
  5387 is also integrated with the Simplifier as a solver.  Diagnostic
  5388 command 'print_orders' shows the available instances of the reasoner
  5389 in the current context.
  5390 
  5391 * Localized monotonicity predicate in theory "Orderings"; integrated
  5392 lemmas max_of_mono and min_of_mono with this predicate.
  5393 INCOMPATIBILITY.
  5394 
  5395 * Formulation of theorem "dense" changed slightly due to integration
  5396 with new class dense_linear_order.
  5397 
  5398 * Uniform lattice theory development in HOL.
  5399 
  5400     constants "meet" and "join" now named "inf" and "sup"
  5401     constant "Meet" now named "Inf"
  5402 
  5403     classes "meet_semilorder" and "join_semilorder" now named
  5404       "lower_semilattice" and "upper_semilattice"
  5405     class "lorder" now named "lattice"
  5406     class "comp_lat" now named "complete_lattice"
  5407 
  5408     Instantiation of lattice classes allows explicit definitions
  5409     for "inf" and "sup" operations (or "Inf" and "Sup" for complete lattices).
  5410 
  5411   INCOMPATIBILITY.  Theorem renames:
  5412 
  5413     meet_left_le            ~> inf_le1
  5414     meet_right_le           ~> inf_le2
  5415     join_left_le            ~> sup_ge1
  5416     join_right_le           ~> sup_ge2
  5417     meet_join_le            ~> inf_sup_ord
  5418     le_meetI                ~> le_infI
  5419     join_leI                ~> le_supI
  5420     le_meet                 ~> le_inf_iff
  5421     le_join                 ~> ge_sup_conv
  5422     meet_idempotent         ~> inf_idem
  5423     join_idempotent         ~> sup_idem
  5424     meet_comm               ~> inf_commute
  5425     join_comm               ~> sup_commute
  5426     meet_leI1               ~> le_infI1
  5427     meet_leI2               ~> le_infI2
  5428     le_joinI1               ~> le_supI1
  5429     le_joinI2               ~> le_supI2
  5430     meet_assoc              ~> inf_assoc
  5431     join_assoc              ~> sup_assoc
  5432     meet_left_comm          ~> inf_left_commute
  5433     meet_left_idempotent    ~> inf_left_idem
  5434     join_left_comm          ~> sup_left_commute
  5435     join_left_idempotent    ~> sup_left_idem
  5436     meet_aci                ~> inf_aci
  5437     join_aci                ~> sup_aci
  5438     le_def_meet             ~> le_iff_inf
  5439     le_def_join             ~> le_iff_sup
  5440     join_absorp2            ~> sup_absorb2
  5441     join_absorp1            ~> sup_absorb1
  5442     meet_absorp1            ~> inf_absorb1
  5443     meet_absorp2            ~> inf_absorb2
  5444     meet_join_absorp        ~> inf_sup_absorb
  5445     join_meet_absorp        ~> sup_inf_absorb
  5446     distrib_join_le         ~> distrib_sup_le
  5447     distrib_meet_le         ~> distrib_inf_le
  5448 
  5449     add_meet_distrib_left   ~> add_inf_distrib_left
  5450     add_join_distrib_left   ~> add_sup_distrib_left
  5451     is_join_neg_meet        ~> is_join_neg_inf
  5452     is_meet_neg_join        ~> is_meet_neg_sup
  5453     add_meet_distrib_right  ~> add_inf_distrib_right
  5454     add_join_distrib_right  ~> add_sup_distrib_right
  5455     add_meet_join_distribs  ~> add_sup_inf_distribs
  5456     join_eq_neg_meet        ~> sup_eq_neg_inf
  5457     meet_eq_neg_join        ~> inf_eq_neg_sup
  5458     add_eq_meet_join        ~> add_eq_inf_sup
  5459     meet_0_imp_0            ~> inf_0_imp_0
  5460     join_0_imp_0            ~> sup_0_imp_0
  5461     meet_0_eq_0             ~> inf_0_eq_0
  5462     join_0_eq_0             ~> sup_0_eq_0
  5463     neg_meet_eq_join        ~> neg_inf_eq_sup
  5464     neg_join_eq_meet        ~> neg_sup_eq_inf
  5465     join_eq_if              ~> sup_eq_if
  5466 
  5467     mono_meet               ~> mono_inf
  5468     mono_join               ~> mono_sup
  5469     meet_bool_eq            ~> inf_bool_eq
  5470     join_bool_eq            ~> sup_bool_eq
  5471     meet_fun_eq             ~> inf_fun_eq
  5472     join_fun_eq             ~> sup_fun_eq
  5473     meet_set_eq             ~> inf_set_eq
  5474     join_set_eq             ~> sup_set_eq
  5475     meet1_iff               ~> inf1_iff
  5476     meet2_iff               ~> inf2_iff
  5477     meet1I                  ~> inf1I
  5478     meet2I                  ~> inf2I
  5479     meet1D1                 ~> inf1D1
  5480     meet2D1                 ~> inf2D1
  5481     meet1D2                 ~> inf1D2
  5482     meet2D2                 ~> inf2D2
  5483     meet1E                  ~> inf1E
  5484     meet2E                  ~> inf2E
  5485     join1_iff               ~> sup1_iff
  5486     join2_iff               ~> sup2_iff
  5487     join1I1                 ~> sup1I1
  5488     join2I1                 ~> sup2I1
  5489     join1I1                 ~> sup1I1
  5490     join2I2                 ~> sup1I2
  5491     join1CI                 ~> sup1CI
  5492     join2CI                 ~> sup2CI
  5493     join1E                  ~> sup1E
  5494     join2E                  ~> sup2E
  5495 
  5496     is_meet_Meet            ~> is_meet_Inf
  5497     Meet_bool_def           ~> Inf_bool_def
  5498     Meet_fun_def            ~> Inf_fun_def
  5499     Meet_greatest           ~> Inf_greatest
  5500     Meet_lower              ~> Inf_lower
  5501     Meet_set_def            ~> Inf_set_def
  5502 
  5503     Sup_def                 ~> Sup_Inf
  5504     Sup_bool_eq             ~> Sup_bool_def
  5505     Sup_fun_eq              ~> Sup_fun_def
  5506     Sup_set_eq              ~> Sup_set_def
  5507 
  5508     listsp_meetI            ~> listsp_infI
  5509     listsp_meet_eq          ~> listsp_inf_eq
  5510 
  5511     meet_min                ~> inf_min
  5512     join_max                ~> sup_max
  5513 
  5514 * Added syntactic class "size"; overloaded constant "size" now has
  5515 type "'a::size ==> bool"
  5516 
  5517 * Internal reorganisation of `size' of datatypes: size theorems
  5518 "foo.size" are no longer subsumed by "foo.simps" (but are still
  5519 simplification rules by default!); theorems "prod.size" now named
  5520 "*.size".
  5521 
  5522 * Class "div" now inherits from class "times" rather than "type".
  5523 INCOMPATIBILITY.
  5524 
  5525 * HOL/Finite_Set: "name-space" locales Lattice, Distrib_lattice,
  5526 Linorder etc.  have disappeared; operations defined in terms of
  5527 fold_set now are named Inf_fin, Sup_fin.  INCOMPATIBILITY.
  5528 
  5529 * HOL/Nat: neq0_conv no longer declared as iff.  INCOMPATIBILITY.
  5530 
  5531 * HOL-Word: New extensive library and type for generic, fixed size
  5532 machine words, with arithmetic, bit-wise, shifting and rotating
  5533 operations, reflection into int, nat, and bool lists, automation for
  5534 linear arithmetic (by automatic reflection into nat or int), including
  5535 lemmas on overflow and monotonicity.  Instantiated to all appropriate
  5536 arithmetic type classes, supporting automatic simplification of
  5537 numerals on all operations.
  5538 
  5539 * Library/Boolean_Algebra: locales for abstract boolean algebras.
  5540 
  5541 * Library/Numeral_Type: numbers as types, e.g. TYPE(32).
  5542 
  5543 * Code generator library theories:
  5544   - Code_Integer represents HOL integers by big integer literals in target
  5545     languages.
  5546   - Code_Char represents HOL characters by character literals in target
  5547     languages.
  5548   - Code_Char_chr like Code_Char, but also offers treatment of character
  5549     codes; includes Code_Integer.
  5550   - Executable_Set allows to generate code for finite sets using lists.
  5551   - Executable_Rat implements rational numbers as triples (sign, enumerator,
  5552     denominator).
  5553   - Executable_Real implements a subset of real numbers, namly those
  5554     representable by rational numbers.
  5555   - Efficient_Nat implements natural numbers by integers, which in general will
  5556     result in higher efficency; pattern matching with 0/Suc is eliminated;
  5557     includes Code_Integer.
  5558   - Code_Index provides an additional datatype index which is mapped to
  5559     target-language built-in integers.
  5560   - Code_Message provides an additional datatype message_string which is isomorphic to
  5561     strings; messages are mapped to target-language strings.
  5562 
  5563 * New package for inductive predicates
  5564 
  5565   An n-ary predicate p with m parameters z_1, ..., z_m can now be defined via
  5566 
  5567     inductive
  5568       p :: "U_1 => ... => U_m => T_1 => ... => T_n => bool"
  5569       for z_1 :: U_1 and ... and z_n :: U_m
  5570     where
  5571       rule_1: "... ==> p z_1 ... z_m t_1_1 ... t_1_n"
  5572     | ...
  5573 
  5574   with full support for type-inference, rather than
  5575 
  5576     consts s :: "U_1 => ... => U_m => (T_1 * ... * T_n) set"
  5577 
  5578     abbreviation p :: "U_1 => ... => U_m => T_1 => ... => T_n => bool"
  5579     where "p z_1 ... z_m x_1 ... x_n == (x_1, ..., x_n) : s z_1 ... z_m"
  5580 
  5581     inductive "s z_1 ... z_m"
  5582     intros
  5583       rule_1: "... ==> (t_1_1, ..., t_1_n) : s z_1 ... z_m"
  5584       ...
  5585 
  5586   For backward compatibility, there is a wrapper allowing inductive
  5587   sets to be defined with the new package via
  5588 
  5589     inductive_set
  5590       s :: "U_1 => ... => U_m => (T_1 * ... * T_n) set"
  5591       for z_1 :: U_1 and ... and z_n :: U_m
  5592     where
  5593       rule_1: "... ==> (t_1_1, ..., t_1_n) : s z_1 ... z_m"
  5594     | ...
  5595 
  5596   or
  5597 
  5598     inductive_set
  5599       s :: "U_1 => ... => U_m => (T_1 * ... * T_n) set"
  5600       and p :: "U_1 => ... => U_m => T_1 => ... => T_n => bool"
  5601       for z_1 :: U_1 and ... and z_n :: U_m
  5602     where
  5603       "p z_1 ... z_m x_1 ... x_n == (x_1, ..., x_n) : s z_1 ... z_m"
  5604     | rule_1: "... ==> p z_1 ... z_m t_1_1 ... t_1_n"
  5605     | ...
  5606 
  5607   if the additional syntax "p ..." is required.
  5608 
  5609   Numerous examples can be found in the subdirectories src/HOL/Auth,
  5610   src/HOL/Bali, src/HOL/Induct, and src/HOL/MicroJava.
  5611 
  5612   INCOMPATIBILITIES:
  5613 
  5614   - Since declaration and definition of inductive sets or predicates
  5615     is no longer separated, abbreviations involving the newly
  5616     introduced sets or predicates must be specified together with the
  5617     introduction rules after the 'where' keyword (see above), rather
  5618     than before the actual inductive definition.
  5619 
  5620   - The variables in induction and elimination rules are now
  5621     quantified in the order of their occurrence in the introduction
  5622     rules, rather than in alphabetical order. Since this may break
  5623     some proofs, these proofs either have to be repaired, e.g. by
  5624     reordering the variables a_i_1 ... a_i_{k_i} in Isar 'case'
  5625     statements of the form
  5626 
  5627       case (rule_i a_i_1 ... a_i_{k_i})
  5628 
  5629     or the old order of quantification has to be restored by explicitly adding
  5630     meta-level quantifiers in the introduction rules, i.e.
  5631 
  5632       | rule_i: "!!a_i_1 ... a_i_{k_i}. ... ==> p z_1 ... z_m t_i_1 ... t_i_n"
  5633 
  5634   - The format of the elimination rules is now
  5635 
  5636       p z_1 ... z_m x_1 ... x_n ==>
  5637         (!!a_1_1 ... a_1_{k_1}. x_1 = t_1_1 ==> ... ==> x_n = t_1_n ==> ... ==> P)
  5638         ==> ... ==> P
  5639 
  5640     for predicates and
  5641 
  5642       (x_1, ..., x_n) : s z_1 ... z_m ==>
  5643         (!!a_1_1 ... a_1_{k_1}. x_1 = t_1_1 ==> ... ==> x_n = t_1_n ==> ... ==> P)
  5644         ==> ... ==> P
  5645 
  5646     for sets rather than
  5647 
  5648       x : s z_1 ... z_m ==>
  5649         (!!a_1_1 ... a_1_{k_1}. x = (t_1_1, ..., t_1_n) ==> ... ==> P)
  5650         ==> ... ==> P
  5651 
  5652     This may require terms in goals to be expanded to n-tuples
  5653     (e.g. using case_tac or simplification with the split_paired_all
  5654     rule) before the above elimination rule is applicable.
  5655 
  5656   - The elimination or case analysis rules for (mutually) inductive
  5657     sets or predicates are now called "p_1.cases" ... "p_k.cases". The
  5658     list of rules "p_1_..._p_k.elims" is no longer available.
  5659 
  5660 * New package "function"/"fun" for general recursive functions,
  5661 supporting mutual and nested recursion, definitions in local contexts,
  5662 more general pattern matching and partiality. See HOL/ex/Fundefs.thy
  5663 for small examples, and the separate tutorial on the function
  5664 package. The old recdef "package" is still available as before, but
  5665 users are encouraged to use the new package.
  5666 
  5667 * Method "lexicographic_order" automatically synthesizes termination
  5668 relations as lexicographic combinations of size measures. 
  5669 
  5670 * Case-expressions allow arbitrary constructor-patterns (including
  5671 "_") and take their order into account, like in functional
  5672 programming.  Internally, this is translated into nested
  5673 case-expressions; missing cases are added and mapped to the predefined
  5674 constant "undefined". In complicated cases printing may no longer show
  5675 the original input but the internal form. Lambda-abstractions allow
  5676 the same form of pattern matching: "% pat1 => e1 | ..." is an
  5677 abbreviation for "%x. case x of pat1 => e1 | ..." where x is a new
  5678 variable.
  5679 
  5680 * IntDef: The constant "int :: nat => int" has been removed; now "int"
  5681 is an abbreviation for "of_nat :: nat => int". The simplification
  5682 rules for "of_nat" have been changed to work like "int" did
  5683 previously.  Potential INCOMPATIBILITY:
  5684   - "of_nat (Suc m)" simplifies to "1 + of_nat m" instead of "of_nat m + 1"
  5685   - of_nat_diff and of_nat_mult are no longer default simp rules
  5686 
  5687 * Method "algebra" solves polynomial equations over (semi)rings using
  5688 Groebner bases. The (semi)ring structure is defined by locales and the
  5689 tool setup depends on that generic context. Installing the method for
  5690 a specific type involves instantiating the locale and possibly adding
  5691 declarations for computation on the coefficients.  The method is
  5692 already instantiated for natural numbers and for the axiomatic class
  5693 of idoms with numerals.  See also the paper by Chaieb and Wenzel at
  5694 CALCULEMUS 2007 for the general principles underlying this
  5695 architecture of context-aware proof-tools.
  5696 
  5697 * Method "ferrack" implements quantifier elimination over
  5698 special-purpose dense linear orders using locales (analogous to
  5699 "algebra"). The method is already installed for class
  5700 {ordered_field,recpower,number_ring} which subsumes real, hyperreal,
  5701 rat, etc.
  5702 
  5703 * Former constant "List.op @" now named "List.append".  Use ML
  5704 antiquotations @{const_name List.append} or @{term " ... @ ... "} to
  5705 circumvent possible incompatibilities when working on ML level.
  5706 
  5707 * primrec: missing cases mapped to "undefined" instead of "arbitrary".
  5708 
  5709 * New function listsum :: 'a list => 'a for arbitrary monoids.
  5710 Special syntax: "SUM x <- xs. f x" (and latex variants)
  5711 
  5712 * New syntax for Haskell-like list comprehension (input only), eg.
  5713 [(x,y). x <- xs, y <- ys, x ~= y], see also src/HOL/List.thy.
  5714 
  5715 * The special syntax for function "filter" has changed from [x :
  5716 xs. P] to [x <- xs. P] to avoid an ambiguity caused by list
  5717 comprehension syntax, and for uniformity.  INCOMPATIBILITY.
  5718 
  5719 * [a..b] is now defined for arbitrary linear orders.  It used to be
  5720 defined on nat only, as an abbreviation for [a..<Suc b]
  5721 INCOMPATIBILITY.
  5722 
  5723 * Renamed lemma "set_take_whileD"  to "set_takeWhileD".
  5724 
  5725 * New functions "sorted" and "sort" in src/HOL/List.thy.
  5726 
  5727 * New lemma collection field_simps (an extension of ring_simps) for
  5728 manipulating (in)equations involving division. Multiplies with all
  5729 denominators that can be proved to be non-zero (in equations) or
  5730 positive/negative (in inequations).
  5731 
  5732 * Lemma collections ring_eq_simps, group_eq_simps and ring_distrib
  5733 have been improved and renamed to ring_simps, group_simps and
  5734 ring_distribs.  Removed lemmas field_xyz in theory Ring_and_Field
  5735 because they were subsumed by lemmas xyz.  INCOMPATIBILITY.
  5736 
  5737 * Theory Library/Commutative_Ring: switched from recdef to function
  5738 package; constants add, mul, pow now curried.  Infix syntax for
  5739 algebraic operations.
  5740 
  5741 * Dropped redundant lemma def_imp_eq in favor of meta_eq_to_obj_eq.
  5742 INCOMPATIBILITY.
  5743 
  5744 * Dropped redundant lemma if_def2 in favor of if_bool_eq_conj.
  5745 INCOMPATIBILITY.
  5746 
  5747 * HOL/records: generalised field-update to take a function on the
  5748 field rather than the new value: r(|A := x|) is translated to A_update
  5749 (K x) r The K-combinator that is internally used is called K_record.
  5750 INCOMPATIBILITY: Usage of the plain update functions has to be
  5751 adapted.
  5752  
  5753 * Class "semiring_0" now contains annihilation axioms x * 0 = 0 and 0
  5754 * x = 0, which are required for a semiring.  Richer structures do not
  5755 inherit from semiring_0 anymore, because this property is a theorem
  5756 there, not an axiom.  INCOMPATIBILITY: In instances of semiring_0,
  5757 there is more to prove, but this is mostly trivial.
  5758 
  5759 * Class "recpower" is generalized to arbitrary monoids, not just
  5760 commutative semirings.  INCOMPATIBILITY: may need to incorporate
  5761 commutativity or semiring properties additionally.
  5762 
  5763 * Constant "List.list_all2" in List.thy now uses authentic syntax.
  5764 INCOMPATIBILITY: translations containing list_all2 may go wrong,
  5765 better use 'abbreviation'.
  5766 
  5767 * Renamed constant "List.op mem" to "List.member".  INCOMPATIBILITY.
  5768 
  5769 * Numeral syntax: type 'bin' which was a mere type copy of 'int' has
  5770 been abandoned in favour of plain 'int'.  INCOMPATIBILITY --
  5771 significant changes for setting up numeral syntax for types:
  5772   - New constants Numeral.pred and Numeral.succ instead
  5773       of former Numeral.bin_pred and Numeral.bin_succ.
  5774   - Use integer operations instead of bin_add, bin_mult and so on.
  5775   - Numeral simplification theorems named Numeral.numeral_simps instead of Bin_simps.
  5776   - ML structure Bin_Simprocs now named Int_Numeral_Base_Simprocs.
  5777 
  5778 See src/HOL/Integ/IntArith.thy for an example setup.
  5779 
  5780 * Command 'normal_form' computes the normal form of a term that may
  5781 contain free variables.  For example ``normal_form "rev [a, b, c]"''
  5782 produces ``[b, c, a]'' (without proof).  This command is suitable for
  5783 heavy-duty computations because the functions are compiled to ML
  5784 first.  Correspondingly, a method "normalization" is provided.  See
  5785 further src/HOL/ex/NormalForm.thy and src/Tools/nbe.ML.
  5786 
  5787 * Alternative iff syntax "A <-> B" for equality on bool (with priority
  5788 25 like -->); output depends on the "iff" print_mode, the default is
  5789 "A = B" (with priority 50).
  5790 
  5791 * Relations less (<) and less_eq (<=) are also available on type bool.
  5792 Modified syntax to disallow nesting without explicit parentheses,
  5793 e.g. "(x < y) < z" or "x < (y < z)", but NOT "x < y < z".  Potential
  5794 INCOMPATIBILITY.
  5795 
  5796 * "LEAST x:A. P" expands to "LEAST x. x:A & P" (input only).
  5797 
  5798 * Relation composition operator "op O" now has precedence 75 and binds
  5799 stronger than union and intersection. INCOMPATIBILITY.
  5800 
  5801 * The old set interval syntax "{m..n(}" (and relatives) has been
  5802 removed.  Use "{m..<n}" (and relatives) instead.
  5803 
  5804 * In the context of the assumption "~(s = t)" the Simplifier rewrites
  5805 "t = s" to False (by simproc "neq").  INCOMPATIBILITY, consider using
  5806 ``declare [[simproc del: neq]]''.
  5807 
  5808 * Simplifier: "m dvd n" where m and n are numbers is evaluated to
  5809 True/False.
  5810 
  5811 * Theorem Cons_eq_map_conv no longer declared as "simp".
  5812 
  5813 * Theorem setsum_mult renamed to setsum_right_distrib.
  5814 
  5815 * Prefer ex1I over ex_ex1I in single-step reasoning, e.g. by the
  5816 ``rule'' method.
  5817 
  5818 * Reimplemented methods "sat" and "satx", with several improvements:
  5819 goals no longer need to be stated as "<prems> ==> False", equivalences
  5820 (i.e. "=" on type bool) are handled, variable names of the form
  5821 "lit_<n>" are no longer reserved, significant speedup.
  5822 
  5823 * Methods "sat" and "satx" can now replay MiniSat proof traces.
  5824 zChaff is still supported as well.
  5825 
  5826 * 'inductive' and 'datatype': provide projections of mutual rules,
  5827 bundled as foo_bar.inducts;
  5828 
  5829 * Library: moved theories Parity, GCD, Binomial, Infinite_Set to
  5830 Library.
  5831 
  5832 * Library: moved theory Accessible_Part to main HOL.
  5833 
  5834 * Library: added theory Coinductive_List of potentially infinite lists
  5835 as greatest fixed-point.
  5836 
  5837 * Library: added theory AssocList which implements (finite) maps as
  5838 association lists.
  5839 
  5840 * Method "evaluation" solves goals (i.e. a boolean expression)
  5841 efficiently by compiling it to ML.  The goal is "proved" (via an
  5842 oracle) if it evaluates to True.
  5843 
  5844 * Linear arithmetic now splits certain operators (e.g. min, max, abs)
  5845 also when invoked by the simplifier.  This results in the Simplifier
  5846 being more powerful on arithmetic goals.  INCOMPATIBILITY.
  5847 Configuration option fast_arith_split_limit=0 recovers the old
  5848 behavior.
  5849 
  5850 * Support for hex (0x20) and binary (0b1001) numerals.
  5851 
  5852 * New method: reify eqs (t), where eqs are equations for an
  5853 interpretation I :: 'a list => 'b => 'c and t::'c is an optional
  5854 parameter, computes a term s::'b and a list xs::'a list and proves the
  5855 theorem I xs s = t. This is also known as reification or quoting. The
  5856 resulting theorem is applied to the subgoal to substitute t with I xs
  5857 s.  If t is omitted, the subgoal itself is reified.
  5858 
  5859 * New method: reflection corr_thm eqs (t). The parameters eqs and (t)
  5860 are as explained above. corr_thm is a theorem for I vs (f t) = I vs t,
  5861 where f is supposed to be a computable function (in the sense of code
  5862 generattion). The method uses reify to compute s and xs as above then
  5863 applies corr_thm and uses normalization by evaluation to "prove" f s =
  5864 r and finally gets the theorem t = r, which is again applied to the
  5865 subgoal. An Example is available in src/HOL/ex/ReflectionEx.thy.
  5866 
  5867 * Reflection: Automatic reification now handels binding, an example is
  5868 available in src/HOL/ex/ReflectionEx.thy
  5869 
  5870 * HOL-Statespace: ``State Spaces: The Locale Way'' introduces a
  5871 command 'statespace' that is similar to 'record', but introduces an
  5872 abstract specification based on the locale infrastructure instead of
  5873 HOL types.  This leads to extra flexibility in composing state spaces,
  5874 in particular multiple inheritance and renaming of components.
  5875 
  5876 
  5877 *** HOL-Complex ***
  5878 
  5879 * Hyperreal: Functions root and sqrt are now defined on negative real
  5880 inputs so that root n (- x) = - root n x and sqrt (- x) = - sqrt x.
  5881 Nonnegativity side conditions have been removed from many lemmas, so
  5882 that more subgoals may now be solved by simplification; potential
  5883 INCOMPATIBILITY.
  5884 
  5885 * Real: new type classes formalize real normed vector spaces and
  5886 algebras, using new overloaded constants scaleR :: real => 'a => 'a
  5887 and norm :: 'a => real.
  5888 
  5889 * Real: constant of_real :: real => 'a::real_algebra_1 injects from
  5890 reals into other types. The overloaded constant Reals :: 'a set is now
  5891 defined as range of_real; potential INCOMPATIBILITY.
  5892 
  5893 * Real: proper support for ML code generation, including 'quickcheck'.
  5894 Reals are implemented as arbitrary precision rationals.
  5895 
  5896 * Hyperreal: Several constants that previously worked only for the
  5897 reals have been generalized, so they now work over arbitrary vector
  5898 spaces. Type annotations may need to be added in some cases; potential
  5899 INCOMPATIBILITY.
  5900 
  5901   Infinitesimal  :: ('a::real_normed_vector) star set
  5902   HFinite        :: ('a::real_normed_vector) star set
  5903   HInfinite      :: ('a::real_normed_vector) star set
  5904   approx         :: ('a::real_normed_vector) star => 'a star => bool
  5905   monad          :: ('a::real_normed_vector) star => 'a star set
  5906   galaxy         :: ('a::real_normed_vector) star => 'a star set
  5907   (NS)LIMSEQ     :: [nat => 'a::real_normed_vector, 'a] => bool
  5908   (NS)convergent :: (nat => 'a::real_normed_vector) => bool
  5909   (NS)Bseq       :: (nat => 'a::real_normed_vector) => bool
  5910   (NS)Cauchy     :: (nat => 'a::real_normed_vector) => bool
  5911   (NS)LIM        :: ['a::real_normed_vector => 'b::real_normed_vector, 'a, 'b] => bool
  5912   is(NS)Cont     :: ['a::real_normed_vector => 'b::real_normed_vector, 'a] => bool
  5913   deriv          :: ['a::real_normed_field => 'a, 'a, 'a] => bool
  5914   sgn            :: 'a::real_normed_vector => 'a
  5915   exp            :: 'a::{recpower,real_normed_field,banach} => 'a
  5916 
  5917 * Complex: Some complex-specific constants are now abbreviations for
  5918 overloaded ones: complex_of_real = of_real, cmod = norm, hcmod =
  5919 hnorm.  Other constants have been entirely removed in favor of the
  5920 polymorphic versions (INCOMPATIBILITY):
  5921 
  5922   approx        <-- capprox
  5923   HFinite       <-- CFinite
  5924   HInfinite     <-- CInfinite
  5925   Infinitesimal <-- CInfinitesimal
  5926   monad         <-- cmonad
  5927   galaxy        <-- cgalaxy
  5928   (NS)LIM       <-- (NS)CLIM, (NS)CRLIM
  5929   is(NS)Cont    <-- is(NS)Contc, is(NS)contCR
  5930   (ns)deriv     <-- (ns)cderiv
  5931 
  5932 
  5933 *** HOL-Algebra ***
  5934 
  5935 * Formalisation of ideals and the quotient construction over rings.
  5936 
  5937 * Order and lattice theory no longer based on records.
  5938 INCOMPATIBILITY.
  5939 
  5940 * Renamed lemmas least_carrier -> least_closed and greatest_carrier ->
  5941 greatest_closed.  INCOMPATIBILITY.
  5942 
  5943 * Method algebra is now set up via an attribute.  For examples see
  5944 Ring.thy.  INCOMPATIBILITY: the method is now weaker on combinations
  5945 of algebraic structures.
  5946 
  5947 * Renamed theory CRing to Ring.
  5948 
  5949 
  5950 *** HOL-Nominal ***
  5951 
  5952 * Substantial, yet incomplete support for nominal datatypes (binding
  5953 structures) based on HOL-Nominal logic.  See src/HOL/Nominal and
  5954 src/HOL/Nominal/Examples.  Prospective users should consult
  5955 http://isabelle.in.tum.de/nominal/
  5956 
  5957 
  5958 *** ML ***
  5959 
  5960 * ML basics: just one true type int, which coincides with IntInf.int
  5961 (even on SML/NJ).
  5962 
  5963 * ML within Isar: antiquotations allow to embed statically-checked
  5964 formal entities in the source, referring to the context available at
  5965 compile-time.  For example:
  5966 
  5967 ML {* @{sort "{zero,one}"} *}
  5968 ML {* @{typ "'a => 'b"} *}
  5969 ML {* @{term "%x. x"} *}
  5970 ML {* @{prop "x == y"} *}
  5971 ML {* @{ctyp "'a => 'b"} *}
  5972 ML {* @{cterm "%x. x"} *}
  5973 ML {* @{cprop "x == y"} *}
  5974 ML {* @{thm asm_rl} *}
  5975 ML {* @{thms asm_rl} *}
  5976 ML {* @{type_name c} *}
  5977 ML {* @{type_syntax c} *}
  5978 ML {* @{const_name c} *}
  5979 ML {* @{const_syntax c} *}
  5980 ML {* @{context} *}
  5981 ML {* @{theory} *}
  5982 ML {* @{theory Pure} *}
  5983 ML {* @{theory_ref} *}
  5984 ML {* @{theory_ref Pure} *}
  5985 ML {* @{simpset} *}
  5986 ML {* @{claset} *}
  5987 ML {* @{clasimpset} *}
  5988 
  5989 The same works for sources being ``used'' within an Isar context.
  5990 
  5991 * ML in Isar: improved error reporting; extra verbosity with
  5992 ML_Context.trace enabled.
  5993 
  5994 * Pure/General/table.ML: the join operations now works via exceptions
  5995 DUP/SAME instead of type option. This is simpler in simple cases, and
  5996 admits slightly more efficient complex applications.
  5997 
  5998 * Pure: 'advanced' translation functions (parse_translation etc.) now
  5999 use Context.generic instead of just theory.
  6000 
  6001 * Pure: datatype Context.generic joins theory/Proof.context and
  6002 provides some facilities for code that works in either kind of
  6003 context, notably GenericDataFun for uniform theory and proof data.
  6004 
  6005 * Pure: simplified internal attribute type, which is now always
  6006 Context.generic * thm -> Context.generic * thm. Global (theory) vs.
  6007 local (Proof.context) attributes have been discontinued, while
  6008 minimizing code duplication. Thm.rule_attribute and
  6009 Thm.declaration_attribute build canonical attributes; see also structure
  6010 Context for further operations on Context.generic, notably
  6011 GenericDataFun. INCOMPATIBILITY, need to adapt attribute type
  6012 declarations and definitions.
  6013 
  6014 * Context data interfaces (Theory/Proof/GenericDataFun): removed
  6015 name/print, uninitialized data defaults to ad-hoc copy of empty value,
  6016 init only required for impure data. INCOMPATIBILITY: empty really need
  6017 to be empty (no dependencies on theory content!)
  6018 
  6019 * Pure/kernel: consts certification ignores sort constraints given in
  6020 signature declarations. (This information is not relevant to the
  6021 logic, but only for type inference.) SIGNIFICANT INTERNAL CHANGE,
  6022 potential INCOMPATIBILITY.
  6023 
  6024 * Pure: axiomatic type classes are now purely definitional, with
  6025 explicit proofs of class axioms and super class relations performed
  6026 internally. See Pure/axclass.ML for the main internal interfaces --
  6027 notably AxClass.define_class supercedes AxClass.add_axclass, and
  6028 AxClass.axiomatize_class/classrel/arity supersede
  6029 Sign.add_classes/classrel/arities.
  6030 
  6031 * Pure/Isar: Args/Attrib parsers operate on Context.generic --
  6032 global/local versions on theory vs. Proof.context have been
  6033 discontinued; Attrib.syntax and Method.syntax have been adapted
  6034 accordingly.  INCOMPATIBILITY, need to adapt parser expressions for
  6035 attributes, methods, etc.
  6036 
  6037 * Pure: several functions of signature "... -> theory -> theory * ..."
  6038 have been reoriented to "... -> theory -> ... * theory" in order to
  6039 allow natural usage in combination with the ||>, ||>>, |-> and
  6040 fold_map combinators.
  6041 
  6042 * Pure: official theorem names (closed derivations) and additional
  6043 comments (tags) are now strictly separate.  Name hints -- which are
  6044 maintained as tags -- may be attached any time without affecting the
  6045 derivation.
  6046 
  6047 * Pure: primitive rule lift_rule now takes goal cterm instead of an
  6048 actual goal state (thm).  Use Thm.lift_rule (Thm.cprem_of st i) to
  6049 achieve the old behaviour.
  6050 
  6051 * Pure: the "Goal" constant is now called "prop", supporting a
  6052 slightly more general idea of ``protecting'' meta-level rule
  6053 statements.
  6054 
  6055 * Pure: Logic.(un)varify only works in a global context, which is now
  6056 enforced instead of silently assumed.  INCOMPATIBILITY, may use
  6057 Logic.legacy_(un)varify as temporary workaround.
  6058 
  6059 * Pure: structure Name provides scalable operations for generating
  6060 internal variable names, notably Name.variants etc.  This replaces
  6061 some popular functions from term.ML:
  6062 
  6063   Term.variant		->  Name.variant
  6064   Term.variantlist	->  Name.variant_list
  6065   Term.invent_names	->  Name.invent_list
  6066 
  6067 Note that low-level renaming rarely occurs in new code -- operations
  6068 from structure Variable are used instead (see below).
  6069 
  6070 * Pure: structure Variable provides fundamental operations for proper
  6071 treatment of fixed/schematic variables in a context.  For example,
  6072 Variable.import introduces fixes for schematics of given facts and
  6073 Variable.export reverses the effect (up to renaming) -- this replaces
  6074 various freeze_thaw operations.
  6075 
  6076 * Pure: structure Goal provides simple interfaces for
  6077 init/conclude/finish and tactical prove operations (replacing former
  6078 Tactic.prove).  Goal.prove is the canonical way to prove results
  6079 within a given context; Goal.prove_global is a degraded version for
  6080 theory level goals, including a global Drule.standard.  Note that
  6081 OldGoals.prove_goalw_cterm has long been obsolete, since it is
  6082 ill-behaved in a local proof context (e.g. with local fixes/assumes or
  6083 in a locale context).
  6084 
  6085 * Pure/Syntax: generic interfaces for parsing (Syntax.parse_term etc.)
  6086 and type checking (Syntax.check_term etc.), with common combinations
  6087 (Syntax.read_term etc.). These supersede former Sign.read_term etc.
  6088 which are considered legacy and await removal.
  6089 
  6090 * Pure/Syntax: generic interfaces for type unchecking
  6091 (Syntax.uncheck_terms etc.) and unparsing (Syntax.unparse_term etc.),
  6092 with common combinations (Syntax.pretty_term, Syntax.string_of_term
  6093 etc.).  Former Sign.pretty_term, Sign.string_of_term etc. are still
  6094 available for convenience, but refer to the very same operations using
  6095 a mere theory instead of a full context.
  6096 
  6097 * Isar: simplified treatment of user-level errors, using exception
  6098 ERROR of string uniformly.  Function error now merely raises ERROR,
  6099 without any side effect on output channels.  The Isar toplevel takes
  6100 care of proper display of ERROR exceptions.  ML code may use plain
  6101 handle/can/try; cat_error may be used to concatenate errors like this:
  6102 
  6103   ... handle ERROR msg => cat_error msg "..."
  6104 
  6105 Toplevel ML code (run directly or through the Isar toplevel) may be
  6106 embedded into the Isar toplevel with exception display/debug like
  6107 this:
  6108 
  6109   Isar.toplevel (fn () => ...)
  6110 
  6111 INCOMPATIBILITY, removed special transform_error facilities, removed
  6112 obsolete variants of user-level exceptions (ERROR_MESSAGE,
  6113 Context.PROOF, ProofContext.CONTEXT, Proof.STATE, ProofHistory.FAIL)
  6114 -- use plain ERROR instead.
  6115 
  6116 * Isar: theory setup now has type (theory -> theory), instead of a
  6117 list.  INCOMPATIBILITY, may use #> to compose setup functions.
  6118 
  6119 * Isar: ML toplevel pretty printer for type Proof.context, subject to
  6120 ProofContext.debug/verbose flags.
  6121 
  6122 * Isar: Toplevel.theory_to_proof admits transactions that modify the
  6123 theory before entering a proof state.  Transactions now always see a
  6124 quasi-functional intermediate checkpoint, both in interactive and
  6125 batch mode.
  6126 
  6127 * Isar: simplified interfaces for outer syntax.  Renamed
  6128 OuterSyntax.add_keywords to OuterSyntax.keywords.  Removed
  6129 OuterSyntax.add_parsers -- this functionality is now included in
  6130 OuterSyntax.command etc.  INCOMPATIBILITY.
  6131 
  6132 * Simplifier: the simpset of a running simplification process now
  6133 contains a proof context (cf. Simplifier.the_context), which is the
  6134 very context that the initial simpset has been retrieved from (by
  6135 simpset_of/local_simpset_of).  Consequently, all plug-in components
  6136 (solver, looper etc.) may depend on arbitrary proof data.
  6137 
  6138 * Simplifier.inherit_context inherits the proof context (plus the
  6139 local bounds) of the current simplification process; any simproc
  6140 etc. that calls the Simplifier recursively should do this!  Removed
  6141 former Simplifier.inherit_bounds, which is already included here --
  6142 INCOMPATIBILITY.  Tools based on low-level rewriting may even have to
  6143 specify an explicit context using Simplifier.context/theory_context.
  6144 
  6145 * Simplifier/Classical Reasoner: more abstract interfaces
  6146 change_simpset/claset for modifying the simpset/claset reference of a
  6147 theory; raw versions simpset/claset_ref etc. have been discontinued --
  6148 INCOMPATIBILITY.
  6149 
  6150 * Provers: more generic wrt. syntax of object-logics, avoid hardwired
  6151 "Trueprop" etc.
  6152 
  6153 
  6154 *** System ***
  6155 
  6156 * settings: the default heap location within ISABELLE_HOME_USER now
  6157 includes ISABELLE_IDENTIFIER.  This simplifies use of multiple
  6158 Isabelle installations.
  6159 
  6160 * isabelle-process: option -S (secure mode) disables some critical
  6161 operations, notably runtime compilation and evaluation of ML source
  6162 code.
  6163 
  6164 * Basic Isabelle mode for jEdit, see Isabelle/lib/jedit/.
  6165 
  6166 * Support for parallel execution, using native multicore support of
  6167 Poly/ML 5.1.  The theory loader exploits parallelism when processing
  6168 independent theories, according to the given theory header
  6169 specifications. The maximum number of worker threads is specified via
  6170 usedir option -M or the "max-threads" setting in Proof General. A
  6171 speedup factor of 1.5--3.5 can be expected on a 4-core machine, and up
  6172 to 6 on a 8-core machine.  User-code needs to observe certain
  6173 guidelines for thread-safe programming, see appendix A in the Isar
  6174 Implementation manual.
  6175 
  6176 
  6177 
  6178 New in Isabelle2005 (October 2005)
  6179 ----------------------------------
  6180 
  6181 *** General ***
  6182 
  6183 * Theory headers: the new header syntax for Isar theories is
  6184 
  6185   theory <name>
  6186   imports <theory1> ... <theoryN>
  6187   uses <file1> ... <fileM>
  6188   begin
  6189 
  6190 where the 'uses' part is optional.  The previous syntax
  6191 
  6192   theory <name> = <theory1> + ... + <theoryN>:
  6193 
  6194 will disappear in the next release.  Use isatool fixheaders to convert
  6195 existing theory files.  Note that there is no change in ancient
  6196 non-Isar theories now, but these will disappear soon.
  6197 
  6198 * Theory loader: parent theories can now also be referred to via
  6199 relative and absolute paths.
  6200 
  6201 * Command 'find_theorems' searches for a list of criteria instead of a
  6202 list of constants. Known criteria are: intro, elim, dest, name:string,
  6203 simp:term, and any term. Criteria can be preceded by '-' to select
  6204 theorems that do not match. Intro, elim, dest select theorems that
  6205 match the current goal, name:s selects theorems whose fully qualified
  6206 name contain s, and simp:term selects all simplification rules whose
  6207 lhs match term.  Any other term is interpreted as pattern and selects
  6208 all theorems matching the pattern. Available in ProofGeneral under
  6209 'ProofGeneral -> Find Theorems' or C-c C-f.  Example:
  6210 
  6211   C-c C-f (100) "(_::nat) + _ + _" intro -name: "HOL."
  6212 
  6213 prints the last 100 theorems matching the pattern "(_::nat) + _ + _",
  6214 matching the current goal as introduction rule and not having "HOL."
  6215 in their name (i.e. not being defined in theory HOL).
  6216 
  6217 * Command 'thms_containing' has been discontinued in favour of
  6218 'find_theorems'; INCOMPATIBILITY.
  6219 
  6220 * Communication with Proof General is now 8bit clean, which means that
  6221 Unicode text in UTF-8 encoding may be used within theory texts (both
  6222 formal and informal parts).  Cf. option -U of the Isabelle Proof
  6223 General interface.  Here are some simple examples (cf. src/HOL/ex):
  6224 
  6225   http://isabelle.in.tum.de/library/HOL/ex/Hebrew.html
  6226   http://isabelle.in.tum.de/library/HOL/ex/Chinese.html
  6227 
  6228 * Improved efficiency of the Simplifier and, to a lesser degree, the
  6229 Classical Reasoner.  Typical big applications run around 2 times
  6230 faster.
  6231 
  6232 
  6233 *** Document preparation ***
  6234 
  6235 * Commands 'display_drafts' and 'print_drafts' perform simple output
  6236 of raw sources.  Only those symbols that do not require additional
  6237 LaTeX packages (depending on comments in isabellesym.sty) are
  6238 displayed properly, everything else is left verbatim.  isatool display
  6239 and isatool print are used as front ends (these are subject to the
  6240