NEWS
author wenzelm
Mon Jun 25 11:07:51 2012 +0200 (2012-06-25)
changeset 48120 9fe0e71052a0
parent 48111 33414f2e82ab
child 48171 28a6d67c93f0
permissions -rw-r--r--
updated "isar-ref" manual, reduced remaining material in "ref" manual.
     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 *** System ***
    17 
    18 * Discontinued support for Poly/ML 5.2.1, which was the last version
    19 without exception positions and advanced ML compiler/toplevel
    20 configuration.
    21 
    22 
    23 *** HOL ***
    24 
    25 * Simproc for rewriting set comprehensions into pointfree expressions
    26 
    27 * Quickcheck:
    28 
    29   - added an optimisation for equality premises.
    30     It is switched on by default, and can be switched off by setting
    31     the configuration quickcheck_optimise_equality to false.    
    32 
    33 * The SMT solver Z3 has now by default a restricted set of directly
    34 supported features. For the full set of features (div/mod, nonlinear
    35 arithmetic, datatypes/records) with potential proof reconstruction
    36 failures, enable the configuration option "z3_with_extensions".
    37 Minor INCOMPATIBILITY.
    38 
    39 * Sledgehammer:
    40 
    41   - Rationalized type encodings ("type_enc" option).
    42 
    43 
    44 
    45 New in Isabelle2012 (May 2012)
    46 ------------------------------
    47 
    48 *** General ***
    49 
    50 * Prover IDE (PIDE) improvements:
    51 
    52   - more robust Sledgehammer integration (as before the sledgehammer
    53     command-line needs to be typed into the source buffer)
    54   - markup for bound variables
    55   - markup for types of term variables (displayed as tooltips)
    56   - support for user-defined Isar commands within the running session
    57   - improved support for Unicode outside original 16bit range
    58     e.g. glyph for \<A> (thanks to jEdit 4.5.1)
    59 
    60 * Forward declaration of outer syntax keywords within the theory
    61 header -- minor INCOMPATIBILITY for user-defined commands.  Allow new
    62 commands to be used in the same theory where defined.
    63 
    64 * Auxiliary contexts indicate block structure for specifications with
    65 additional parameters and assumptions.  Such unnamed contexts may be
    66 nested within other targets, like 'theory', 'locale', 'class',
    67 'instantiation' etc.  Results from the local context are generalized
    68 accordingly and applied to the enclosing target context.  Example:
    69 
    70   context
    71     fixes x y z :: 'a
    72     assumes xy: "x = y" and yz: "y = z"
    73   begin
    74 
    75   lemma my_trans: "x = z" using xy yz by simp
    76 
    77   end
    78 
    79   thm my_trans
    80 
    81 The most basic application is to factor-out context elements of
    82 several fixes/assumes/shows theorem statements, e.g. see
    83 ~~/src/HOL/Isar_Examples/Group_Context.thy
    84 
    85 Any other local theory specification element works within the "context
    86 ... begin ... end" block as well.
    87 
    88 * Bundled declarations associate attributed fact expressions with a
    89 given name in the context.  These may be later included in other
    90 contexts.  This allows to manage context extensions casually, without
    91 the logical dependencies of locales and locale interpretation.  See
    92 commands 'bundle', 'include', 'including' etc. in the isar-ref manual.
    93 
    94 * Commands 'lemmas' and 'theorems' allow local variables using 'for'
    95 declaration, and results are standardized before being stored.  Thus
    96 old-style "standard" after instantiation or composition of facts
    97 becomes obsolete.  Minor INCOMPATIBILITY, due to potential change of
    98 indices of schematic variables.
    99 
   100 * Rule attributes in local theory declarations (e.g. locale or class)
   101 are now statically evaluated: the resulting theorem is stored instead
   102 of the original expression.  INCOMPATIBILITY in rare situations, where
   103 the historic accident of dynamic re-evaluation in interpretations
   104 etc. was exploited.
   105 
   106 * New tutorial "Programming and Proving in Isabelle/HOL"
   107 ("prog-prove").  It completely supersedes "A Tutorial Introduction to
   108 Structured Isar Proofs" ("isar-overview"), which has been removed.  It
   109 also supersedes "Isabelle/HOL, A Proof Assistant for Higher-Order
   110 Logic" as the recommended beginners tutorial, but does not cover all
   111 of the material of that old tutorial.
   112 
   113 * Updated and extended reference manuals: "isar-ref",
   114 "implementation", "system"; reduced remaining material in old "ref"
   115 manual.
   116 
   117 
   118 *** Pure ***
   119 
   120 * Command 'definition' no longer exports the foundational "raw_def"
   121 into the user context.  Minor INCOMPATIBILITY, may use the regular
   122 "def" result with attribute "abs_def" to imitate the old version.
   123 
   124 * Attribute "abs_def" turns an equation of the form "f x y == t" into
   125 "f == %x y. t", which ensures that "simp" or "unfold" steps always
   126 expand it.  This also works for object-logic equality.  (Formerly
   127 undocumented feature.)
   128 
   129 * Sort constraints are now propagated in simultaneous statements, just
   130 like type constraints.  INCOMPATIBILITY in rare situations, where
   131 distinct sorts used to be assigned accidentally.  For example:
   132 
   133   lemma "P (x::'a::foo)" and "Q (y::'a::bar)"  -- "now illegal"
   134 
   135   lemma "P (x::'a)" and "Q (y::'a::bar)"
   136     -- "now uniform 'a::bar instead of default sort for first occurrence (!)"
   137 
   138 * Rule composition via attribute "OF" (or ML functions OF/MRS) is more
   139 tolerant against multiple unifiers, as long as the final result is
   140 unique.  (As before, rules are composed in canonical right-to-left
   141 order to accommodate newly introduced premises.)
   142 
   143 * Renamed some inner syntax categories:
   144 
   145     num ~> num_token
   146     xnum ~> xnum_token
   147     xstr ~> str_token
   148 
   149 Minor INCOMPATIBILITY.  Note that in practice "num_const" or
   150 "num_position" etc. are mainly used instead (which also include
   151 position information via constraints).
   152 
   153 * Simplified configuration options for syntax ambiguity: see
   154 "syntax_ambiguity_warning" and "syntax_ambiguity_limit" in isar-ref
   155 manual.  Minor INCOMPATIBILITY.
   156 
   157 * Discontinued configuration option "syntax_positions": atomic terms
   158 in parse trees are always annotated by position constraints.
   159 
   160 * Old code generator for SML and its commands 'code_module',
   161 'code_library', 'consts_code', 'types_code' have been discontinued.
   162 Use commands of the generic code generator instead.  INCOMPATIBILITY.
   163 
   164 * Redundant attribute "code_inline" has been discontinued. Use
   165 "code_unfold" instead.  INCOMPATIBILITY.
   166 
   167 * Dropped attribute "code_unfold_post" in favor of the its dual
   168 "code_abbrev", which yields a common pattern in definitions like
   169 
   170   definition [code_abbrev]: "f = t"
   171 
   172 INCOMPATIBILITY.
   173 
   174 * Obsolete 'types' command has been discontinued.  Use 'type_synonym'
   175 instead.  INCOMPATIBILITY.
   176 
   177 * Discontinued old "prems" fact, which used to refer to the accidental
   178 collection of foundational premises in the context (already marked as
   179 legacy since Isabelle2011).
   180 
   181 
   182 *** HOL ***
   183 
   184 * Type 'a set is now a proper type constructor (just as before
   185 Isabelle2008).  Definitions mem_def and Collect_def have disappeared.
   186 Non-trivial INCOMPATIBILITY.  For developments keeping predicates and
   187 sets separate, it is often sufficient to rephrase some set S that has
   188 been accidentally used as predicates by "%x. x : S", and some
   189 predicate P that has been accidentally used as set by "{x. P x}".
   190 Corresponding proofs in a first step should be pruned from any
   191 tinkering with former theorems mem_def and Collect_def as far as
   192 possible.
   193 
   194 For developments which deliberately mix predicates and sets, a
   195 planning step is necessary to determine what should become a predicate
   196 and what a set.  It can be helpful to carry out that step in
   197 Isabelle2011-1 before jumping right into the current release.
   198 
   199 * Code generation by default implements sets as container type rather
   200 than predicates.  INCOMPATIBILITY.
   201 
   202 * New type synonym 'a rel = ('a * 'a) set
   203 
   204 * The representation of numerals has changed.  Datatype "num"
   205 represents strictly positive binary numerals, along with functions
   206 "numeral :: num => 'a" and "neg_numeral :: num => 'a" to represent
   207 positive and negated numeric literals, respectively.  See also
   208 definitions in ~~/src/HOL/Num.thy.  Potential INCOMPATIBILITY, some
   209 user theories may require adaptations as follows:
   210 
   211   - Theorems with number_ring or number_semiring constraints: These
   212     classes are gone; use comm_ring_1 or comm_semiring_1 instead.
   213 
   214   - Theories defining numeric types: Remove number, number_semiring,
   215     and number_ring instances. Defer all theorems about numerals until
   216     after classes one and semigroup_add have been instantiated.
   217 
   218   - Numeral-only simp rules: Replace each rule having a "number_of v"
   219     pattern with two copies, one for numeral and one for neg_numeral.
   220 
   221   - Theorems about subclasses of semiring_1 or ring_1: These classes
   222     automatically support numerals now, so more simp rules and
   223     simprocs may now apply within the proof.
   224 
   225   - Definitions and theorems using old constructors Pls/Min/Bit0/Bit1:
   226     Redefine using other integer operations.
   227 
   228 * Transfer: New package intended to generalize the existing
   229 "descending" method and related theorem attributes from the Quotient
   230 package.  (Not all functionality is implemented yet, but future
   231 development will focus on Transfer as an eventual replacement for the
   232 corresponding parts of the Quotient package.)
   233 
   234   - transfer_rule attribute: Maintains a collection of transfer rules,
   235     which relate constants at two different types. Transfer rules may
   236     relate different type instances of the same polymorphic constant,
   237     or they may relate an operation on a raw type to a corresponding
   238     operation on an abstract type (quotient or subtype). For example:
   239 
   240     ((A ===> B) ===> list_all2 A ===> list_all2 B) map map
   241     (cr_int ===> cr_int ===> cr_int) (%(x,y) (u,v). (x+u, y+v)) plus_int
   242 
   243   - transfer method: Replaces a subgoal on abstract types with an
   244     equivalent subgoal on the corresponding raw types. Constants are
   245     replaced with corresponding ones according to the transfer rules.
   246     Goals are generalized over all free variables by default; this is
   247     necessary for variables whose types change, but can be overridden
   248     for specific variables with e.g. "transfer fixing: x y z".  The
   249     variant transfer' method allows replacing a subgoal with one that
   250     is logically stronger (rather than equivalent).
   251 
   252   - relator_eq attribute: Collects identity laws for relators of
   253     various type constructors, e.g. "list_all2 (op =) = (op =)".  The
   254     transfer method uses these lemmas to infer transfer rules for
   255     non-polymorphic constants on the fly.
   256 
   257   - transfer_prover method: Assists with proving a transfer rule for a
   258     new constant, provided the constant is defined in terms of other
   259     constants that already have transfer rules. It should be applied
   260     after unfolding the constant definitions.
   261 
   262   - HOL/ex/Transfer_Int_Nat.thy: Example theory demonstrating transfer
   263     from type nat to type int.
   264 
   265 * Lifting: New package intended to generalize the quotient_definition
   266 facility of the Quotient package; designed to work with Transfer.
   267 
   268   - lift_definition command: Defines operations on an abstract type in
   269     terms of a corresponding operation on a representation
   270     type.  Example syntax:
   271 
   272     lift_definition dlist_insert :: "'a => 'a dlist => 'a dlist"
   273       is List.insert
   274 
   275     Users must discharge a respectfulness proof obligation when each
   276     constant is defined. (For a type copy, i.e. a typedef with UNIV,
   277     the proof is discharged automatically.) The obligation is
   278     presented in a user-friendly, readable form; a respectfulness
   279     theorem in the standard format and a transfer rule are generated
   280     by the package.
   281 
   282   - Integration with code_abstype: For typedefs (e.g. subtypes
   283     corresponding to a datatype invariant, such as dlist),
   284     lift_definition generates a code certificate theorem and sets up
   285     code generation for each constant.
   286 
   287   - setup_lifting command: Sets up the Lifting package to work with a
   288     user-defined type. The user must provide either a quotient theorem
   289     or a type_definition theorem.  The package configures transfer
   290     rules for equality and quantifiers on the type, and sets up the
   291     lift_definition command to work with the type.
   292 
   293   - Usage examples: See Quotient_Examples/Lift_DList.thy,
   294     Quotient_Examples/Lift_RBT.thy, Quotient_Examples/Lift_FSet.thy,
   295     Word/Word.thy and Library/Float.thy.
   296 
   297 * Quotient package:
   298 
   299   - The 'quotient_type' command now supports a 'morphisms' option with
   300     rep and abs functions, similar to typedef.
   301 
   302   - 'quotient_type' sets up new types to work with the Lifting and
   303     Transfer packages, as with 'setup_lifting'.
   304 
   305   - The 'quotient_definition' command now requires the user to prove a
   306     respectfulness property at the point where the constant is
   307     defined, similar to lift_definition; INCOMPATIBILITY.
   308 
   309   - Renamed predicate 'Quotient' to 'Quotient3', and renamed theorems
   310     accordingly, INCOMPATIBILITY.
   311 
   312 * New diagnostic command 'find_unused_assms' to find potentially
   313 superfluous assumptions in theorems using Quickcheck.
   314 
   315 * Quickcheck:
   316 
   317   - Quickcheck returns variable assignments as counterexamples, which
   318     allows to reveal the underspecification of functions under test.
   319     For example, refuting "hd xs = x", it presents the variable
   320     assignment xs = [] and x = a1 as a counterexample, assuming that
   321     any property is false whenever "hd []" occurs in it.
   322 
   323     These counterexample are marked as potentially spurious, as
   324     Quickcheck also returns "xs = []" as a counterexample to the
   325     obvious theorem "hd xs = hd xs".
   326 
   327     After finding a potentially spurious counterexample, Quickcheck
   328     continues searching for genuine ones.
   329 
   330     By default, Quickcheck shows potentially spurious and genuine
   331     counterexamples. The option "genuine_only" sets quickcheck to only
   332     show genuine counterexamples.
   333 
   334   - The command 'quickcheck_generator' creates random and exhaustive
   335     value generators for a given type and operations.
   336 
   337     It generates values by using the operations as if they were
   338     constructors of that type.
   339 
   340   - Support for multisets.
   341 
   342   - Added "use_subtype" options.
   343 
   344   - Added "quickcheck_locale" configuration to specify how to process
   345     conjectures in a locale context.
   346 
   347 * Nitpick: Fixed infinite loop caused by the 'peephole_optim' option
   348 and affecting 'rat' and 'real'.
   349 
   350 * Sledgehammer:
   351   - Integrated more tightly with SPASS, as described in the ITP 2012
   352     paper "More SPASS with Isabelle".
   353   - Made it try "smt" as a fallback if "metis" fails or times out.
   354   - Added support for the following provers: Alt-Ergo (via Why3 and
   355     TFF1), iProver, iProver-Eq.
   356   - Replaced remote E-SInE with remote Satallax in the default setup.
   357   - Sped up the minimizer.
   358   - Added "lam_trans", "uncurry_aliases", and "minimize" options.
   359   - Renamed "slicing" ("no_slicing") option to "slice" ("dont_slice").
   360   - Renamed "sound" option to "strict".
   361 
   362 * Metis: Added possibility to specify lambda translations scheme as a
   363 parenthesized argument (e.g., "by (metis (lifting) ...)").
   364 
   365 * SMT: Renamed "smt_fixed" option to "smt_read_only_certificates".
   366 
   367 * Command 'try0': Renamed from 'try_methods'. INCOMPATIBILITY.
   368 
   369 * New "case_product" attribute to generate a case rule doing multiple
   370 case distinctions at the same time.  E.g.
   371 
   372   list.exhaust [case_product nat.exhaust]
   373 
   374 produces a rule which can be used to perform case distinction on both
   375 a list and a nat.
   376 
   377 * New "eventually_elim" method as a generalized variant of the
   378 eventually_elim* rules.  Supports structured proofs.
   379 
   380 * Typedef with implicit set definition is considered legacy.  Use
   381 "typedef (open)" form instead, which will eventually become the
   382 default.
   383 
   384 * Record: code generation can be switched off manually with
   385 
   386   declare [[record_coden = false]]  -- "default true"
   387 
   388 * Datatype: type parameters allow explicit sort constraints.
   389 
   390 * Concrete syntax for case expressions includes constraints for source
   391 positions, and thus produces Prover IDE markup for its bindings.
   392 INCOMPATIBILITY for old-style syntax translations that augment the
   393 pattern notation; e.g. see src/HOL/HOLCF/One.thy for translations of
   394 one_case.
   395 
   396 * Clarified attribute "mono_set": pure declaration without modifying
   397 the result of the fact expression.
   398 
   399 * More default pred/set conversions on a couple of relation operations
   400 and predicates.  Added powers of predicate relations.  Consolidation
   401 of some relation theorems:
   402 
   403   converse_def ~> converse_unfold
   404   rel_comp_def ~> relcomp_unfold
   405   symp_def ~> (modified, use symp_def and sym_def instead)
   406   transp_def ~> transp_trans
   407   Domain_def ~> Domain_unfold
   408   Range_def ~> Domain_converse [symmetric]
   409 
   410 Generalized theorems INF_INT_eq, INF_INT_eq2, SUP_UN_eq, SUP_UN_eq2.
   411 
   412 See theory "Relation" for examples for making use of pred/set
   413 conversions by means of attributes "to_set" and "to_pred".
   414 
   415 INCOMPATIBILITY.
   416 
   417 * Renamed facts about the power operation on relations, i.e., relpow
   418 to match the constant's name:
   419 
   420   rel_pow_1 ~> relpow_1
   421   rel_pow_0_I ~> relpow_0_I
   422   rel_pow_Suc_I ~> relpow_Suc_I
   423   rel_pow_Suc_I2 ~> relpow_Suc_I2
   424   rel_pow_0_E ~> relpow_0_E
   425   rel_pow_Suc_E ~> relpow_Suc_E
   426   rel_pow_E ~> relpow_E
   427   rel_pow_Suc_D2 ~> relpow_Suc_D2
   428   rel_pow_Suc_E2 ~> relpow_Suc_E2
   429   rel_pow_Suc_D2' ~> relpow_Suc_D2'
   430   rel_pow_E2 ~> relpow_E2
   431   rel_pow_add ~> relpow_add
   432   rel_pow_commute ~> relpow
   433   rel_pow_empty ~> relpow_empty:
   434   rtrancl_imp_UN_rel_pow ~> rtrancl_imp_UN_relpow
   435   rel_pow_imp_rtrancl ~> relpow_imp_rtrancl
   436   rtrancl_is_UN_rel_pow ~> rtrancl_is_UN_relpow
   437   rtrancl_imp_rel_pow ~> rtrancl_imp_relpow
   438   rel_pow_fun_conv ~> relpow_fun_conv
   439   rel_pow_finite_bounded1 ~> relpow_finite_bounded1
   440   rel_pow_finite_bounded ~> relpow_finite_bounded
   441   rtrancl_finite_eq_rel_pow ~> rtrancl_finite_eq_relpow
   442   trancl_finite_eq_rel_pow ~> trancl_finite_eq_relpow
   443   single_valued_rel_pow ~> single_valued_relpow
   444 
   445 INCOMPATIBILITY.
   446 
   447 * Theory Relation: Consolidated constant name for relation composition
   448 and corresponding theorem names:
   449 
   450   - Renamed constant rel_comp to relcomp.
   451 
   452   - Dropped abbreviation pred_comp. Use relcompp instead.
   453 
   454   - Renamed theorems:
   455 
   456     rel_compI ~> relcompI
   457     rel_compEpair ~> relcompEpair
   458     rel_compE ~> relcompE
   459     pred_comp_rel_comp_eq ~> relcompp_relcomp_eq
   460     rel_comp_empty1 ~> relcomp_empty1
   461     rel_comp_mono ~> relcomp_mono
   462     rel_comp_subset_Sigma ~> relcomp_subset_Sigma
   463     rel_comp_distrib ~> relcomp_distrib
   464     rel_comp_distrib2 ~> relcomp_distrib2
   465     rel_comp_UNION_distrib ~> relcomp_UNION_distrib
   466     rel_comp_UNION_distrib2 ~> relcomp_UNION_distrib2
   467     single_valued_rel_comp ~> single_valued_relcomp
   468     rel_comp_def ~> relcomp_unfold
   469     converse_rel_comp ~> converse_relcomp
   470     pred_compI ~> relcomppI
   471     pred_compE ~> relcomppE
   472     pred_comp_bot1 ~> relcompp_bot1
   473     pred_comp_bot2 ~> relcompp_bot2
   474     transp_pred_comp_less_eq ~> transp_relcompp_less_eq
   475     pred_comp_mono ~> relcompp_mono
   476     pred_comp_distrib ~> relcompp_distrib
   477     pred_comp_distrib2 ~> relcompp_distrib2
   478     converse_pred_comp ~> converse_relcompp
   479 
   480     finite_rel_comp ~> finite_relcomp
   481 
   482     set_rel_comp ~> set_relcomp
   483 
   484 INCOMPATIBILITY.
   485 
   486 * Theory Divides: Discontinued redundant theorems about div and mod.
   487 INCOMPATIBILITY, use the corresponding generic theorems instead.
   488 
   489   DIVISION_BY_ZERO ~> div_by_0, mod_by_0
   490   zdiv_self ~> div_self
   491   zmod_self ~> mod_self
   492   zdiv_zero ~> div_0
   493   zmod_zero ~> mod_0
   494   zdiv_zmod_equality ~> div_mod_equality2
   495   zdiv_zmod_equality2 ~> div_mod_equality
   496   zmod_zdiv_trivial ~> mod_div_trivial
   497   zdiv_zminus_zminus ~> div_minus_minus
   498   zmod_zminus_zminus ~> mod_minus_minus
   499   zdiv_zminus2 ~> div_minus_right
   500   zmod_zminus2 ~> mod_minus_right
   501   zdiv_minus1_right ~> div_minus1_right
   502   zmod_minus1_right ~> mod_minus1_right
   503   zdvd_mult_div_cancel ~> dvd_mult_div_cancel
   504   zmod_zmult1_eq ~> mod_mult_right_eq
   505   zpower_zmod ~> power_mod
   506   zdvd_zmod ~> dvd_mod
   507   zdvd_zmod_imp_zdvd ~> dvd_mod_imp_dvd
   508   mod_mult_distrib ~> mult_mod_left
   509   mod_mult_distrib2 ~> mult_mod_right
   510 
   511 * Removed redundant theorems nat_mult_2 and nat_mult_2_right; use
   512 generic mult_2 and mult_2_right instead. INCOMPATIBILITY.
   513 
   514 * Finite_Set.fold now qualified.  INCOMPATIBILITY.
   515 
   516 * Consolidated theorem names concerning fold combinators:
   517 
   518   inf_INFI_fold_inf ~> inf_INF_fold_inf
   519   sup_SUPR_fold_sup ~> sup_SUP_fold_sup
   520   INFI_fold_inf ~> INF_fold_inf
   521   SUPR_fold_sup ~> SUP_fold_sup
   522   union_set ~> union_set_fold
   523   minus_set ~> minus_set_fold
   524   INFI_set_fold ~> INF_set_fold
   525   SUPR_set_fold ~> SUP_set_fold
   526   INF_code ~> INF_set_foldr
   527   SUP_code ~> SUP_set_foldr
   528   foldr.simps ~> foldr.simps (in point-free formulation)
   529   foldr_fold_rev ~> foldr_conv_fold
   530   foldl_fold ~> foldl_conv_fold
   531   foldr_foldr ~> foldr_conv_foldl
   532   foldl_foldr ~> foldl_conv_foldr
   533   fold_set_remdups ~> fold_set_fold_remdups
   534   fold_set ~> fold_set_fold
   535   fold1_set ~> fold1_set_fold
   536 
   537 INCOMPATIBILITY.
   538 
   539 * Dropped rarely useful theorems concerning fold combinators:
   540 foldl_apply, foldl_fun_comm, foldl_rev, fold_weak_invariant,
   541 rev_foldl_cons, fold_set_remdups, fold_set, fold_set1,
   542 concat_conv_foldl, foldl_weak_invariant, foldl_invariant,
   543 foldr_invariant, foldl_absorb0, foldl_foldr1_lemma, foldl_foldr1,
   544 listsum_conv_fold, listsum_foldl, sort_foldl_insort, foldl_assoc,
   545 foldr_conv_foldl, start_le_sum, elem_le_sum, sum_eq_0_conv.
   546 INCOMPATIBILITY.  For the common phrases "%xs. List.foldr plus xs 0"
   547 and "List.foldl plus 0", prefer "List.listsum".  Otherwise it can be
   548 useful to boil down "List.foldr" and "List.foldl" to "List.fold" by
   549 unfolding "foldr_conv_fold" and "foldl_conv_fold".
   550 
   551 * Dropped lemmas minus_set_foldr, union_set_foldr, union_coset_foldr,
   552 inter_coset_foldr, Inf_fin_set_foldr, Sup_fin_set_foldr,
   553 Min_fin_set_foldr, Max_fin_set_foldr, Inf_set_foldr, Sup_set_foldr,
   554 INF_set_foldr, SUP_set_foldr.  INCOMPATIBILITY.  Prefer corresponding
   555 lemmas over fold rather than foldr, or make use of lemmas
   556 fold_conv_foldr and fold_rev.
   557 
   558 * Congruence rules Option.map_cong and Option.bind_cong for recursion
   559 through option types.
   560 
   561 * "Transitive_Closure.ntrancl": bounded transitive closure on
   562 relations.
   563 
   564 * Constant "Set.not_member" now qualified.  INCOMPATIBILITY.
   565 
   566 * Theory Int: Discontinued many legacy theorems specific to type int.
   567 INCOMPATIBILITY, use the corresponding generic theorems instead.
   568 
   569   zminus_zminus ~> minus_minus
   570   zminus_0 ~> minus_zero
   571   zminus_zadd_distrib ~> minus_add_distrib
   572   zadd_commute ~> add_commute
   573   zadd_assoc ~> add_assoc
   574   zadd_left_commute ~> add_left_commute
   575   zadd_ac ~> add_ac
   576   zmult_ac ~> mult_ac
   577   zadd_0 ~> add_0_left
   578   zadd_0_right ~> add_0_right
   579   zadd_zminus_inverse2 ~> left_minus
   580   zmult_zminus ~> mult_minus_left
   581   zmult_commute ~> mult_commute
   582   zmult_assoc ~> mult_assoc
   583   zadd_zmult_distrib ~> left_distrib
   584   zadd_zmult_distrib2 ~> right_distrib
   585   zdiff_zmult_distrib ~> left_diff_distrib
   586   zdiff_zmult_distrib2 ~> right_diff_distrib
   587   zmult_1 ~> mult_1_left
   588   zmult_1_right ~> mult_1_right
   589   zle_refl ~> order_refl
   590   zle_trans ~> order_trans
   591   zle_antisym ~> order_antisym
   592   zle_linear ~> linorder_linear
   593   zless_linear ~> linorder_less_linear
   594   zadd_left_mono ~> add_left_mono
   595   zadd_strict_right_mono ~> add_strict_right_mono
   596   zadd_zless_mono ~> add_less_le_mono
   597   int_0_less_1 ~> zero_less_one
   598   int_0_neq_1 ~> zero_neq_one
   599   zless_le ~> less_le
   600   zpower_zadd_distrib ~> power_add
   601   zero_less_zpower_abs_iff ~> zero_less_power_abs_iff
   602   zero_le_zpower_abs ~> zero_le_power_abs
   603 
   604 * Theory Deriv: Renamed
   605 
   606   DERIV_nonneg_imp_nonincreasing ~> DERIV_nonneg_imp_nondecreasing
   607 
   608 * Theory Library/Multiset: Improved code generation of multisets.
   609 
   610 * Theory HOL/Library/Set_Algebras: Addition and multiplication on sets
   611 are expressed via type classes again. The special syntax
   612 \<oplus>/\<otimes> has been replaced by plain +/*. Removed constant
   613 setsum_set, which is now subsumed by Big_Operators.setsum.
   614 INCOMPATIBILITY.
   615 
   616 * Theory HOL/Library/Diagonalize has been removed. INCOMPATIBILITY,
   617 use theory HOL/Library/Nat_Bijection instead.
   618 
   619 * Theory HOL/Library/RBT_Impl: Backing implementation of red-black
   620 trees is now inside a type class context.  Names of affected
   621 operations and lemmas have been prefixed by rbt_.  INCOMPATIBILITY for
   622 theories working directly with raw red-black trees, adapt the names as
   623 follows:
   624 
   625   Operations:
   626   bulkload -> rbt_bulkload
   627   del_from_left -> rbt_del_from_left
   628   del_from_right -> rbt_del_from_right
   629   del -> rbt_del
   630   delete -> rbt_delete
   631   ins -> rbt_ins
   632   insert -> rbt_insert
   633   insertw -> rbt_insert_with
   634   insert_with_key -> rbt_insert_with_key
   635   map_entry -> rbt_map_entry
   636   lookup -> rbt_lookup
   637   sorted -> rbt_sorted
   638   tree_greater -> rbt_greater
   639   tree_less -> rbt_less
   640   tree_less_symbol -> rbt_less_symbol
   641   union -> rbt_union
   642   union_with -> rbt_union_with
   643   union_with_key -> rbt_union_with_key
   644 
   645   Lemmas:
   646   balance_left_sorted -> balance_left_rbt_sorted
   647   balance_left_tree_greater -> balance_left_rbt_greater
   648   balance_left_tree_less -> balance_left_rbt_less
   649   balance_right_sorted -> balance_right_rbt_sorted
   650   balance_right_tree_greater -> balance_right_rbt_greater
   651   balance_right_tree_less -> balance_right_rbt_less
   652   balance_sorted -> balance_rbt_sorted
   653   balance_tree_greater -> balance_rbt_greater
   654   balance_tree_less -> balance_rbt_less
   655   bulkload_is_rbt -> rbt_bulkload_is_rbt
   656   combine_sorted -> combine_rbt_sorted
   657   combine_tree_greater -> combine_rbt_greater
   658   combine_tree_less -> combine_rbt_less
   659   delete_in_tree -> rbt_delete_in_tree
   660   delete_is_rbt -> rbt_delete_is_rbt
   661   del_from_left_tree_greater -> rbt_del_from_left_rbt_greater
   662   del_from_left_tree_less -> rbt_del_from_left_rbt_less
   663   del_from_right_tree_greater -> rbt_del_from_right_rbt_greater
   664   del_from_right_tree_less -> rbt_del_from_right_rbt_less
   665   del_in_tree -> rbt_del_in_tree
   666   del_inv1_inv2 -> rbt_del_inv1_inv2
   667   del_sorted -> rbt_del_rbt_sorted
   668   del_tree_greater -> rbt_del_rbt_greater
   669   del_tree_less -> rbt_del_rbt_less
   670   dom_lookup_Branch -> dom_rbt_lookup_Branch
   671   entries_lookup -> entries_rbt_lookup
   672   finite_dom_lookup -> finite_dom_rbt_lookup
   673   insert_sorted -> rbt_insert_rbt_sorted
   674   insertw_is_rbt -> rbt_insertw_is_rbt
   675   insertwk_is_rbt -> rbt_insertwk_is_rbt
   676   insertwk_sorted -> rbt_insertwk_rbt_sorted
   677   insertw_sorted -> rbt_insertw_rbt_sorted
   678   ins_sorted -> ins_rbt_sorted
   679   ins_tree_greater -> ins_rbt_greater
   680   ins_tree_less -> ins_rbt_less
   681   is_rbt_sorted -> is_rbt_rbt_sorted
   682   lookup_balance -> rbt_lookup_balance
   683   lookup_bulkload -> rbt_lookup_rbt_bulkload
   684   lookup_delete -> rbt_lookup_rbt_delete
   685   lookup_Empty -> rbt_lookup_Empty
   686   lookup_from_in_tree -> rbt_lookup_from_in_tree
   687   lookup_in_tree -> rbt_lookup_in_tree
   688   lookup_ins -> rbt_lookup_ins
   689   lookup_insert -> rbt_lookup_rbt_insert
   690   lookup_insertw -> rbt_lookup_rbt_insertw
   691   lookup_insertwk -> rbt_lookup_rbt_insertwk
   692   lookup_keys -> rbt_lookup_keys
   693   lookup_map -> rbt_lookup_map
   694   lookup_map_entry -> rbt_lookup_rbt_map_entry
   695   lookup_tree_greater -> rbt_lookup_rbt_greater
   696   lookup_tree_less -> rbt_lookup_rbt_less
   697   lookup_union -> rbt_lookup_rbt_union
   698   map_entry_color_of -> rbt_map_entry_color_of
   699   map_entry_inv1 -> rbt_map_entry_inv1
   700   map_entry_inv2 -> rbt_map_entry_inv2
   701   map_entry_is_rbt -> rbt_map_entry_is_rbt
   702   map_entry_sorted -> rbt_map_entry_rbt_sorted
   703   map_entry_tree_greater -> rbt_map_entry_rbt_greater
   704   map_entry_tree_less -> rbt_map_entry_rbt_less
   705   map_tree_greater -> map_rbt_greater
   706   map_tree_less -> map_rbt_less
   707   map_sorted -> map_rbt_sorted
   708   paint_sorted -> paint_rbt_sorted
   709   paint_lookup -> paint_rbt_lookup
   710   paint_tree_greater -> paint_rbt_greater
   711   paint_tree_less -> paint_rbt_less
   712   sorted_entries -> rbt_sorted_entries
   713   tree_greater_eq_trans -> rbt_greater_eq_trans
   714   tree_greater_nit -> rbt_greater_nit
   715   tree_greater_prop -> rbt_greater_prop
   716   tree_greater_simps -> rbt_greater_simps
   717   tree_greater_trans -> rbt_greater_trans
   718   tree_less_eq_trans -> rbt_less_eq_trans
   719   tree_less_nit -> rbt_less_nit
   720   tree_less_prop -> rbt_less_prop
   721   tree_less_simps -> rbt_less_simps
   722   tree_less_trans -> rbt_less_trans
   723   tree_ord_props -> rbt_ord_props
   724   union_Branch -> rbt_union_Branch
   725   union_is_rbt -> rbt_union_is_rbt
   726   unionw_is_rbt -> rbt_unionw_is_rbt
   727   unionwk_is_rbt -> rbt_unionwk_is_rbt
   728   unionwk_sorted -> rbt_unionwk_rbt_sorted
   729 
   730 * Theory HOL/Library/Float: Floating point numbers are now defined as
   731 a subset of the real numbers.  All operations are defined using the
   732 lifing-framework and proofs use the transfer method.  INCOMPATIBILITY.
   733 
   734   Changed Operations:
   735   float_abs -> abs
   736   float_nprt -> nprt
   737   float_pprt -> pprt
   738   pow2 -> use powr
   739   round_down -> float_round_down
   740   round_up -> float_round_up
   741   scale -> exponent
   742 
   743   Removed Operations:
   744   ceiling_fl, lb_mult, lb_mod, ub_mult, ub_mod
   745 
   746   Renamed Lemmas:
   747   abs_float_def -> Float.compute_float_abs
   748   bitlen_ge0 -> bitlen_nonneg
   749   bitlen.simps -> Float.compute_bitlen
   750   float_components -> Float_mantissa_exponent
   751   float_divl.simps -> Float.compute_float_divl
   752   float_divr.simps -> Float.compute_float_divr
   753   float_eq_odd -> mult_powr_eq_mult_powr_iff
   754   float_power -> real_of_float_power
   755   lapprox_posrat_def -> Float.compute_lapprox_posrat
   756   lapprox_rat.simps -> Float.compute_lapprox_rat
   757   le_float_def' -> Float.compute_float_le
   758   le_float_def -> less_eq_float.rep_eq
   759   less_float_def' -> Float.compute_float_less
   760   less_float_def -> less_float.rep_eq
   761   normfloat_def -> Float.compute_normfloat
   762   normfloat_imp_odd_or_zero -> mantissa_not_dvd and mantissa_noteq_0
   763   normfloat -> normfloat_def
   764   normfloat_unique -> use normfloat_def
   765   number_of_float_Float -> Float.compute_float_numeral, Float.compute_float_neg_numeral
   766   one_float_def -> Float.compute_float_one
   767   plus_float_def -> Float.compute_float_plus
   768   rapprox_posrat_def -> Float.compute_rapprox_posrat
   769   rapprox_rat.simps -> Float.compute_rapprox_rat
   770   real_of_float_0 -> zero_float.rep_eq
   771   real_of_float_1 -> one_float.rep_eq
   772   real_of_float_abs -> abs_float.rep_eq
   773   real_of_float_add -> plus_float.rep_eq
   774   real_of_float_minus -> uminus_float.rep_eq
   775   real_of_float_mult -> times_float.rep_eq
   776   real_of_float_simp -> Float.rep_eq
   777   real_of_float_sub -> minus_float.rep_eq
   778   round_down.simps -> Float.compute_float_round_down
   779   round_up.simps -> Float.compute_float_round_up
   780   times_float_def -> Float.compute_float_times
   781   uminus_float_def -> Float.compute_float_uminus
   782   zero_float_def -> Float.compute_float_zero
   783 
   784   Lemmas not necessary anymore, use the transfer method:
   785   bitlen_B0, bitlen_B1, bitlen_ge1, bitlen_Min, bitlen_Pls, float_divl,
   786   float_divr, float_le_simp, float_less1_mantissa_bound,
   787   float_less_simp, float_less_zero, float_le_zero,
   788   float_pos_less1_e_neg, float_pos_m_pos, float_split, float_split2,
   789   floor_pos_exp, lapprox_posrat, lapprox_posrat_bottom, lapprox_rat,
   790   lapprox_rat_bottom, normalized_float, rapprox_posrat,
   791   rapprox_posrat_le1, rapprox_rat, real_of_float_ge0_exp,
   792   real_of_float_neg_exp, real_of_float_nge0_exp, round_down floor_fl,
   793   round_up, zero_le_float, zero_less_float
   794 
   795 * New theory HOL/Library/DAList provides an abstract type for
   796 association lists with distinct keys.
   797 
   798 * Session HOL/IMP: Added new theory of abstract interpretation of
   799 annotated commands.
   800 
   801 * Session HOL-Import: Re-implementation from scratch is faster,
   802 simpler, and more scalable.  Requires a proof bundle, which is
   803 available as an external component.  Discontinued old (and mostly
   804 dead) Importer for HOL4 and HOL Light.  INCOMPATIBILITY.
   805 
   806 * Session HOL-Word: Discontinued many redundant theorems specific to
   807 type 'a word. INCOMPATIBILITY, use the corresponding generic theorems
   808 instead.
   809 
   810   word_sub_alt ~> word_sub_wi
   811   word_add_alt ~> word_add_def
   812   word_mult_alt ~> word_mult_def
   813   word_minus_alt ~> word_minus_def
   814   word_0_alt ~> word_0_wi
   815   word_1_alt ~> word_1_wi
   816   word_add_0 ~> add_0_left
   817   word_add_0_right ~> add_0_right
   818   word_mult_1 ~> mult_1_left
   819   word_mult_1_right ~> mult_1_right
   820   word_add_commute ~> add_commute
   821   word_add_assoc ~> add_assoc
   822   word_add_left_commute ~> add_left_commute
   823   word_mult_commute ~> mult_commute
   824   word_mult_assoc ~> mult_assoc
   825   word_mult_left_commute ~> mult_left_commute
   826   word_left_distrib ~> left_distrib
   827   word_right_distrib ~> right_distrib
   828   word_left_minus ~> left_minus
   829   word_diff_0_right ~> diff_0_right
   830   word_diff_self ~> diff_self
   831   word_sub_def ~> diff_minus
   832   word_diff_minus ~> diff_minus
   833   word_add_ac ~> add_ac
   834   word_mult_ac ~> mult_ac
   835   word_plus_ac0 ~> add_0_left add_0_right add_ac
   836   word_times_ac1 ~> mult_1_left mult_1_right mult_ac
   837   word_order_trans ~> order_trans
   838   word_order_refl ~> order_refl
   839   word_order_antisym ~> order_antisym
   840   word_order_linear ~> linorder_linear
   841   lenw1_zero_neq_one ~> zero_neq_one
   842   word_number_of_eq ~> number_of_eq
   843   word_of_int_add_hom ~> wi_hom_add
   844   word_of_int_sub_hom ~> wi_hom_sub
   845   word_of_int_mult_hom ~> wi_hom_mult
   846   word_of_int_minus_hom ~> wi_hom_neg
   847   word_of_int_succ_hom ~> wi_hom_succ
   848   word_of_int_pred_hom ~> wi_hom_pred
   849   word_of_int_0_hom ~> word_0_wi
   850   word_of_int_1_hom ~> word_1_wi
   851 
   852 * Session HOL-Word: New proof method "word_bitwise" for splitting
   853 machine word equalities and inequalities into logical circuits,
   854 defined in HOL/Word/WordBitwise.thy.  Supports addition, subtraction,
   855 multiplication, shifting by constants, bitwise operators and numeric
   856 constants.  Requires fixed-length word types, not 'a word.  Solves
   857 many standard word identities outright and converts more into first
   858 order problems amenable to blast or similar.  See also examples in
   859 HOL/Word/Examples/WordExamples.thy.
   860 
   861 * Session HOL-Probability: Introduced the type "'a measure" to
   862 represent measures, this replaces the records 'a algebra and 'a
   863 measure_space.  The locales based on subset_class now have two
   864 locale-parameters the space \<Omega> and the set of measurable sets M.
   865 The product of probability spaces uses now the same constant as the
   866 finite product of sigma-finite measure spaces "PiM :: ('i => 'a)
   867 measure".  Most constants are defined now outside of locales and gain
   868 an additional parameter, like null_sets, almost_eventually or \<mu>'.
   869 Measure space constructions for distributions and densities now got
   870 their own constants distr and density.  Instead of using locales to
   871 describe measure spaces with a finite space, the measure count_space
   872 and point_measure is introduced.  INCOMPATIBILITY.
   873 
   874   Renamed constants:
   875   measure -> emeasure
   876   finite_measure.\<mu>' -> measure
   877   product_algebra_generator -> prod_algebra
   878   product_prob_space.emb -> prod_emb
   879   product_prob_space.infprod_algebra -> PiM
   880 
   881   Removed locales:
   882   completeable_measure_space
   883   finite_measure_space
   884   finite_prob_space
   885   finite_product_finite_prob_space
   886   finite_product_sigma_algebra
   887   finite_sigma_algebra
   888   measure_space
   889   pair_finite_prob_space
   890   pair_finite_sigma_algebra
   891   pair_finite_space
   892   pair_sigma_algebra
   893   product_sigma_algebra
   894 
   895   Removed constants:
   896   conditional_space
   897   distribution -> use distr measure, or distributed predicate
   898   image_space
   899   joint_distribution -> use distr measure, or distributed predicate
   900   pair_measure_generator
   901   product_prob_space.infprod_algebra -> use PiM
   902   subvimage
   903 
   904   Replacement theorems:
   905   finite_additivity_sufficient -> ring_of_sets.countably_additiveI_finite
   906   finite_measure.empty_measure -> measure_empty
   907   finite_measure.finite_continuity_from_above -> finite_measure.finite_Lim_measure_decseq
   908   finite_measure.finite_continuity_from_below -> finite_measure.finite_Lim_measure_incseq
   909   finite_measure.finite_measure_countably_subadditive -> finite_measure.finite_measure_subadditive_countably
   910   finite_measure.finite_measure_eq -> finite_measure.emeasure_eq_measure
   911   finite_measure.finite_measure -> finite_measure.emeasure_finite
   912   finite_measure.finite_measure_finite_singleton -> finite_measure.finite_measure_eq_setsum_singleton
   913   finite_measure.positive_measure' -> measure_nonneg
   914   finite_measure.real_measure -> finite_measure.emeasure_real
   915   finite_product_prob_space.finite_measure_times -> finite_product_prob_space.finite_measure_PiM_emb
   916   finite_product_sigma_algebra.in_P -> sets_PiM_I_finite
   917   finite_product_sigma_algebra.P_empty -> space_PiM_empty, sets_PiM_empty
   918   information_space.conditional_entropy_eq -> information_space.conditional_entropy_simple_distributed
   919   information_space.conditional_entropy_positive -> information_space.conditional_entropy_nonneg_simple
   920   information_space.conditional_mutual_information_eq_mutual_information -> information_space.conditional_mutual_information_eq_mutual_information_simple
   921   information_space.conditional_mutual_information_generic_positive -> information_space.conditional_mutual_information_nonneg_simple
   922   information_space.conditional_mutual_information_positive -> information_space.conditional_mutual_information_nonneg_simple
   923   information_space.entropy_commute -> information_space.entropy_commute_simple
   924   information_space.entropy_eq -> information_space.entropy_simple_distributed
   925   information_space.entropy_generic_eq -> information_space.entropy_simple_distributed
   926   information_space.entropy_positive -> information_space.entropy_nonneg_simple
   927   information_space.entropy_uniform_max -> information_space.entropy_uniform
   928   information_space.KL_eq_0_imp -> information_space.KL_eq_0_iff_eq
   929   information_space.KL_eq_0 -> information_space.KL_same_eq_0
   930   information_space.KL_ge_0 -> information_space.KL_nonneg
   931   information_space.mutual_information_eq -> information_space.mutual_information_simple_distributed
   932   information_space.mutual_information_positive -> information_space.mutual_information_nonneg_simple
   933   Int_stable_cuboids -> Int_stable_atLeastAtMost
   934   Int_stable_product_algebra_generator -> positive_integral
   935   measure_preserving -> equality "distr M N f = N" "f : measurable M N"
   936   measure_space.additive -> emeasure_additive
   937   measure_space.AE_iff_null_set -> AE_iff_null
   938   measure_space.almost_everywhere_def -> eventually_ae_filter
   939   measure_space.almost_everywhere_vimage -> AE_distrD
   940   measure_space.continuity_from_above -> INF_emeasure_decseq
   941   measure_space.continuity_from_above_Lim -> Lim_emeasure_decseq
   942   measure_space.continuity_from_below_Lim -> Lim_emeasure_incseq
   943   measure_space.continuity_from_below -> SUP_emeasure_incseq
   944   measure_space_density -> emeasure_density
   945   measure_space.density_is_absolutely_continuous -> absolutely_continuousI_density
   946   measure_space.integrable_vimage -> integrable_distr
   947   measure_space.integral_translated_density -> integral_density
   948   measure_space.integral_vimage -> integral_distr
   949   measure_space.measure_additive -> plus_emeasure
   950   measure_space.measure_compl -> emeasure_compl
   951   measure_space.measure_countable_increasing -> emeasure_countable_increasing
   952   measure_space.measure_countably_subadditive -> emeasure_subadditive_countably
   953   measure_space.measure_decseq -> decseq_emeasure
   954   measure_space.measure_Diff -> emeasure_Diff
   955   measure_space.measure_Diff_null_set -> emeasure_Diff_null_set
   956   measure_space.measure_eq_0 -> emeasure_eq_0
   957   measure_space.measure_finitely_subadditive -> emeasure_subadditive_finite
   958   measure_space.measure_finite_singleton -> emeasure_eq_setsum_singleton
   959   measure_space.measure_incseq -> incseq_emeasure
   960   measure_space.measure_insert -> emeasure_insert
   961   measure_space.measure_mono -> emeasure_mono
   962   measure_space.measure_not_negative -> emeasure_not_MInf
   963   measure_space.measure_preserving_Int_stable -> measure_eqI_generator_eq
   964   measure_space.measure_setsum -> setsum_emeasure
   965   measure_space.measure_setsum_split -> setsum_emeasure_cover
   966   measure_space.measure_space_vimage -> emeasure_distr
   967   measure_space.measure_subadditive_finite -> emeasure_subadditive_finite
   968   measure_space.measure_subadditive -> subadditive
   969   measure_space.measure_top -> emeasure_space
   970   measure_space.measure_UN_eq_0 -> emeasure_UN_eq_0
   971   measure_space.measure_Un_null_set -> emeasure_Un_null_set
   972   measure_space.positive_integral_translated_density -> positive_integral_density
   973   measure_space.positive_integral_vimage -> positive_integral_distr
   974   measure_space.real_continuity_from_above -> Lim_measure_decseq
   975   measure_space.real_continuity_from_below -> Lim_measure_incseq
   976   measure_space.real_measure_countably_subadditive -> measure_subadditive_countably
   977   measure_space.real_measure_Diff -> measure_Diff
   978   measure_space.real_measure_finite_Union -> measure_finite_Union
   979   measure_space.real_measure_setsum_singleton -> measure_eq_setsum_singleton
   980   measure_space.real_measure_subadditive -> measure_subadditive
   981   measure_space.real_measure_Union -> measure_Union
   982   measure_space.real_measure_UNION -> measure_UNION
   983   measure_space.simple_function_vimage -> simple_function_comp
   984   measure_space.simple_integral_vimage -> simple_integral_distr
   985   measure_space.simple_integral_vimage -> simple_integral_distr
   986   measure_unique_Int_stable -> measure_eqI_generator_eq
   987   measure_unique_Int_stable_vimage -> measure_eqI_generator_eq
   988   pair_sigma_algebra.measurable_cut_fst -> sets_Pair1
   989   pair_sigma_algebra.measurable_cut_snd -> sets_Pair2
   990   pair_sigma_algebra.measurable_pair_image_fst -> measurable_Pair1
   991   pair_sigma_algebra.measurable_pair_image_snd -> measurable_Pair2
   992   pair_sigma_algebra.measurable_product_swap -> measurable_pair_swap_iff
   993   pair_sigma_algebra.pair_sigma_algebra_measurable -> measurable_pair_swap
   994   pair_sigma_algebra.pair_sigma_algebra_swap_measurable -> measurable_pair_swap'
   995   pair_sigma_algebra.sets_swap -> sets_pair_swap
   996   pair_sigma_finite.measure_cut_measurable_fst -> pair_sigma_finite.measurable_emeasure_Pair1
   997   pair_sigma_finite.measure_cut_measurable_snd -> pair_sigma_finite.measurable_emeasure_Pair2
   998   pair_sigma_finite.measure_preserving_swap -> pair_sigma_finite.distr_pair_swap
   999   pair_sigma_finite.pair_measure_alt2 -> pair_sigma_finite.emeasure_pair_measure_alt2
  1000   pair_sigma_finite.pair_measure_alt -> pair_sigma_finite.emeasure_pair_measure_alt
  1001   pair_sigma_finite.pair_measure_times -> pair_sigma_finite.emeasure_pair_measure_Times
  1002   prob_space.indep_distribution_eq_measure -> prob_space.indep_vars_iff_distr_eq_PiM
  1003   prob_space.indep_var_distributionD -> prob_space.indep_var_distribution_eq
  1004   prob_space.measure_space_1 -> prob_space.emeasure_space_1
  1005   prob_space.prob_space_vimage -> prob_space_distr
  1006   prob_space.random_variable_restrict -> measurable_restrict
  1007   prob_space_unique_Int_stable -> measure_eqI_prob_space
  1008   product_algebraE -> prod_algebraE_all
  1009   product_algebra_generator_der -> prod_algebra_eq_finite
  1010   product_algebra_generator_into_space -> prod_algebra_sets_into_space
  1011   product_algebraI -> sets_PiM_I_finite
  1012   product_measure_exists -> product_sigma_finite.sigma_finite
  1013   product_prob_space.finite_index_eq_finite_product -> product_prob_space.sets_PiM_generator
  1014   product_prob_space.finite_measure_infprod_emb_Pi -> product_prob_space.measure_PiM_emb
  1015   product_prob_space.infprod_spec -> product_prob_space.emeasure_PiM_emb_not_empty
  1016   product_prob_space.measurable_component -> measurable_component_singleton
  1017   product_prob_space.measurable_emb -> measurable_prod_emb
  1018   product_prob_space.measurable_into_infprod_algebra -> measurable_PiM_single
  1019   product_prob_space.measurable_singleton_infprod -> measurable_component_singleton
  1020   product_prob_space.measure_emb -> emeasure_prod_emb
  1021   product_prob_space.measure_preserving_restrict -> product_prob_space.distr_restrict
  1022   product_sigma_algebra.product_algebra_into_space -> space_closed
  1023   product_sigma_finite.measure_fold -> product_sigma_finite.distr_merge
  1024   product_sigma_finite.measure_preserving_component_singelton -> product_sigma_finite.distr_singleton
  1025   product_sigma_finite.measure_preserving_merge -> product_sigma_finite.distr_merge
  1026   sequence_space.measure_infprod -> sequence_space.measure_PiM_countable
  1027   sets_product_algebra -> sets_PiM
  1028   sigma_algebra.measurable_sigma -> measurable_measure_of
  1029   sigma_finite_measure.disjoint_sigma_finite -> sigma_finite_disjoint
  1030   sigma_finite_measure.RN_deriv_vimage -> sigma_finite_measure.RN_deriv_distr
  1031   sigma_product_algebra_sigma_eq -> sigma_prod_algebra_sigma_eq
  1032   space_product_algebra -> space_PiM
  1033 
  1034 * Session HOL-TPTP: support to parse and import TPTP problems (all
  1035 languages) into Isabelle/HOL.
  1036 
  1037 
  1038 *** FOL ***
  1039 
  1040 * New "case_product" attribute (see HOL).
  1041 
  1042 
  1043 *** ZF ***
  1044 
  1045 * Greater support for structured proofs involving induction or case
  1046 analysis.
  1047 
  1048 * Much greater use of mathematical symbols.
  1049 
  1050 * Removal of many ML theorem bindings.  INCOMPATIBILITY.
  1051 
  1052 
  1053 *** ML ***
  1054 
  1055 * Antiquotation @{keyword "name"} produces a parser for outer syntax
  1056 from a minor keyword introduced via theory header declaration.
  1057 
  1058 * Antiquotation @{command_spec "name"} produces the
  1059 Outer_Syntax.command_spec from a major keyword introduced via theory
  1060 header declaration; it can be passed to Outer_Syntax.command etc.
  1061 
  1062 * Local_Theory.define no longer hard-wires default theorem name
  1063 "foo_def", but retains the binding as given.  If that is Binding.empty
  1064 / Attrib.empty_binding, the result is not registered as user-level
  1065 fact.  The Local_Theory.define_internal variant allows to specify a
  1066 non-empty name (used for the foundation in the background theory),
  1067 while omitting the fact binding in the user-context.  Potential
  1068 INCOMPATIBILITY for derived definitional packages: need to specify
  1069 naming policy for primitive definitions more explicitly.
  1070 
  1071 * Renamed Thm.capply to Thm.apply, and Thm.cabs to Thm.lambda in
  1072 conformance with similar operations in structure Term and Logic.
  1073 
  1074 * Antiquotation @{attributes [...]} embeds attribute source
  1075 representation into the ML text, which is particularly useful with
  1076 declarations like Local_Theory.note.
  1077 
  1078 * Structure Proof_Context follows standard naming scheme.  Old
  1079 ProofContext has been discontinued.  INCOMPATIBILITY.
  1080 
  1081 * Refined Local_Theory.declaration {syntax, pervasive}, with subtle
  1082 change of semantics: update is applied to auxiliary local theory
  1083 context as well.
  1084 
  1085 * Modernized some old-style infix operations:
  1086 
  1087   addeqcongs    ~> Simplifier.add_eqcong
  1088   deleqcongs    ~> Simplifier.del_eqcong
  1089   addcongs      ~> Simplifier.add_cong
  1090   delcongs      ~> Simplifier.del_cong
  1091   setmksimps    ~> Simplifier.set_mksimps
  1092   setmkcong     ~> Simplifier.set_mkcong
  1093   setmksym      ~> Simplifier.set_mksym
  1094   setmkeqTrue   ~> Simplifier.set_mkeqTrue
  1095   settermless   ~> Simplifier.set_termless
  1096   setsubgoaler  ~> Simplifier.set_subgoaler
  1097   addsplits     ~> Splitter.add_split
  1098   delsplits     ~> Splitter.del_split
  1099 
  1100 
  1101 *** System ***
  1102 
  1103 * USER_HOME settings variable points to cross-platform user home
  1104 directory, which coincides with HOME on POSIX systems only.  Likewise,
  1105 the Isabelle path specification "~" now expands to $USER_HOME, instead
  1106 of former $HOME.  A different default for USER_HOME may be set
  1107 explicitly in shell environment, before Isabelle settings are
  1108 evaluated.  Minor INCOMPATIBILITY: need to adapt Isabelle path where
  1109 the generic user home was intended.
  1110 
  1111 * ISABELLE_HOME_WINDOWS refers to ISABELLE_HOME in windows file name
  1112 notation, which is useful for the jEdit file browser, for example.
  1113 
  1114 * ISABELLE_JDK_HOME settings variable points to JDK with javac and jar
  1115 (not just JRE).
  1116 
  1117 
  1118 
  1119 New in Isabelle2011-1 (October 2011)
  1120 ------------------------------------
  1121 
  1122 *** General ***
  1123 
  1124 * Improved Isabelle/jEdit Prover IDE (PIDE), which can be invoked as
  1125 "isabelle jedit" or "ISABELLE_HOME/Isabelle" on the command line.
  1126 
  1127   - Management of multiple theory files directly from the editor
  1128     buffer store -- bypassing the file-system (no requirement to save
  1129     files for checking).
  1130 
  1131   - Markup of formal entities within the text buffer, with semantic
  1132     highlighting, tooltips and hyperlinks to jump to defining source
  1133     positions.
  1134 
  1135   - Improved text rendering, with sub/superscripts in the source
  1136     buffer (including support for copy/paste wrt. output panel, HTML
  1137     theory output and other non-Isabelle text boxes).
  1138 
  1139   - Refined scheduling of proof checking and printing of results,
  1140     based on interactive editor view.  (Note: jEdit folding and
  1141     narrowing allows to restrict buffer perspectives explicitly.)
  1142 
  1143   - Reduced CPU performance requirements, usable on machines with few
  1144     cores.
  1145 
  1146   - Reduced memory requirements due to pruning of unused document
  1147     versions (garbage collection).
  1148 
  1149 See also ~~/src/Tools/jEdit/README.html for further information,
  1150 including some remaining limitations.
  1151 
  1152 * Theory loader: source files are exclusively located via the master
  1153 directory of each theory node (where the .thy file itself resides).
  1154 The global load path (such as src/HOL/Library) has been discontinued.
  1155 Note that the path element ~~ may be used to reference theories in the
  1156 Isabelle home folder -- for instance, "~~/src/HOL/Library/FuncSet".
  1157 INCOMPATIBILITY.
  1158 
  1159 * Theory loader: source files are identified by content via SHA1
  1160 digests.  Discontinued former path/modtime identification and optional
  1161 ISABELLE_FILE_IDENT plugin scripts.
  1162 
  1163 * Parallelization of nested Isar proofs is subject to
  1164 Goal.parallel_proofs_threshold (default 100).  See also isabelle
  1165 usedir option -Q.
  1166 
  1167 * Name space: former unsynchronized references are now proper
  1168 configuration options, with more conventional names:
  1169 
  1170   long_names   ~> names_long
  1171   short_names  ~> names_short
  1172   unique_names ~> names_unique
  1173 
  1174 Minor INCOMPATIBILITY, need to declare options in context like this:
  1175 
  1176   declare [[names_unique = false]]
  1177 
  1178 * Literal facts `prop` may contain dummy patterns, e.g. `_ = _`.  Note
  1179 that the result needs to be unique, which means fact specifications
  1180 may have to be refined after enriching a proof context.
  1181 
  1182 * Attribute "case_names" has been refined: the assumptions in each case
  1183 can be named now by following the case name with [name1 name2 ...].
  1184 
  1185 * Isabelle/Isar reference manual has been updated and extended:
  1186   - "Synopsis" provides a catalog of main Isar language concepts.
  1187   - Formal references in syntax diagrams, via @{rail} antiquotation.
  1188   - Updated material from classic "ref" manual, notably about
  1189     "Classical Reasoner".
  1190 
  1191 
  1192 *** HOL ***
  1193 
  1194 * Class bot and top require underlying partial order rather than
  1195 preorder: uniqueness of bot and top is guaranteed.  INCOMPATIBILITY.
  1196 
  1197 * Class complete_lattice: generalized a couple of lemmas from sets;
  1198 generalized theorems INF_cong and SUP_cong.  New type classes for
  1199 complete boolean algebras and complete linear orders.  Lemmas
  1200 Inf_less_iff, less_Sup_iff, INF_less_iff, less_SUP_iff now reside in
  1201 class complete_linorder.
  1202 
  1203 Changed proposition of lemmas Inf_bool_def, Sup_bool_def, Inf_fun_def,
  1204 Sup_fun_def, Inf_apply, Sup_apply.
  1205 
  1206 Removed redundant lemmas (the right hand side gives hints how to
  1207 replace them for (metis ...), or (simp only: ...) proofs):
  1208 
  1209   Inf_singleton ~> Inf_insert [where A="{}", unfolded Inf_empty inf_top_right]
  1210   Sup_singleton ~> Sup_insert [where A="{}", unfolded Sup_empty sup_bot_right]
  1211   Inf_binary ~> Inf_insert, Inf_empty, and inf_top_right
  1212   Sup_binary ~> Sup_insert, Sup_empty, and sup_bot_right
  1213   Int_eq_Inter ~> Inf_insert, Inf_empty, and inf_top_right
  1214   Un_eq_Union ~> Sup_insert, Sup_empty, and sup_bot_right
  1215   Inter_def ~> INF_def, image_def
  1216   Union_def ~> SUP_def, image_def
  1217   INT_eq ~> INF_def, and image_def
  1218   UN_eq ~> SUP_def, and image_def
  1219   INF_subset ~> INF_superset_mono [OF _ order_refl]
  1220 
  1221 More consistent and comprehensive names:
  1222 
  1223   INTER_eq_Inter_image ~> INF_def
  1224   UNION_eq_Union_image ~> SUP_def
  1225   INFI_def ~> INF_def
  1226   SUPR_def ~> SUP_def
  1227   INF_leI ~> INF_lower
  1228   INF_leI2 ~> INF_lower2
  1229   le_INFI ~> INF_greatest
  1230   le_SUPI ~> SUP_upper
  1231   le_SUPI2 ~> SUP_upper2
  1232   SUP_leI ~> SUP_least
  1233   INFI_bool_eq ~> INF_bool_eq
  1234   SUPR_bool_eq ~> SUP_bool_eq
  1235   INFI_apply ~> INF_apply
  1236   SUPR_apply ~> SUP_apply
  1237   INTER_def ~> INTER_eq
  1238   UNION_def ~> UNION_eq
  1239 
  1240 INCOMPATIBILITY.
  1241 
  1242 * Renamed theory Complete_Lattice to Complete_Lattices.
  1243 INCOMPATIBILITY.
  1244 
  1245 * Theory Complete_Lattices: lemmas Inf_eq_top_iff, INF_eq_top_iff,
  1246 INF_image, Inf_insert, INF_top, Inf_top_conv, INF_top_conv, SUP_bot,
  1247 Sup_bot_conv, SUP_bot_conv, Sup_eq_top_iff, SUP_eq_top_iff, SUP_image,
  1248 Sup_insert are now declared as [simp].  INCOMPATIBILITY.
  1249 
  1250 * Theory Lattice: lemmas compl_inf_bot, compl_le_comp_iff,
  1251 compl_sup_top, inf_idem, inf_left_idem, inf_sup_absorb, sup_idem,
  1252 sup_inf_absob, sup_left_idem are now declared as [simp].  Minor
  1253 INCOMPATIBILITY.
  1254 
  1255 * Added syntactic classes "inf" and "sup" for the respective
  1256 constants.  INCOMPATIBILITY: Changes in the argument order of the
  1257 (mostly internal) locale predicates for some derived classes.
  1258 
  1259 * Theorem collections ball_simps and bex_simps do not contain theorems
  1260 referring to UNION any longer; these have been moved to collection
  1261 UN_ball_bex_simps.  INCOMPATIBILITY.
  1262 
  1263 * Theory Archimedean_Field: floor now is defined as parameter of a
  1264 separate type class floor_ceiling.
  1265 
  1266 * Theory Finite_Set: more coherent development of fold_set locales:
  1267 
  1268     locale fun_left_comm ~> locale comp_fun_commute
  1269     locale fun_left_comm_idem ~> locale comp_fun_idem
  1270 
  1271 Both use point-free characterization; interpretation proofs may need
  1272 adjustment.  INCOMPATIBILITY.
  1273 
  1274 * Theory Limits: Type "'a net" has been renamed to "'a filter", in
  1275 accordance with standard mathematical terminology. INCOMPATIBILITY.
  1276 
  1277 * Theory Complex_Main: The locale interpretations for the
  1278 bounded_linear and bounded_bilinear locales have been removed, in
  1279 order to reduce the number of duplicate lemmas. Users must use the
  1280 original names for distributivity theorems, potential INCOMPATIBILITY.
  1281 
  1282   divide.add ~> add_divide_distrib
  1283   divide.diff ~> diff_divide_distrib
  1284   divide.setsum ~> setsum_divide_distrib
  1285   mult.add_right ~> right_distrib
  1286   mult.diff_right ~> right_diff_distrib
  1287   mult_right.setsum ~> setsum_right_distrib
  1288   mult_left.diff ~> left_diff_distrib
  1289 
  1290 * Theory Complex_Main: Several redundant theorems have been removed or
  1291 replaced by more general versions. INCOMPATIBILITY.
  1292 
  1293   real_diff_def ~> minus_real_def
  1294   real_divide_def ~> divide_real_def
  1295   real_less_def ~> less_le
  1296   real_abs_def ~> abs_real_def
  1297   real_sgn_def ~> sgn_real_def
  1298   real_mult_commute ~> mult_commute
  1299   real_mult_assoc ~> mult_assoc
  1300   real_mult_1 ~> mult_1_left
  1301   real_add_mult_distrib ~> left_distrib
  1302   real_zero_not_eq_one ~> zero_neq_one
  1303   real_mult_inverse_left ~> left_inverse
  1304   INVERSE_ZERO ~> inverse_zero
  1305   real_le_refl ~> order_refl
  1306   real_le_antisym ~> order_antisym
  1307   real_le_trans ~> order_trans
  1308   real_le_linear ~> linear
  1309   real_le_eq_diff ~> le_iff_diff_le_0
  1310   real_add_left_mono ~> add_left_mono
  1311   real_mult_order ~> mult_pos_pos
  1312   real_mult_less_mono2 ~> mult_strict_left_mono
  1313   real_of_int_real_of_nat ~> real_of_int_of_nat_eq
  1314   real_0_le_divide_iff ~> zero_le_divide_iff
  1315   realpow_two_disj ~> power2_eq_iff
  1316   real_squared_diff_one_factored ~> square_diff_one_factored
  1317   realpow_two_diff ~> square_diff_square_factored
  1318   reals_complete2 ~> complete_real
  1319   real_sum_squared_expand ~> power2_sum
  1320   exp_ln_eq ~> ln_unique
  1321   expi_add ~> exp_add
  1322   expi_zero ~> exp_zero
  1323   lemma_DERIV_subst ~> DERIV_cong
  1324   LIMSEQ_Zfun_iff ~> tendsto_Zfun_iff
  1325   LIMSEQ_const ~> tendsto_const
  1326   LIMSEQ_norm ~> tendsto_norm
  1327   LIMSEQ_add ~> tendsto_add
  1328   LIMSEQ_minus ~> tendsto_minus
  1329   LIMSEQ_minus_cancel ~> tendsto_minus_cancel
  1330   LIMSEQ_diff ~> tendsto_diff
  1331   bounded_linear.LIMSEQ ~> bounded_linear.tendsto
  1332   bounded_bilinear.LIMSEQ ~> bounded_bilinear.tendsto
  1333   LIMSEQ_mult ~> tendsto_mult
  1334   LIMSEQ_inverse ~> tendsto_inverse
  1335   LIMSEQ_divide ~> tendsto_divide
  1336   LIMSEQ_pow ~> tendsto_power
  1337   LIMSEQ_setsum ~> tendsto_setsum
  1338   LIMSEQ_setprod ~> tendsto_setprod
  1339   LIMSEQ_norm_zero ~> tendsto_norm_zero_iff
  1340   LIMSEQ_rabs_zero ~> tendsto_rabs_zero_iff
  1341   LIMSEQ_imp_rabs ~> tendsto_rabs
  1342   LIMSEQ_add_minus ~> tendsto_add [OF _ tendsto_minus]
  1343   LIMSEQ_add_const ~> tendsto_add [OF _ tendsto_const]
  1344   LIMSEQ_diff_const ~> tendsto_diff [OF _ tendsto_const]
  1345   LIMSEQ_Complex ~> tendsto_Complex
  1346   LIM_ident ~> tendsto_ident_at
  1347   LIM_const ~> tendsto_const
  1348   LIM_add ~> tendsto_add
  1349   LIM_add_zero ~> tendsto_add_zero
  1350   LIM_minus ~> tendsto_minus
  1351   LIM_diff ~> tendsto_diff
  1352   LIM_norm ~> tendsto_norm
  1353   LIM_norm_zero ~> tendsto_norm_zero
  1354   LIM_norm_zero_cancel ~> tendsto_norm_zero_cancel
  1355   LIM_norm_zero_iff ~> tendsto_norm_zero_iff
  1356   LIM_rabs ~> tendsto_rabs
  1357   LIM_rabs_zero ~> tendsto_rabs_zero
  1358   LIM_rabs_zero_cancel ~> tendsto_rabs_zero_cancel
  1359   LIM_rabs_zero_iff ~> tendsto_rabs_zero_iff
  1360   LIM_compose ~> tendsto_compose
  1361   LIM_mult ~> tendsto_mult
  1362   LIM_scaleR ~> tendsto_scaleR
  1363   LIM_of_real ~> tendsto_of_real
  1364   LIM_power ~> tendsto_power
  1365   LIM_inverse ~> tendsto_inverse
  1366   LIM_sgn ~> tendsto_sgn
  1367   isCont_LIM_compose ~> isCont_tendsto_compose
  1368   bounded_linear.LIM ~> bounded_linear.tendsto
  1369   bounded_linear.LIM_zero ~> bounded_linear.tendsto_zero
  1370   bounded_bilinear.LIM ~> bounded_bilinear.tendsto
  1371   bounded_bilinear.LIM_prod_zero ~> bounded_bilinear.tendsto_zero
  1372   bounded_bilinear.LIM_left_zero ~> bounded_bilinear.tendsto_left_zero
  1373   bounded_bilinear.LIM_right_zero ~> bounded_bilinear.tendsto_right_zero
  1374   LIM_inverse_fun ~> tendsto_inverse [OF tendsto_ident_at]
  1375 
  1376 * Theory Complex_Main: The definition of infinite series was
  1377 generalized.  Now it is defined on the type class {topological_space,
  1378 comm_monoid_add}.  Hence it is useable also for extended real numbers.
  1379 
  1380 * Theory Complex_Main: The complex exponential function "expi" is now
  1381 a type-constrained abbreviation for "exp :: complex => complex"; thus
  1382 several polymorphic lemmas about "exp" are now applicable to "expi".
  1383 
  1384 * Code generation:
  1385 
  1386   - Theory Library/Code_Char_ord provides native ordering of
  1387     characters in the target language.
  1388 
  1389   - Commands code_module and code_library are legacy, use export_code
  1390     instead.
  1391 
  1392   - Method "evaluation" is legacy, use method "eval" instead.
  1393 
  1394   - Legacy evaluator "SML" is deactivated by default.  May be
  1395     reactivated by the following theory command:
  1396 
  1397       setup {* Value.add_evaluator ("SML", Codegen.eval_term) *}
  1398 
  1399 * Declare ext [intro] by default.  Rare INCOMPATIBILITY.
  1400 
  1401 * New proof method "induction" that gives induction hypotheses the
  1402 name "IH", thus distinguishing them from further hypotheses that come
  1403 from rule induction.  The latter are still called "hyps".  Method
  1404 "induction" is a thin wrapper around "induct" and follows the same
  1405 syntax.
  1406 
  1407 * Method "fastsimp" has been renamed to "fastforce", but "fastsimp" is
  1408 still available as a legacy feature for some time.
  1409 
  1410 * Nitpick:
  1411   - Added "need" and "total_consts" options.
  1412   - Reintroduced "show_skolems" option by popular demand.
  1413   - Renamed attribute: nitpick_def ~> nitpick_unfold.
  1414     INCOMPATIBILITY.
  1415 
  1416 * Sledgehammer:
  1417   - Use quasi-sound (and efficient) translations by default.
  1418   - Added support for the following provers: E-ToFoF, LEO-II,
  1419     Satallax, SNARK, Waldmeister, and Z3 with TPTP syntax.
  1420   - Automatically preplay and minimize proofs before showing them if
  1421     this can be done within reasonable time.
  1422   - sledgehammer available_provers ~> sledgehammer supported_provers.
  1423     INCOMPATIBILITY.
  1424   - Added "preplay_timeout", "slicing", "type_enc", "sound",
  1425     "max_mono_iters", and "max_new_mono_instances" options.
  1426   - Removed "explicit_apply" and "full_types" options as well as "Full
  1427     Types" Proof General menu item. INCOMPATIBILITY.
  1428 
  1429 * Metis:
  1430   - Removed "metisF" -- use "metis" instead. INCOMPATIBILITY.
  1431   - Obsoleted "metisFT" -- use "metis (full_types)" instead.
  1432     INCOMPATIBILITY.
  1433 
  1434 * Command 'try':
  1435   - Renamed 'try_methods' and added "simp:", "intro:", "dest:", and
  1436     "elim:" options. INCOMPATIBILITY.
  1437   - Introduced 'try' that not only runs 'try_methods' but also
  1438     'solve_direct', 'sledgehammer', 'quickcheck', and 'nitpick'.
  1439 
  1440 * Quickcheck:
  1441   - Added "eval" option to evaluate terms for the found counterexample
  1442     (currently only supported by the default (exhaustive) tester).
  1443   - Added post-processing of terms to obtain readable counterexamples
  1444     (currently only supported by the default (exhaustive) tester).
  1445   - New counterexample generator quickcheck[narrowing] enables
  1446     narrowing-based testing.  Requires the Glasgow Haskell compiler
  1447     with its installation location defined in the Isabelle settings
  1448     environment as ISABELLE_GHC.
  1449   - Removed quickcheck tester "SML" based on the SML code generator
  1450     (formly in HOL/Library).
  1451 
  1452 * Function package: discontinued option "tailrec".  INCOMPATIBILITY,
  1453 use 'partial_function' instead.
  1454 
  1455 * Theory Library/Extended_Reals replaces now the positive extended
  1456 reals found in probability theory. This file is extended by
  1457 Multivariate_Analysis/Extended_Real_Limits.
  1458 
  1459 * Theory Library/Old_Recdef: old 'recdef' package has been moved here,
  1460 from where it must be imported explicitly if it is really required.
  1461 INCOMPATIBILITY.
  1462 
  1463 * Theory Library/Wfrec: well-founded recursion combinator "wfrec" has
  1464 been moved here.  INCOMPATIBILITY.
  1465 
  1466 * Theory Library/Saturated provides type of numbers with saturated
  1467 arithmetic.
  1468 
  1469 * Theory Library/Product_Lattice defines a pointwise ordering for the
  1470 product type 'a * 'b, and provides instance proofs for various order
  1471 and lattice type classes.
  1472 
  1473 * Theory Library/Countable now provides the "countable_datatype" proof
  1474 method for proving "countable" class instances for datatypes.
  1475 
  1476 * Theory Library/Cset_Monad allows do notation for computable sets
  1477 (cset) via the generic monad ad-hoc overloading facility.
  1478 
  1479 * Library: Theories of common data structures are split into theories
  1480 for implementation, an invariant-ensuring type, and connection to an
  1481 abstract type. INCOMPATIBILITY.
  1482 
  1483   - RBT is split into RBT and RBT_Mapping.
  1484   - AssocList is split and renamed into AList and AList_Mapping.
  1485   - DList is split into DList_Impl, DList, and DList_Cset.
  1486   - Cset is split into Cset and List_Cset.
  1487 
  1488 * Theory Library/Nat_Infinity has been renamed to
  1489 Library/Extended_Nat, with name changes of the following types and
  1490 constants:
  1491 
  1492   type inat   ~> type enat
  1493   Fin         ~> enat
  1494   Infty       ~> infinity (overloaded)
  1495   iSuc        ~> eSuc
  1496   the_Fin     ~> the_enat
  1497 
  1498 Every theorem name containing "inat", "Fin", "Infty", or "iSuc" has
  1499 been renamed accordingly. INCOMPATIBILITY.
  1500 
  1501 * Session Multivariate_Analysis: The euclidean_space type class now
  1502 fixes a constant "Basis :: 'a set" consisting of the standard
  1503 orthonormal basis for the type. Users now have the option of
  1504 quantifying over this set instead of using the "basis" function, e.g.
  1505 "ALL x:Basis. P x" vs "ALL i<DIM('a). P (basis i)".
  1506 
  1507 * Session Multivariate_Analysis: Type "('a, 'b) cart" has been renamed
  1508 to "('a, 'b) vec" (the syntax "'a ^ 'b" remains unaffected). Constants
  1509 "Cart_nth" and "Cart_lambda" have been respectively renamed to
  1510 "vec_nth" and "vec_lambda"; theorems mentioning those names have
  1511 changed to match. Definition theorems for overloaded constants now use
  1512 the standard "foo_vec_def" naming scheme. A few other theorems have
  1513 been renamed as follows (INCOMPATIBILITY):
  1514 
  1515   Cart_eq          ~> vec_eq_iff
  1516   dist_nth_le_cart ~> dist_vec_nth_le
  1517   tendsto_vector   ~> vec_tendstoI
  1518   Cauchy_vector    ~> vec_CauchyI
  1519 
  1520 * Session Multivariate_Analysis: Several duplicate theorems have been
  1521 removed, and other theorems have been renamed or replaced with more
  1522 general versions. INCOMPATIBILITY.
  1523 
  1524   finite_choice ~> finite_set_choice
  1525   eventually_conjI ~> eventually_conj
  1526   eventually_and ~> eventually_conj_iff
  1527   eventually_false ~> eventually_False
  1528   setsum_norm ~> norm_setsum
  1529   Lim_sequentially ~> LIMSEQ_def
  1530   Lim_ident_at ~> LIM_ident
  1531   Lim_const ~> tendsto_const
  1532   Lim_cmul ~> tendsto_scaleR [OF tendsto_const]
  1533   Lim_neg ~> tendsto_minus
  1534   Lim_add ~> tendsto_add
  1535   Lim_sub ~> tendsto_diff
  1536   Lim_mul ~> tendsto_scaleR
  1537   Lim_vmul ~> tendsto_scaleR [OF _ tendsto_const]
  1538   Lim_null_norm ~> tendsto_norm_zero_iff [symmetric]
  1539   Lim_linear ~> bounded_linear.tendsto
  1540   Lim_component ~> tendsto_euclidean_component
  1541   Lim_component_cart ~> tendsto_vec_nth
  1542   Lim_inner ~> tendsto_inner [OF tendsto_const]
  1543   dot_lsum ~> inner_setsum_left
  1544   dot_rsum ~> inner_setsum_right
  1545   continuous_cmul ~> continuous_scaleR [OF continuous_const]
  1546   continuous_neg ~> continuous_minus
  1547   continuous_sub ~> continuous_diff
  1548   continuous_vmul ~> continuous_scaleR [OF _ continuous_const]
  1549   continuous_mul ~> continuous_scaleR
  1550   continuous_inv ~> continuous_inverse
  1551   continuous_at_within_inv ~> continuous_at_within_inverse
  1552   continuous_at_inv ~> continuous_at_inverse
  1553   continuous_at_norm ~> continuous_norm [OF continuous_at_id]
  1554   continuous_at_infnorm ~> continuous_infnorm [OF continuous_at_id]
  1555   continuous_at_component ~> continuous_component [OF continuous_at_id]
  1556   continuous_on_neg ~> continuous_on_minus
  1557   continuous_on_sub ~> continuous_on_diff
  1558   continuous_on_cmul ~> continuous_on_scaleR [OF continuous_on_const]
  1559   continuous_on_vmul ~> continuous_on_scaleR [OF _ continuous_on_const]
  1560   continuous_on_mul ~> continuous_on_scaleR
  1561   continuous_on_mul_real ~> continuous_on_mult
  1562   continuous_on_inner ~> continuous_on_inner [OF continuous_on_const]
  1563   continuous_on_norm ~> continuous_on_norm [OF continuous_on_id]
  1564   continuous_on_inverse ~> continuous_on_inv
  1565   uniformly_continuous_on_neg ~> uniformly_continuous_on_minus
  1566   uniformly_continuous_on_sub ~> uniformly_continuous_on_diff
  1567   subset_interior ~> interior_mono
  1568   subset_closure ~> closure_mono
  1569   closure_univ ~> closure_UNIV
  1570   real_arch_lt ~> reals_Archimedean2
  1571   real_arch ~> reals_Archimedean3
  1572   real_abs_norm ~> abs_norm_cancel
  1573   real_abs_sub_norm ~> norm_triangle_ineq3
  1574   norm_cauchy_schwarz_abs ~> Cauchy_Schwarz_ineq2
  1575 
  1576 * Session HOL-Probability:
  1577   - Caratheodory's extension lemma is now proved for ring_of_sets.
  1578   - Infinite products of probability measures are now available.
  1579   - Sigma closure is independent, if the generator is independent
  1580   - Use extended reals instead of positive extended
  1581     reals. INCOMPATIBILITY.
  1582 
  1583 * Session HOLCF: Discontinued legacy theorem names, INCOMPATIBILITY.
  1584 
  1585   expand_fun_below ~> fun_below_iff
  1586   below_fun_ext ~> fun_belowI
  1587   expand_cfun_eq ~> cfun_eq_iff
  1588   ext_cfun ~> cfun_eqI
  1589   expand_cfun_below ~> cfun_below_iff
  1590   below_cfun_ext ~> cfun_belowI
  1591   monofun_fun_fun ~> fun_belowD
  1592   monofun_fun_arg ~> monofunE
  1593   monofun_lub_fun ~> adm_monofun [THEN admD]
  1594   cont_lub_fun ~> adm_cont [THEN admD]
  1595   cont2cont_Rep_CFun ~> cont2cont_APP
  1596   cont_Rep_CFun_app ~> cont_APP_app
  1597   cont_Rep_CFun_app_app ~> cont_APP_app_app
  1598   cont_cfun_fun ~> cont_Rep_cfun1 [THEN contE]
  1599   cont_cfun_arg ~> cont_Rep_cfun2 [THEN contE]
  1600   contlub_cfun ~> lub_APP [symmetric]
  1601   contlub_LAM ~> lub_LAM [symmetric]
  1602   thelubI ~> lub_eqI
  1603   UU_I ~> bottomI
  1604   lift_distinct1 ~> lift.distinct(1)
  1605   lift_distinct2 ~> lift.distinct(2)
  1606   Def_not_UU ~> lift.distinct(2)
  1607   Def_inject ~> lift.inject
  1608   below_UU_iff ~> below_bottom_iff
  1609   eq_UU_iff ~> eq_bottom_iff
  1610 
  1611 
  1612 *** Document preparation ***
  1613 
  1614 * Antiquotation @{rail} layouts railroad syntax diagrams, see also
  1615 isar-ref manual, both for description and actual application of the
  1616 same.
  1617 
  1618 * Antiquotation @{value} evaluates the given term and presents its
  1619 result.
  1620 
  1621 * Antiquotations: term style "isub" provides ad-hoc conversion of
  1622 variables x1, y23 into subscripted form x\<^isub>1,
  1623 y\<^isub>2\<^isub>3.
  1624 
  1625 * Predefined LaTeX macros for Isabelle symbols \<bind> and \<then>
  1626 (e.g. see ~~/src/HOL/Library/Monad_Syntax.thy).
  1627 
  1628 * Localized \isabellestyle switch can be used within blocks or groups
  1629 like this:
  1630 
  1631   \isabellestyle{it}  %preferred default
  1632   {\isabellestylett @{text "typewriter stuff"}}
  1633 
  1634 * Discontinued special treatment of hard tabulators.  Implicit
  1635 tab-width is now defined as 1.  Potential INCOMPATIBILITY for visual
  1636 layouts.
  1637 
  1638 
  1639 *** ML ***
  1640 
  1641 * The inner syntax of sort/type/term/prop supports inlined YXML
  1642 representations within quoted string tokens.  By encoding logical
  1643 entities via Term_XML (in ML or Scala) concrete syntax can be
  1644 bypassed, which is particularly useful for producing bits of text
  1645 under external program control.
  1646 
  1647 * Antiquotations for ML and document preparation are managed as theory
  1648 data, which requires explicit setup.
  1649 
  1650 * Isabelle_Process.is_active allows tools to check if the official
  1651 process wrapper is running (Isabelle/Scala/jEdit) or the old TTY loop
  1652 (better known as Proof General).
  1653 
  1654 * Structure Proof_Context follows standard naming scheme.  Old
  1655 ProofContext is still available for some time as legacy alias.
  1656 
  1657 * Structure Timing provides various operations for timing; supersedes
  1658 former start_timing/end_timing etc.
  1659 
  1660 * Path.print is the official way to show file-system paths to users
  1661 (including quotes etc.).
  1662 
  1663 * Inner syntax: identifiers in parse trees of generic categories
  1664 "logic", "aprop", "idt" etc. carry position information (disguised as
  1665 type constraints).  Occasional INCOMPATIBILITY with non-compliant
  1666 translations that choke on unexpected type constraints.  Positions can
  1667 be stripped in ML translations via Syntax.strip_positions /
  1668 Syntax.strip_positions_ast, or via the syntax constant
  1669 "_strip_positions" within parse trees.  As last resort, positions can
  1670 be disabled via the configuration option Syntax.positions, which is
  1671 called "syntax_positions" in Isar attribute syntax.
  1672 
  1673 * Discontinued special status of various ML structures that contribute
  1674 to structure Syntax (Ast, Lexicon, Mixfix, Parser, Printer etc.): less
  1675 pervasive content, no inclusion in structure Syntax.  INCOMPATIBILITY,
  1676 refer directly to Ast.Constant, Lexicon.is_identifier,
  1677 Syntax_Trans.mk_binder_tr etc.
  1678 
  1679 * Typed print translation: discontinued show_sorts argument, which is
  1680 already available via context of "advanced" translation.
  1681 
  1682 * Refined PARALLEL_GOALS tactical: degrades gracefully for schematic
  1683 goal states; body tactic needs to address all subgoals uniformly.
  1684 
  1685 * Slightly more special eq_list/eq_set, with shortcut involving
  1686 pointer equality (assumes that eq relation is reflexive).
  1687 
  1688 * Classical tactics use proper Proof.context instead of historic types
  1689 claset/clasimpset.  Old-style declarations like addIs, addEs, addDs
  1690 operate directly on Proof.context.  Raw type claset retains its use as
  1691 snapshot of the classical context, which can be recovered via
  1692 (put_claset HOL_cs) etc.  Type clasimpset has been discontinued.
  1693 INCOMPATIBILITY, classical tactics and derived proof methods require
  1694 proper Proof.context.
  1695 
  1696 
  1697 *** System ***
  1698 
  1699 * Discontinued support for Poly/ML 5.2, which was the last version
  1700 without proper multithreading and TimeLimit implementation.
  1701 
  1702 * Discontinued old lib/scripts/polyml-platform, which has been
  1703 obsolete since Isabelle2009-2.
  1704 
  1705 * Various optional external tools are referenced more robustly and
  1706 uniformly by explicit Isabelle settings as follows:
  1707 
  1708   ISABELLE_CSDP   (formerly CSDP_EXE)
  1709   ISABELLE_GHC    (formerly EXEC_GHC or GHC_PATH)
  1710   ISABELLE_OCAML  (formerly EXEC_OCAML)
  1711   ISABELLE_SWIPL  (formerly EXEC_SWIPL)
  1712   ISABELLE_YAP    (formerly EXEC_YAP)
  1713 
  1714 Note that automated detection from the file-system or search path has
  1715 been discontinued.  INCOMPATIBILITY.
  1716 
  1717 * Scala layer provides JVM method invocation service for static
  1718 methods of type (String)String, see Invoke_Scala.method in ML.  For
  1719 example:
  1720 
  1721   Invoke_Scala.method "java.lang.System.getProperty" "java.home"
  1722 
  1723 Together with YXML.string_of_body/parse_body and XML.Encode/Decode
  1724 this allows to pass structured values between ML and Scala.
  1725 
  1726 * The IsabelleText fonts includes some further glyphs to support the
  1727 Prover IDE.  Potential INCOMPATIBILITY: users who happen to have
  1728 installed a local copy (which is normally *not* required) need to
  1729 delete or update it from ~~/lib/fonts/.
  1730 
  1731 
  1732 
  1733 New in Isabelle2011 (January 2011)
  1734 ----------------------------------
  1735 
  1736 *** General ***
  1737 
  1738 * Experimental Prover IDE based on Isabelle/Scala and jEdit (see
  1739 src/Tools/jEdit).  This also serves as IDE for Isabelle/ML, with
  1740 useful tooltips and hyperlinks produced from its static analysis.  The
  1741 bundled component provides an executable Isabelle tool that can be run
  1742 like this:
  1743 
  1744   Isabelle2011/bin/isabelle jedit
  1745 
  1746 * Significantly improved Isabelle/Isar implementation manual.
  1747 
  1748 * System settings: ISABELLE_HOME_USER now includes ISABELLE_IDENTIFIER
  1749 (and thus refers to something like $HOME/.isabelle/Isabelle2011),
  1750 while the default heap location within that directory lacks that extra
  1751 suffix.  This isolates multiple Isabelle installations from each
  1752 other, avoiding problems with old settings in new versions.
  1753 INCOMPATIBILITY, need to copy/upgrade old user settings manually.
  1754 
  1755 * Source files are always encoded as UTF-8, instead of old-fashioned
  1756 ISO-Latin-1.  INCOMPATIBILITY.  Isabelle LaTeX documents might require
  1757 the following package declarations:
  1758 
  1759   \usepackage[utf8]{inputenc}
  1760   \usepackage{textcomp}
  1761 
  1762 * Explicit treatment of UTF-8 sequences as Isabelle symbols, such that
  1763 a Unicode character is treated as a single symbol, not a sequence of
  1764 non-ASCII bytes as before.  Since Isabelle/ML string literals may
  1765 contain symbols without further backslash escapes, Unicode can now be
  1766 used here as well.  Recall that Symbol.explode in ML provides a
  1767 consistent view on symbols, while raw explode (or String.explode)
  1768 merely give a byte-oriented representation.
  1769 
  1770 * Theory loader: source files are primarily located via the master
  1771 directory of each theory node (where the .thy file itself resides).
  1772 The global load path is still partially available as legacy feature.
  1773 Minor INCOMPATIBILITY due to subtle change in file lookup: use
  1774 explicit paths, relatively to the theory.
  1775 
  1776 * Special treatment of ML file names has been discontinued.
  1777 Historically, optional extensions .ML or .sml were added on demand --
  1778 at the cost of clarity of file dependencies.  Recall that Isabelle/ML
  1779 files exclusively use the .ML extension.  Minor INCOMPATIBILTY.
  1780 
  1781 * Various options that affect pretty printing etc. are now properly
  1782 handled within the context via configuration options, instead of
  1783 unsynchronized references or print modes.  There are both ML Config.T
  1784 entities and Isar declaration attributes to access these.
  1785 
  1786   ML (Config.T)                 Isar (attribute)
  1787 
  1788   eta_contract                  eta_contract
  1789   show_brackets                 show_brackets
  1790   show_sorts                    show_sorts
  1791   show_types                    show_types
  1792   show_question_marks           show_question_marks
  1793   show_consts                   show_consts
  1794   show_abbrevs                  show_abbrevs
  1795 
  1796   Syntax.ast_trace              syntax_ast_trace
  1797   Syntax.ast_stat               syntax_ast_stat
  1798   Syntax.ambiguity_level        syntax_ambiguity_level
  1799 
  1800   Goal_Display.goals_limit      goals_limit
  1801   Goal_Display.show_main_goal   show_main_goal
  1802 
  1803   Method.rule_trace             rule_trace
  1804 
  1805   Thy_Output.display            thy_output_display
  1806   Thy_Output.quotes             thy_output_quotes
  1807   Thy_Output.indent             thy_output_indent
  1808   Thy_Output.source             thy_output_source
  1809   Thy_Output.break              thy_output_break
  1810 
  1811 Note that corresponding "..._default" references in ML may only be
  1812 changed globally at the ROOT session setup, but *not* within a theory.
  1813 The option "show_abbrevs" supersedes the former print mode
  1814 "no_abbrevs" with inverted meaning.
  1815 
  1816 * More systematic naming of some configuration options.
  1817 INCOMPATIBILITY.
  1818 
  1819   trace_simp  ~>  simp_trace
  1820   debug_simp  ~>  simp_debug
  1821 
  1822 * Support for real valued configuration options, using simplistic
  1823 floating-point notation that coincides with the inner syntax for
  1824 float_token.
  1825 
  1826 * Support for real valued preferences (with approximative PGIP type):
  1827 front-ends need to accept "pgint" values in float notation.
  1828 INCOMPATIBILITY.
  1829 
  1830 * The IsabelleText font now includes Cyrillic, Hebrew, Arabic from
  1831 DejaVu Sans.
  1832 
  1833 * Discontinued support for Poly/ML 5.0 and 5.1 versions.
  1834 
  1835 
  1836 *** Pure ***
  1837 
  1838 * Command 'type_synonym' (with single argument) replaces somewhat
  1839 outdated 'types', which is still available as legacy feature for some
  1840 time.
  1841 
  1842 * Command 'nonterminal' (with 'and' separated list of arguments)
  1843 replaces somewhat outdated 'nonterminals'.  INCOMPATIBILITY.
  1844 
  1845 * Command 'notepad' replaces former 'example_proof' for
  1846 experimentation in Isar without any result.  INCOMPATIBILITY.
  1847 
  1848 * Locale interpretation commands 'interpret' and 'sublocale' accept
  1849 lists of equations to map definitions in a locale to appropriate
  1850 entities in the context of the interpretation.  The 'interpretation'
  1851 command already provided this functionality.
  1852 
  1853 * Diagnostic command 'print_dependencies' prints the locale instances
  1854 that would be activated if the specified expression was interpreted in
  1855 the current context.  Variant "print_dependencies!" assumes a context
  1856 without interpretations.
  1857 
  1858 * Diagnostic command 'print_interps' prints interpretations in proofs
  1859 in addition to interpretations in theories.
  1860 
  1861 * Discontinued obsolete 'global' and 'local' commands to manipulate
  1862 the theory name space.  Rare INCOMPATIBILITY.  The ML functions
  1863 Sign.root_path and Sign.local_path may be applied directly where this
  1864 feature is still required for historical reasons.
  1865 
  1866 * Discontinued obsolete 'constdefs' command.  INCOMPATIBILITY, use
  1867 'definition' instead.
  1868 
  1869 * The "prems" fact, which refers to the accidental collection of
  1870 foundational premises in the context, is now explicitly marked as
  1871 legacy feature and will be discontinued soon.  Consider using "assms"
  1872 of the head statement or reference facts by explicit names.
  1873 
  1874 * Document antiquotations @{class} and @{type} print classes and type
  1875 constructors.
  1876 
  1877 * Document antiquotation @{file} checks file/directory entries within
  1878 the local file system.
  1879 
  1880 
  1881 *** HOL ***
  1882 
  1883 * Coercive subtyping: functions can be declared as coercions and type
  1884 inference will add them as necessary upon input of a term.  Theory
  1885 Complex_Main declares real :: nat => real and real :: int => real as
  1886 coercions. A coercion function f is declared like this:
  1887 
  1888   declare [[coercion f]]
  1889 
  1890 To lift coercions through type constructors (e.g. from nat => real to
  1891 nat list => real list), map functions can be declared, e.g.
  1892 
  1893   declare [[coercion_map map]]
  1894 
  1895 Currently coercion inference is activated only in theories including
  1896 real numbers, i.e. descendants of Complex_Main.  This is controlled by
  1897 the configuration option "coercion_enabled", e.g. it can be enabled in
  1898 other theories like this:
  1899 
  1900   declare [[coercion_enabled]]
  1901 
  1902 * Command 'partial_function' provides basic support for recursive
  1903 function definitions over complete partial orders.  Concrete instances
  1904 are provided for i) the option type, ii) tail recursion on arbitrary
  1905 types, and iii) the heap monad of Imperative_HOL.  See
  1906 src/HOL/ex/Fundefs.thy and src/HOL/Imperative_HOL/ex/Linked_Lists.thy
  1907 for examples.
  1908 
  1909 * Function package: f.psimps rules are no longer implicitly declared
  1910 as [simp].  INCOMPATIBILITY.
  1911 
  1912 * Datatype package: theorems generated for executable equality (class
  1913 "eq") carry proper names and are treated as default code equations.
  1914 
  1915 * Inductive package: now offers command 'inductive_simps' to
  1916 automatically derive instantiated and simplified equations for
  1917 inductive predicates, similar to 'inductive_cases'.
  1918 
  1919 * Command 'enriched_type' allows to register properties of the
  1920 functorial structure of types.
  1921 
  1922 * Improved infrastructure for term evaluation using code generator
  1923 techniques, in particular static evaluation conversions.
  1924 
  1925 * Code generator: Scala (2.8 or higher) has been added to the target
  1926 languages.
  1927 
  1928 * Code generator: globbing constant expressions "*" and "Theory.*"
  1929 have been replaced by the more idiomatic "_" and "Theory._".
  1930 INCOMPATIBILITY.
  1931 
  1932 * Code generator: export_code without explicit file declaration prints
  1933 to standard output.  INCOMPATIBILITY.
  1934 
  1935 * Code generator: do not print function definitions for case
  1936 combinators any longer.
  1937 
  1938 * Code generator: simplification with rules determined with
  1939 src/Tools/Code/code_simp.ML and method "code_simp".
  1940 
  1941 * Code generator for records: more idiomatic representation of record
  1942 types.  Warning: records are not covered by ancient SML code
  1943 generation any longer.  INCOMPATIBILITY.  In cases of need, a suitable
  1944 rep_datatype declaration helps to succeed then:
  1945 
  1946   record 'a foo = ...
  1947   ...
  1948   rep_datatype foo_ext ...
  1949 
  1950 * Records: logical foundation type for records does not carry a
  1951 '_type' suffix any longer (obsolete due to authentic syntax).
  1952 INCOMPATIBILITY.
  1953 
  1954 * Quickcheck now by default uses exhaustive testing instead of random
  1955 testing.  Random testing can be invoked by "quickcheck [random]",
  1956 exhaustive testing by "quickcheck [exhaustive]".
  1957 
  1958 * Quickcheck instantiates polymorphic types with small finite
  1959 datatypes by default. This enables a simple execution mechanism to
  1960 handle quantifiers and function equality over the finite datatypes.
  1961 
  1962 * Quickcheck random generator has been renamed from "code" to
  1963 "random".  INCOMPATIBILITY.
  1964 
  1965 * Quickcheck now has a configurable time limit which is set to 30
  1966 seconds by default. This can be changed by adding [timeout = n] to the
  1967 quickcheck command. The time limit for Auto Quickcheck is still set
  1968 independently.
  1969 
  1970 * Quickcheck in locales considers interpretations of that locale for
  1971 counter example search.
  1972 
  1973 * Sledgehammer:
  1974   - Added "smt" and "remote_smt" provers based on the "smt" proof
  1975     method. See the Sledgehammer manual for details ("isabelle doc
  1976     sledgehammer").
  1977   - Renamed commands:
  1978     sledgehammer atp_info ~> sledgehammer running_provers
  1979     sledgehammer atp_kill ~> sledgehammer kill_provers
  1980     sledgehammer available_atps ~> sledgehammer available_provers
  1981     INCOMPATIBILITY.
  1982   - Renamed options:
  1983     sledgehammer [atps = ...] ~> sledgehammer [provers = ...]
  1984     sledgehammer [atp = ...] ~> sledgehammer [prover = ...]
  1985     sledgehammer [timeout = 77 s] ~> sledgehammer [timeout = 77]
  1986     (and "ms" and "min" are no longer supported)
  1987     INCOMPATIBILITY.
  1988 
  1989 * Nitpick:
  1990   - Renamed options:
  1991     nitpick [timeout = 77 s] ~> nitpick [timeout = 77]
  1992     nitpick [tac_timeout = 777 ms] ~> nitpick [tac_timeout = 0.777]
  1993     INCOMPATIBILITY.
  1994   - Added support for partial quotient types.
  1995   - Added local versions of the "Nitpick.register_xxx" functions.
  1996   - Added "whack" option.
  1997   - Allow registration of quotient types as codatatypes.
  1998   - Improved "merge_type_vars" option to merge more types.
  1999   - Removed unsound "fast_descrs" option.
  2000   - Added custom symmetry breaking for datatypes, making it possible to reach
  2001     higher cardinalities.
  2002   - Prevent the expansion of too large definitions.
  2003 
  2004 * Proof methods "metis" and "meson" now have configuration options
  2005 "meson_trace", "metis_trace", and "metis_verbose" that can be enabled
  2006 to diagnose these tools. E.g.
  2007 
  2008     using [[metis_trace = true]]
  2009 
  2010 * Auto Solve: Renamed "Auto Solve Direct".  The tool is now available
  2011 manually as command 'solve_direct'.
  2012 
  2013 * The default SMT solver Z3 must be enabled explicitly (due to
  2014 licensing issues) by setting the environment variable
  2015 Z3_NON_COMMERCIAL in etc/settings of the component, for example.  For
  2016 commercial applications, the SMT solver CVC3 is provided as fall-back;
  2017 changing the SMT solver is done via the configuration option
  2018 "smt_solver".
  2019 
  2020 * Remote SMT solvers need to be referred to by the "remote_" prefix,
  2021 i.e. "remote_cvc3" and "remote_z3".
  2022 
  2023 * Added basic SMT support for datatypes, records, and typedefs using
  2024 the oracle mode (no proofs).  Direct support of pairs has been dropped
  2025 in exchange (pass theorems fst_conv snd_conv pair_collapse to the SMT
  2026 support for a similar behavior).  Minor INCOMPATIBILITY.
  2027 
  2028 * Changed SMT configuration options:
  2029   - Renamed:
  2030     z3_proofs ~> smt_oracle (with inverted meaning)
  2031     z3_trace_assms ~> smt_trace_used_facts
  2032     INCOMPATIBILITY.
  2033   - Added:
  2034     smt_verbose
  2035     smt_random_seed
  2036     smt_datatypes
  2037     smt_infer_triggers
  2038     smt_monomorph_limit
  2039     cvc3_options
  2040     remote_cvc3_options
  2041     remote_z3_options
  2042     yices_options
  2043 
  2044 * Boogie output files (.b2i files) need to be declared in the theory
  2045 header.
  2046 
  2047 * Simplification procedure "list_to_set_comprehension" rewrites list
  2048 comprehensions applied to List.set to set comprehensions.  Occasional
  2049 INCOMPATIBILITY, may be deactivated like this:
  2050 
  2051   declare [[simproc del: list_to_set_comprehension]]
  2052 
  2053 * Removed old version of primrec package.  INCOMPATIBILITY.
  2054 
  2055 * Removed simplifier congruence rule of "prod_case", as has for long
  2056 been the case with "split".  INCOMPATIBILITY.
  2057 
  2058 * String.literal is a type, but not a datatype.  INCOMPATIBILITY.
  2059 
  2060 * Removed [split_format ... and ... and ...] version of
  2061 [split_format].  Potential INCOMPATIBILITY.
  2062 
  2063 * Predicate "sorted" now defined inductively, with nice induction
  2064 rules.  INCOMPATIBILITY: former sorted.simps now named sorted_simps.
  2065 
  2066 * Constant "contents" renamed to "the_elem", to free the generic name
  2067 contents for other uses.  INCOMPATIBILITY.
  2068 
  2069 * Renamed class eq and constant eq (for code generation) to class
  2070 equal and constant equal, plus renaming of related facts and various
  2071 tuning.  INCOMPATIBILITY.
  2072 
  2073 * Dropped type classes mult_mono and mult_mono1.  INCOMPATIBILITY.
  2074 
  2075 * Removed output syntax "'a ~=> 'b" for "'a => 'b option".
  2076 INCOMPATIBILITY.
  2077 
  2078 * Renamed theory Fset to Cset, type Fset.fset to Cset.set, in order to
  2079 avoid confusion with finite sets.  INCOMPATIBILITY.
  2080 
  2081 * Abandoned locales equiv, congruent and congruent2 for equivalence
  2082 relations.  INCOMPATIBILITY: use equivI rather than equiv_intro (same
  2083 for congruent(2)).
  2084 
  2085 * Some previously unqualified names have been qualified:
  2086 
  2087   types
  2088     bool ~> HOL.bool
  2089     nat ~> Nat.nat
  2090 
  2091   constants
  2092     Trueprop ~> HOL.Trueprop
  2093     True ~> HOL.True
  2094     False ~> HOL.False
  2095     op & ~> HOL.conj
  2096     op | ~> HOL.disj
  2097     op --> ~> HOL.implies
  2098     op = ~> HOL.eq
  2099     Not ~> HOL.Not
  2100     The ~> HOL.The
  2101     All ~> HOL.All
  2102     Ex ~> HOL.Ex
  2103     Ex1 ~> HOL.Ex1
  2104     Let ~> HOL.Let
  2105     If ~> HOL.If
  2106     Ball ~> Set.Ball
  2107     Bex ~> Set.Bex
  2108     Suc ~> Nat.Suc
  2109     Pair ~> Product_Type.Pair
  2110     fst ~> Product_Type.fst
  2111     snd ~> Product_Type.snd
  2112     curry ~> Product_Type.curry
  2113     op : ~> Set.member
  2114     Collect ~> Set.Collect
  2115 
  2116 INCOMPATIBILITY.
  2117 
  2118 * More canonical naming convention for some fundamental definitions:
  2119 
  2120     bot_bool_eq ~> bot_bool_def
  2121     top_bool_eq ~> top_bool_def
  2122     inf_bool_eq ~> inf_bool_def
  2123     sup_bool_eq ~> sup_bool_def
  2124     bot_fun_eq  ~> bot_fun_def
  2125     top_fun_eq  ~> top_fun_def
  2126     inf_fun_eq  ~> inf_fun_def
  2127     sup_fun_eq  ~> sup_fun_def
  2128 
  2129 INCOMPATIBILITY.
  2130 
  2131 * More stylized fact names:
  2132 
  2133   expand_fun_eq ~> fun_eq_iff
  2134   expand_set_eq ~> set_eq_iff
  2135   set_ext       ~> set_eqI
  2136   nat_number    ~> eval_nat_numeral
  2137 
  2138 INCOMPATIBILITY.
  2139 
  2140 * Refactoring of code-generation specific operations in theory List:
  2141 
  2142   constants
  2143     null ~> List.null
  2144 
  2145   facts
  2146     mem_iff ~> member_def
  2147     null_empty ~> null_def
  2148 
  2149 INCOMPATIBILITY.  Note that these were not supposed to be used
  2150 regularly unless for striking reasons; their main purpose was code
  2151 generation.
  2152 
  2153 Various operations from the Haskell prelude are used for generating
  2154 Haskell code.
  2155 
  2156 * Term "bij f" is now an abbreviation of "bij_betw f UNIV UNIV".  Term
  2157 "surj f" is now an abbreviation of "range f = UNIV".  The theorems
  2158 bij_def and surj_def are unchanged.  INCOMPATIBILITY.
  2159 
  2160 * Abolished some non-alphabetic type names: "prod" and "sum" replace
  2161 "*" and "+" respectively.  INCOMPATIBILITY.
  2162 
  2163 * Name "Plus" of disjoint sum operator "<+>" is now hidden.  Write
  2164 "Sum_Type.Plus" instead.
  2165 
  2166 * Constant "split" has been merged with constant "prod_case"; names of
  2167 ML functions, facts etc. involving split have been retained so far,
  2168 though.  INCOMPATIBILITY.
  2169 
  2170 * Dropped old infix syntax "_ mem _" for List.member; use "_ : set _"
  2171 instead.  INCOMPATIBILITY.
  2172 
  2173 * Removed lemma "Option.is_none_none" which duplicates "is_none_def".
  2174 INCOMPATIBILITY.
  2175 
  2176 * Former theory Library/Enum is now part of the HOL-Main image.
  2177 INCOMPATIBILITY: all constants of the Enum theory now have to be
  2178 referred to by its qualified name.
  2179 
  2180   enum    ~>  Enum.enum
  2181   nlists  ~>  Enum.nlists
  2182   product ~>  Enum.product
  2183 
  2184 * Theory Library/Monad_Syntax provides do-syntax for monad types.
  2185 Syntax in Library/State_Monad has been changed to avoid ambiguities.
  2186 INCOMPATIBILITY.
  2187 
  2188 * Theory Library/SetsAndFunctions has been split into
  2189 Library/Function_Algebras and Library/Set_Algebras; canonical names
  2190 for instance definitions for functions; various improvements.
  2191 INCOMPATIBILITY.
  2192 
  2193 * Theory Library/Multiset provides stable quicksort implementation of
  2194 sort_key.
  2195 
  2196 * Theory Library/Multiset: renamed empty_idemp ~> empty_neutral.
  2197 INCOMPATIBILITY.
  2198 
  2199 * Session Multivariate_Analysis: introduced a type class for euclidean
  2200 space.  Most theorems are now stated in terms of euclidean spaces
  2201 instead of finite cartesian products.
  2202 
  2203   types
  2204     real ^ 'n ~>  'a::real_vector
  2205               ~>  'a::euclidean_space
  2206               ~>  'a::ordered_euclidean_space
  2207         (depends on your needs)
  2208 
  2209   constants
  2210      _ $ _        ~> _ $$ _
  2211      \<chi> x. _  ~> \<chi>\<chi> x. _
  2212      CARD('n)     ~> DIM('a)
  2213 
  2214 Also note that the indices are now natural numbers and not from some
  2215 finite type. Finite cartesian products of euclidean spaces, products
  2216 of euclidean spaces the real and complex numbers are instantiated to
  2217 be euclidean_spaces.  INCOMPATIBILITY.
  2218 
  2219 * Session Probability: introduced pextreal as positive extended real
  2220 numbers.  Use pextreal as value for measures.  Introduce the
  2221 Radon-Nikodym derivative, product spaces and Fubini's theorem for
  2222 arbitrary sigma finite measures.  Introduces Lebesgue measure based on
  2223 the integral in Multivariate Analysis.  INCOMPATIBILITY.
  2224 
  2225 * Session Imperative_HOL: revamped, corrected dozens of inadequacies.
  2226 INCOMPATIBILITY.
  2227 
  2228 * Session SPARK (with image HOL-SPARK) provides commands to load and
  2229 prove verification conditions generated by the SPARK Ada program
  2230 verifier.  See also src/HOL/SPARK and src/HOL/SPARK/Examples.
  2231 
  2232 
  2233 *** HOL-Algebra ***
  2234 
  2235 * Theorems for additive ring operations (locale abelian_monoid and
  2236 descendants) are generated by interpretation from their multiplicative
  2237 counterparts.  Names (in particular theorem names) have the mandatory
  2238 qualifier 'add'.  Previous theorem names are redeclared for
  2239 compatibility.
  2240 
  2241 * Structure "int_ring" is now an abbreviation (previously a
  2242 definition).  This fits more natural with advanced interpretations.
  2243 
  2244 
  2245 *** HOLCF ***
  2246 
  2247 * The domain package now runs in definitional mode by default: The
  2248 former command 'new_domain' is now called 'domain'.  To use the domain
  2249 package in its original axiomatic mode, use 'domain (unsafe)'.
  2250 INCOMPATIBILITY.
  2251 
  2252 * The new class "domain" is now the default sort.  Class "predomain"
  2253 is an unpointed version of "domain". Theories can be updated by
  2254 replacing sort annotations as shown below.  INCOMPATIBILITY.
  2255 
  2256   'a::type ~> 'a::countable
  2257   'a::cpo  ~> 'a::predomain
  2258   'a::pcpo ~> 'a::domain
  2259 
  2260 * The old type class "rep" has been superseded by class "domain".
  2261 Accordingly, users of the definitional package must remove any
  2262 "default_sort rep" declarations.  INCOMPATIBILITY.
  2263 
  2264 * The domain package (definitional mode) now supports unpointed
  2265 predomain argument types, as long as they are marked 'lazy'. (Strict
  2266 arguments must be in class "domain".) For example, the following
  2267 domain definition now works:
  2268 
  2269   domain natlist = nil | cons (lazy "nat discr") (lazy "natlist")
  2270 
  2271 * Theory HOLCF/Library/HOL_Cpo provides cpo and predomain class
  2272 instances for types from main HOL: bool, nat, int, char, 'a + 'b,
  2273 'a option, and 'a list.  Additionally, it configures fixrec and the
  2274 domain package to work with these types.  For example:
  2275 
  2276   fixrec isInl :: "('a + 'b) u -> tr"
  2277     where "isInl$(up$(Inl x)) = TT" | "isInl$(up$(Inr y)) = FF"
  2278 
  2279   domain V = VFun (lazy "V -> V") | VCon (lazy "nat") (lazy "V list")
  2280 
  2281 * The "(permissive)" option of fixrec has been replaced with a
  2282 per-equation "(unchecked)" option. See
  2283 src/HOL/HOLCF/Tutorial/Fixrec_ex.thy for examples. INCOMPATIBILITY.
  2284 
  2285 * The "bifinite" class no longer fixes a constant "approx"; the class
  2286 now just asserts that such a function exists.  INCOMPATIBILITY.
  2287 
  2288 * Former type "alg_defl" has been renamed to "defl".  HOLCF no longer
  2289 defines an embedding of type 'a defl into udom by default; instances
  2290 of "bifinite" and "domain" classes are available in
  2291 src/HOL/HOLCF/Library/Defl_Bifinite.thy.
  2292 
  2293 * The syntax "REP('a)" has been replaced with "DEFL('a)".
  2294 
  2295 * The predicate "directed" has been removed.  INCOMPATIBILITY.
  2296 
  2297 * The type class "finite_po" has been removed.  INCOMPATIBILITY.
  2298 
  2299 * The function "cprod_map" has been renamed to "prod_map".
  2300 INCOMPATIBILITY.
  2301 
  2302 * The monadic bind operator on each powerdomain has new binder syntax
  2303 similar to sets, e.g. "\<Union>\<sharp>x\<in>xs. t" represents
  2304 "upper_bind\<cdot>xs\<cdot>(\<Lambda> x. t)".
  2305 
  2306 * The infix syntax for binary union on each powerdomain has changed
  2307 from e.g. "+\<sharp>" to "\<union>\<sharp>", for consistency with set
  2308 syntax.  INCOMPATIBILITY.
  2309 
  2310 * The constant "UU" has been renamed to "bottom".  The syntax "UU" is
  2311 still supported as an input translation.
  2312 
  2313 * Renamed some theorems (the original names are also still available).
  2314 
  2315   expand_fun_below   ~> fun_below_iff
  2316   below_fun_ext      ~> fun_belowI
  2317   expand_cfun_eq     ~> cfun_eq_iff
  2318   ext_cfun           ~> cfun_eqI
  2319   expand_cfun_below  ~> cfun_below_iff
  2320   below_cfun_ext     ~> cfun_belowI
  2321   cont2cont_Rep_CFun ~> cont2cont_APP
  2322 
  2323 * The Abs and Rep functions for various types have changed names.
  2324 Related theorem names have also changed to match. INCOMPATIBILITY.
  2325 
  2326   Rep_CFun  ~> Rep_cfun
  2327   Abs_CFun  ~> Abs_cfun
  2328   Rep_Sprod ~> Rep_sprod
  2329   Abs_Sprod ~> Abs_sprod
  2330   Rep_Ssum  ~> Rep_ssum
  2331   Abs_Ssum  ~> Abs_ssum
  2332 
  2333 * Lemmas with names of the form *_defined_iff or *_strict_iff have
  2334 been renamed to *_bottom_iff.  INCOMPATIBILITY.
  2335 
  2336 * Various changes to bisimulation/coinduction with domain package:
  2337 
  2338   - Definitions of "bisim" constants no longer mention definedness.
  2339   - With mutual recursion, "bisim" predicate is now curried.
  2340   - With mutual recursion, each type gets a separate coind theorem.
  2341   - Variable names in bisim_def and coinduct rules have changed.
  2342 
  2343 INCOMPATIBILITY.
  2344 
  2345 * Case combinators generated by the domain package for type "foo" are
  2346 now named "foo_case" instead of "foo_when".  INCOMPATIBILITY.
  2347 
  2348 * Several theorems have been renamed to more accurately reflect the
  2349 names of constants and types involved.  INCOMPATIBILITY.
  2350 
  2351   thelub_const    ~> lub_const
  2352   lub_const       ~> is_lub_const
  2353   thelubI         ~> lub_eqI
  2354   is_lub_lub      ~> is_lubD2
  2355   lubI            ~> is_lub_lub
  2356   unique_lub      ~> is_lub_unique
  2357   is_ub_lub       ~> is_lub_rangeD1
  2358   lub_bin_chain   ~> is_lub_bin_chain
  2359   lub_fun         ~> is_lub_fun
  2360   thelub_fun      ~> lub_fun
  2361   thelub_cfun     ~> lub_cfun
  2362   thelub_Pair     ~> lub_Pair
  2363   lub_cprod       ~> is_lub_prod
  2364   thelub_cprod    ~> lub_prod
  2365   minimal_cprod   ~> minimal_prod
  2366   inst_cprod_pcpo ~> inst_prod_pcpo
  2367   UU_I            ~> bottomI
  2368   compact_UU      ~> compact_bottom
  2369   deflation_UU    ~> deflation_bottom
  2370   finite_deflation_UU ~> finite_deflation_bottom
  2371 
  2372 * Many legacy theorem names have been discontinued.  INCOMPATIBILITY.
  2373 
  2374   sq_ord_less_eq_trans ~> below_eq_trans
  2375   sq_ord_eq_less_trans ~> eq_below_trans
  2376   refl_less            ~> below_refl
  2377   trans_less           ~> below_trans
  2378   antisym_less         ~> below_antisym
  2379   antisym_less_inverse ~> po_eq_conv [THEN iffD1]
  2380   box_less             ~> box_below
  2381   rev_trans_less       ~> rev_below_trans
  2382   not_less2not_eq      ~> not_below2not_eq
  2383   less_UU_iff          ~> below_UU_iff
  2384   flat_less_iff        ~> flat_below_iff
  2385   adm_less             ~> adm_below
  2386   adm_not_less         ~> adm_not_below
  2387   adm_compact_not_less ~> adm_compact_not_below
  2388   less_fun_def         ~> below_fun_def
  2389   expand_fun_less      ~> fun_below_iff
  2390   less_fun_ext         ~> fun_belowI
  2391   less_discr_def       ~> below_discr_def
  2392   discr_less_eq        ~> discr_below_eq
  2393   less_unit_def        ~> below_unit_def
  2394   less_cprod_def       ~> below_prod_def
  2395   prod_lessI           ~> prod_belowI
  2396   Pair_less_iff        ~> Pair_below_iff
  2397   fst_less_iff         ~> fst_below_iff
  2398   snd_less_iff         ~> snd_below_iff
  2399   expand_cfun_less     ~> cfun_below_iff
  2400   less_cfun_ext        ~> cfun_belowI
  2401   injection_less       ~> injection_below
  2402   less_up_def          ~> below_up_def
  2403   not_Iup_less         ~> not_Iup_below
  2404   Iup_less             ~> Iup_below
  2405   up_less              ~> up_below
  2406   Def_inject_less_eq   ~> Def_below_Def
  2407   Def_less_is_eq       ~> Def_below_iff
  2408   spair_less_iff       ~> spair_below_iff
  2409   less_sprod           ~> below_sprod
  2410   spair_less           ~> spair_below
  2411   sfst_less_iff        ~> sfst_below_iff
  2412   ssnd_less_iff        ~> ssnd_below_iff
  2413   fix_least_less       ~> fix_least_below
  2414   dist_less_one        ~> dist_below_one
  2415   less_ONE             ~> below_ONE
  2416   ONE_less_iff         ~> ONE_below_iff
  2417   less_sinlD           ~> below_sinlD
  2418   less_sinrD           ~> below_sinrD
  2419 
  2420 
  2421 *** FOL and ZF ***
  2422 
  2423 * All constant names are now qualified internally and use proper
  2424 identifiers, e.g. "IFOL.eq" instead of "op =".  INCOMPATIBILITY.
  2425 
  2426 
  2427 *** ML ***
  2428 
  2429 * Antiquotation @{assert} inlines a function bool -> unit that raises
  2430 Fail if the argument is false.  Due to inlining the source position of
  2431 failed assertions is included in the error output.
  2432 
  2433 * Discontinued antiquotation @{theory_ref}, which is obsolete since ML
  2434 text is in practice always evaluated with a stable theory checkpoint.
  2435 Minor INCOMPATIBILITY, use (Theory.check_thy @{theory}) instead.
  2436 
  2437 * Antiquotation @{theory A} refers to theory A from the ancestry of
  2438 the current context, not any accidental theory loader state as before.
  2439 Potential INCOMPATIBILITY, subtle change in semantics.
  2440 
  2441 * Syntax.pretty_priority (default 0) configures the required priority
  2442 of pretty-printed output and thus affects insertion of parentheses.
  2443 
  2444 * Syntax.default_root (default "any") configures the inner syntax
  2445 category (nonterminal symbol) for parsing of terms.
  2446 
  2447 * Former exception Library.UnequalLengths now coincides with
  2448 ListPair.UnequalLengths.
  2449 
  2450 * Renamed structure MetaSimplifier to Raw_Simplifier.  Note that the
  2451 main functionality is provided by structure Simplifier.
  2452 
  2453 * Renamed raw "explode" function to "raw_explode" to emphasize its
  2454 meaning.  Note that internally to Isabelle, Symbol.explode is used in
  2455 almost all situations.
  2456 
  2457 * Discontinued obsolete function sys_error and exception SYS_ERROR.
  2458 See implementation manual for further details on exceptions in
  2459 Isabelle/ML.
  2460 
  2461 * Renamed setmp_noncritical to Unsynchronized.setmp to emphasize its
  2462 meaning.
  2463 
  2464 * Renamed structure PureThy to Pure_Thy and moved most of its
  2465 operations to structure Global_Theory, to emphasize that this is
  2466 rarely-used global-only stuff.
  2467 
  2468 * Discontinued Output.debug.  Minor INCOMPATIBILITY, use plain writeln
  2469 instead (or tracing for high-volume output).
  2470 
  2471 * Configuration option show_question_marks only affects regular pretty
  2472 printing of types and terms, not raw Term.string_of_vname.
  2473 
  2474 * ML_Context.thm and ML_Context.thms are no longer pervasive.  Rare
  2475 INCOMPATIBILITY, superseded by static antiquotations @{thm} and
  2476 @{thms} for most purposes.
  2477 
  2478 * ML structure Unsynchronized is never opened, not even in Isar
  2479 interaction mode as before.  Old Unsynchronized.set etc. have been
  2480 discontinued -- use plain := instead.  This should be *rare* anyway,
  2481 since modern tools always work via official context data, notably
  2482 configuration options.
  2483 
  2484 * Parallel and asynchronous execution requires special care concerning
  2485 interrupts.  Structure Exn provides some convenience functions that
  2486 avoid working directly with raw Interrupt.  User code must not absorb
  2487 interrupts -- intermediate handling (for cleanup etc.) needs to be
  2488 followed by re-raising of the original exception.  Another common
  2489 source of mistakes are "handle _" patterns, which make the meaning of
  2490 the program subject to physical effects of the environment.
  2491 
  2492 
  2493 
  2494 New in Isabelle2009-2 (June 2010)
  2495 ---------------------------------
  2496 
  2497 *** General ***
  2498 
  2499 * Authentic syntax for *all* logical entities (type classes, type
  2500 constructors, term constants): provides simple and robust
  2501 correspondence between formal entities and concrete syntax.  Within
  2502 the parse tree / AST representations, "constants" are decorated by
  2503 their category (class, type, const) and spelled out explicitly with
  2504 their full internal name.
  2505 
  2506 Substantial INCOMPATIBILITY concerning low-level syntax declarations
  2507 and translations (translation rules and translation functions in ML).
  2508 Some hints on upgrading:
  2509 
  2510   - Many existing uses of 'syntax' and 'translations' can be replaced
  2511     by more modern 'type_notation', 'notation' and 'abbreviation',
  2512     which are independent of this issue.
  2513 
  2514   - 'translations' require markup within the AST; the term syntax
  2515     provides the following special forms:
  2516 
  2517       CONST c   -- produces syntax version of constant c from context
  2518       XCONST c  -- literally c, checked as constant from context
  2519       c         -- literally c, if declared by 'syntax'
  2520 
  2521     Plain identifiers are treated as AST variables -- occasionally the
  2522     system indicates accidental variables via the error "rhs contains
  2523     extra variables".
  2524 
  2525     Type classes and type constructors are marked according to their
  2526     concrete syntax.  Some old translations rules need to be written
  2527     for the "type" category, using type constructor application
  2528     instead of pseudo-term application of the default category
  2529     "logic".
  2530 
  2531   - 'parse_translation' etc. in ML may use the following
  2532     antiquotations:
  2533 
  2534       @{class_syntax c}   -- type class c within parse tree / AST
  2535       @{term_syntax c}    -- type constructor c within parse tree / AST
  2536       @{const_syntax c}   -- ML version of "CONST c" above
  2537       @{syntax_const c}   -- literally c (checked wrt. 'syntax' declarations)
  2538 
  2539   - Literal types within 'typed_print_translations', i.e. those *not*
  2540     represented as pseudo-terms are represented verbatim.  Use @{class
  2541     c} or @{type_name c} here instead of the above syntax
  2542     antiquotations.
  2543 
  2544 Note that old non-authentic syntax was based on unqualified base
  2545 names, so all of the above "constant" names would coincide.  Recall
  2546 that 'print_syntax' and ML_command "set Syntax.trace_ast" help to
  2547 diagnose syntax problems.
  2548 
  2549 * Type constructors admit general mixfix syntax, not just infix.
  2550 
  2551 * Concrete syntax may be attached to local entities without a proof
  2552 body, too.  This works via regular mixfix annotations for 'fix',
  2553 'def', 'obtain' etc. or via the explicit 'write' command, which is
  2554 similar to the 'notation' command in theory specifications.
  2555 
  2556 * Discontinued unnamed infix syntax (legacy feature for many years) --
  2557 need to specify constant name and syntax separately.  Internal ML
  2558 datatype constructors have been renamed from InfixName to Infix etc.
  2559 Minor INCOMPATIBILITY.
  2560 
  2561 * Schematic theorem statements need to be explicitly markup as such,
  2562 via commands 'schematic_lemma', 'schematic_theorem',
  2563 'schematic_corollary'.  Thus the relevance of the proof is made
  2564 syntactically clear, which impacts performance in a parallel or
  2565 asynchronous interactive environment.  Minor INCOMPATIBILITY.
  2566 
  2567 * Use of cumulative prems via "!" in some proof methods has been
  2568 discontinued (old legacy feature).
  2569 
  2570 * References 'trace_simp' and 'debug_simp' have been replaced by
  2571 configuration options stored in the context. Enabling tracing (the
  2572 case of debugging is similar) in proofs works via
  2573 
  2574   using [[trace_simp = true]]
  2575 
  2576 Tracing is then active for all invocations of the simplifier in
  2577 subsequent goal refinement steps. Tracing may also still be enabled or
  2578 disabled via the ProofGeneral settings menu.
  2579 
  2580 * Separate commands 'hide_class', 'hide_type', 'hide_const',
  2581 'hide_fact' replace the former 'hide' KIND command.  Minor
  2582 INCOMPATIBILITY.
  2583 
  2584 * Improved parallelism of proof term normalization: usedir -p2 -q0 is
  2585 more efficient than combinations with -q1 or -q2.
  2586 
  2587 
  2588 *** Pure ***
  2589 
  2590 * Proofterms record type-class reasoning explicitly, using the
  2591 "unconstrain" operation internally.  This eliminates all sort
  2592 constraints from a theorem and proof, introducing explicit
  2593 OFCLASS-premises.  On the proof term level, this operation is
  2594 automatically applied at theorem boundaries, such that closed proofs
  2595 are always free of sort constraints.  INCOMPATIBILITY for tools that
  2596 inspect proof terms.
  2597 
  2598 * Local theory specifications may depend on extra type variables that
  2599 are not present in the result type -- arguments TYPE('a) :: 'a itself
  2600 are added internally.  For example:
  2601 
  2602   definition unitary :: bool where "unitary = (ALL (x::'a) y. x = y)"
  2603 
  2604 * Predicates of locales introduced by classes carry a mandatory
  2605 "class" prefix.  INCOMPATIBILITY.
  2606 
  2607 * Vacuous class specifications observe default sort.  INCOMPATIBILITY.
  2608 
  2609 * Old 'axclass' command has been discontinued.  INCOMPATIBILITY, use
  2610 'class' instead.
  2611 
  2612 * Command 'code_reflect' allows to incorporate generated ML code into
  2613 runtime environment; replaces immature code_datatype antiquotation.
  2614 INCOMPATIBILITY.
  2615 
  2616 * Code generator: simple concept for abstract datatypes obeying
  2617 invariants.
  2618 
  2619 * Code generator: details of internal data cache have no impact on the
  2620 user space functionality any longer.
  2621 
  2622 * Methods "unfold_locales" and "intro_locales" ignore non-locale
  2623 subgoals.  This is more appropriate for interpretations with 'where'.
  2624 INCOMPATIBILITY.
  2625 
  2626 * Command 'example_proof' opens an empty proof body.  This allows to
  2627 experiment with Isar, without producing any persistent result.
  2628 
  2629 * Commands 'type_notation' and 'no_type_notation' declare type syntax
  2630 within a local theory context, with explicit checking of the
  2631 constructors involved (in contrast to the raw 'syntax' versions).
  2632 
  2633 * Commands 'types' and 'typedecl' now work within a local theory
  2634 context -- without introducing dependencies on parameters or
  2635 assumptions, which is not possible in Isabelle/Pure.
  2636 
  2637 * Command 'defaultsort' has been renamed to 'default_sort', it works
  2638 within a local theory context.  Minor INCOMPATIBILITY.
  2639 
  2640 
  2641 *** HOL ***
  2642 
  2643 * Command 'typedef' now works within a local theory context -- without
  2644 introducing dependencies on parameters or assumptions, which is not
  2645 possible in Isabelle/Pure/HOL.  Note that the logical environment may
  2646 contain multiple interpretations of local typedefs (with different
  2647 non-emptiness proofs), even in a global theory context.
  2648 
  2649 * New package for quotient types.  Commands 'quotient_type' and
  2650 'quotient_definition' may be used for defining types and constants by
  2651 quotient constructions.  An example is the type of integers created by
  2652 quotienting pairs of natural numbers:
  2653 
  2654   fun
  2655     intrel :: "(nat * nat) => (nat * nat) => bool"
  2656   where
  2657     "intrel (x, y) (u, v) = (x + v = u + y)"
  2658 
  2659   quotient_type int = "nat * nat" / intrel
  2660     by (auto simp add: equivp_def expand_fun_eq)
  2661 
  2662   quotient_definition
  2663     "0::int" is "(0::nat, 0::nat)"
  2664 
  2665 The method "lifting" can be used to lift of theorems from the
  2666 underlying "raw" type to the quotient type.  The example
  2667 src/HOL/Quotient_Examples/FSet.thy includes such a quotient
  2668 construction and provides a reasoning infrastructure for finite sets.
  2669 
  2670 * Renamed Library/Quotient.thy to Library/Quotient_Type.thy to avoid
  2671 clash with new theory Quotient in Main HOL.
  2672 
  2673 * Moved the SMT binding into the main HOL session, eliminating
  2674 separate HOL-SMT session.
  2675 
  2676 * List membership infix mem operation is only an input abbreviation.
  2677 INCOMPATIBILITY.
  2678 
  2679 * Theory Library/Word.thy has been removed.  Use library Word/Word.thy
  2680 for future developements; former Library/Word.thy is still present in
  2681 the AFP entry RSAPPS.
  2682 
  2683 * Theorem Int.int_induct renamed to Int.int_of_nat_induct and is no
  2684 longer shadowed.  INCOMPATIBILITY.
  2685 
  2686 * Dropped theorem duplicate comp_arith; use semiring_norm instead.
  2687 INCOMPATIBILITY.
  2688 
  2689 * Dropped theorem RealPow.real_sq_order; use power2_le_imp_le instead.
  2690 INCOMPATIBILITY.
  2691 
  2692 * Dropped normalizing_semiring etc; use the facts in semiring classes
  2693 instead.  INCOMPATIBILITY.
  2694 
  2695 * Dropped several real-specific versions of lemmas about floor and
  2696 ceiling; use the generic lemmas from theory "Archimedean_Field"
  2697 instead.  INCOMPATIBILITY.
  2698 
  2699   floor_number_of_eq         ~> floor_number_of
  2700   le_floor_eq_number_of      ~> number_of_le_floor
  2701   le_floor_eq_zero           ~> zero_le_floor
  2702   le_floor_eq_one            ~> one_le_floor
  2703   floor_less_eq_number_of    ~> floor_less_number_of
  2704   floor_less_eq_zero         ~> floor_less_zero
  2705   floor_less_eq_one          ~> floor_less_one
  2706   less_floor_eq_number_of    ~> number_of_less_floor
  2707   less_floor_eq_zero         ~> zero_less_floor
  2708   less_floor_eq_one          ~> one_less_floor
  2709   floor_le_eq_number_of      ~> floor_le_number_of
  2710   floor_le_eq_zero           ~> floor_le_zero
  2711   floor_le_eq_one            ~> floor_le_one
  2712   floor_subtract_number_of   ~> floor_diff_number_of
  2713   floor_subtract_one         ~> floor_diff_one
  2714   ceiling_number_of_eq       ~> ceiling_number_of
  2715   ceiling_le_eq_number_of    ~> ceiling_le_number_of
  2716   ceiling_le_zero_eq         ~> ceiling_le_zero
  2717   ceiling_le_eq_one          ~> ceiling_le_one
  2718   less_ceiling_eq_number_of  ~> number_of_less_ceiling
  2719   less_ceiling_eq_zero       ~> zero_less_ceiling
  2720   less_ceiling_eq_one        ~> one_less_ceiling
  2721   ceiling_less_eq_number_of  ~> ceiling_less_number_of
  2722   ceiling_less_eq_zero       ~> ceiling_less_zero
  2723   ceiling_less_eq_one        ~> ceiling_less_one
  2724   le_ceiling_eq_number_of    ~> number_of_le_ceiling
  2725   le_ceiling_eq_zero         ~> zero_le_ceiling
  2726   le_ceiling_eq_one          ~> one_le_ceiling
  2727   ceiling_subtract_number_of ~> ceiling_diff_number_of
  2728   ceiling_subtract_one       ~> ceiling_diff_one
  2729 
  2730 * Theory "Finite_Set": various folding_XXX locales facilitate the
  2731 application of the various fold combinators on finite sets.
  2732 
  2733 * Library theory "RBT" renamed to "RBT_Impl"; new library theory "RBT"
  2734 provides abstract red-black tree type which is backed by "RBT_Impl" as
  2735 implementation.  INCOMPATIBILTY.
  2736 
  2737 * Theory Library/Coinductive_List has been removed -- superseded by
  2738 AFP/thys/Coinductive.
  2739 
  2740 * Theory PReal, including the type "preal" and related operations, has
  2741 been removed.  INCOMPATIBILITY.
  2742 
  2743 * Real: new development using Cauchy Sequences.
  2744 
  2745 * Split off theory "Big_Operators" containing setsum, setprod,
  2746 Inf_fin, Sup_fin, Min, Max from theory Finite_Set.  INCOMPATIBILITY.
  2747 
  2748 * Theory "Rational" renamed to "Rat", for consistency with "Nat",
  2749 "Int" etc.  INCOMPATIBILITY.
  2750 
  2751 * Constant Rat.normalize needs to be qualified.  INCOMPATIBILITY.
  2752 
  2753 * New set of rules "ac_simps" provides combined assoc / commute
  2754 rewrites for all interpretations of the appropriate generic locales.
  2755 
  2756 * Renamed theory "OrderedGroup" to "Groups" and split theory
  2757 "Ring_and_Field" into theories "Rings" and "Fields"; for more
  2758 appropriate and more consistent names suitable for name prefixes
  2759 within the HOL theories.  INCOMPATIBILITY.
  2760 
  2761 * Some generic constants have been put to appropriate theories:
  2762   - less_eq, less: Orderings
  2763   - zero, one, plus, minus, uminus, times, abs, sgn: Groups
  2764   - inverse, divide: Rings
  2765 INCOMPATIBILITY.
  2766 
  2767 * More consistent naming of type classes involving orderings (and
  2768 lattices):
  2769 
  2770     lower_semilattice                   ~> semilattice_inf
  2771     upper_semilattice                   ~> semilattice_sup
  2772 
  2773     dense_linear_order                  ~> dense_linorder
  2774 
  2775     pordered_ab_group_add               ~> ordered_ab_group_add
  2776     pordered_ab_group_add_abs           ~> ordered_ab_group_add_abs
  2777     pordered_ab_semigroup_add           ~> ordered_ab_semigroup_add
  2778     pordered_ab_semigroup_add_imp_le    ~> ordered_ab_semigroup_add_imp_le
  2779     pordered_cancel_ab_semigroup_add    ~> ordered_cancel_ab_semigroup_add
  2780     pordered_cancel_comm_semiring       ~> ordered_cancel_comm_semiring
  2781     pordered_cancel_semiring            ~> ordered_cancel_semiring
  2782     pordered_comm_monoid_add            ~> ordered_comm_monoid_add
  2783     pordered_comm_ring                  ~> ordered_comm_ring
  2784     pordered_comm_semiring              ~> ordered_comm_semiring
  2785     pordered_ring                       ~> ordered_ring
  2786     pordered_ring_abs                   ~> ordered_ring_abs
  2787     pordered_semiring                   ~> ordered_semiring
  2788 
  2789     ordered_ab_group_add                ~> linordered_ab_group_add
  2790     ordered_ab_semigroup_add            ~> linordered_ab_semigroup_add
  2791     ordered_cancel_ab_semigroup_add     ~> linordered_cancel_ab_semigroup_add
  2792     ordered_comm_semiring_strict        ~> linordered_comm_semiring_strict
  2793     ordered_field                       ~> linordered_field
  2794     ordered_field_no_lb                 ~> linordered_field_no_lb
  2795     ordered_field_no_ub                 ~> linordered_field_no_ub
  2796     ordered_field_dense_linear_order    ~> dense_linordered_field
  2797     ordered_idom                        ~> linordered_idom
  2798     ordered_ring                        ~> linordered_ring
  2799     ordered_ring_le_cancel_factor       ~> linordered_ring_le_cancel_factor
  2800     ordered_ring_less_cancel_factor     ~> linordered_ring_less_cancel_factor
  2801     ordered_ring_strict                 ~> linordered_ring_strict
  2802     ordered_semidom                     ~> linordered_semidom
  2803     ordered_semiring                    ~> linordered_semiring
  2804     ordered_semiring_1                  ~> linordered_semiring_1
  2805     ordered_semiring_1_strict           ~> linordered_semiring_1_strict
  2806     ordered_semiring_strict             ~> linordered_semiring_strict
  2807 
  2808   The following slightly odd type classes have been moved to a
  2809   separate theory Library/Lattice_Algebras:
  2810 
  2811     lordered_ab_group_add               ~> lattice_ab_group_add
  2812     lordered_ab_group_add_abs           ~> lattice_ab_group_add_abs
  2813     lordered_ab_group_add_meet          ~> semilattice_inf_ab_group_add
  2814     lordered_ab_group_add_join          ~> semilattice_sup_ab_group_add
  2815     lordered_ring                       ~> lattice_ring
  2816 
  2817 INCOMPATIBILITY.
  2818 
  2819 * Refined field classes:
  2820   - classes division_ring_inverse_zero, field_inverse_zero,
  2821     linordered_field_inverse_zero include rule inverse 0 = 0 --
  2822     subsumes former division_by_zero class;
  2823   - numerous lemmas have been ported from field to division_ring.
  2824 INCOMPATIBILITY.
  2825 
  2826 * Refined algebra theorem collections:
  2827   - dropped theorem group group_simps, use algebra_simps instead;
  2828   - dropped theorem group ring_simps, use field_simps instead;
  2829   - proper theorem collection field_simps subsumes former theorem
  2830     groups field_eq_simps and field_simps;
  2831   - dropped lemma eq_minus_self_iff which is a duplicate for
  2832     equal_neg_zero.
  2833 INCOMPATIBILITY.
  2834 
  2835 * Theory Finite_Set and List: some lemmas have been generalized from
  2836 sets to lattices:
  2837 
  2838   fun_left_comm_idem_inter      ~> fun_left_comm_idem_inf
  2839   fun_left_comm_idem_union      ~> fun_left_comm_idem_sup
  2840   inter_Inter_fold_inter        ~> inf_Inf_fold_inf
  2841   union_Union_fold_union        ~> sup_Sup_fold_sup
  2842   Inter_fold_inter              ~> Inf_fold_inf
  2843   Union_fold_union              ~> Sup_fold_sup
  2844   inter_INTER_fold_inter        ~> inf_INFI_fold_inf
  2845   union_UNION_fold_union        ~> sup_SUPR_fold_sup
  2846   INTER_fold_inter              ~> INFI_fold_inf
  2847   UNION_fold_union              ~> SUPR_fold_sup
  2848 
  2849 * Theory "Complete_Lattice": lemmas top_def and bot_def have been
  2850 replaced by the more convenient lemmas Inf_empty and Sup_empty.
  2851 Dropped lemmas Inf_insert_simp and Sup_insert_simp, which are subsumed
  2852 by Inf_insert and Sup_insert.  Lemmas Inf_UNIV and Sup_UNIV replace
  2853 former Inf_Univ and Sup_Univ.  Lemmas inf_top_right and sup_bot_right
  2854 subsume inf_top and sup_bot respectively.  INCOMPATIBILITY.
  2855 
  2856 * Reorganized theory Multiset: swapped notation of pointwise and
  2857 multiset order:
  2858 
  2859   - pointwise ordering is instance of class order with standard syntax
  2860     <= and <;
  2861   - multiset ordering has syntax <=# and <#; partial order properties
  2862     are provided by means of interpretation with prefix
  2863     multiset_order;
  2864   - less duplication, less historical organization of sections,
  2865     conversion from associations lists to multisets, rudimentary code
  2866     generation;
  2867   - use insert_DiffM2 [symmetric] instead of elem_imp_eq_diff_union,
  2868     if needed.
  2869 
  2870 Renamed:
  2871 
  2872   multiset_eq_conv_count_eq  ~>  multiset_ext_iff
  2873   multi_count_ext  ~>  multiset_ext
  2874   diff_union_inverse2  ~>  diff_union_cancelR
  2875 
  2876 INCOMPATIBILITY.
  2877 
  2878 * Theory Permutation: replaced local "remove" by List.remove1.
  2879 
  2880 * Code generation: ML and OCaml code is decorated with signatures.
  2881 
  2882 * Theory List: added transpose.
  2883 
  2884 * Library/Nat_Bijection.thy is a collection of bijective functions
  2885 between nat and other types, which supersedes the older libraries
  2886 Library/Nat_Int_Bij.thy and HOLCF/NatIso.thy.  INCOMPATIBILITY.
  2887 
  2888   Constants:
  2889   Nat_Int_Bij.nat2_to_nat         ~> prod_encode
  2890   Nat_Int_Bij.nat_to_nat2         ~> prod_decode
  2891   Nat_Int_Bij.int_to_nat_bij      ~> int_encode
  2892   Nat_Int_Bij.nat_to_int_bij      ~> int_decode
  2893   Countable.pair_encode           ~> prod_encode
  2894   NatIso.prod2nat                 ~> prod_encode
  2895   NatIso.nat2prod                 ~> prod_decode
  2896   NatIso.sum2nat                  ~> sum_encode
  2897   NatIso.nat2sum                  ~> sum_decode
  2898   NatIso.list2nat                 ~> list_encode
  2899   NatIso.nat2list                 ~> list_decode
  2900   NatIso.set2nat                  ~> set_encode
  2901   NatIso.nat2set                  ~> set_decode
  2902 
  2903   Lemmas:
  2904   Nat_Int_Bij.bij_nat_to_int_bij  ~> bij_int_decode
  2905   Nat_Int_Bij.nat2_to_nat_inj     ~> inj_prod_encode
  2906   Nat_Int_Bij.nat2_to_nat_surj    ~> surj_prod_encode
  2907   Nat_Int_Bij.nat_to_nat2_inj     ~> inj_prod_decode
  2908   Nat_Int_Bij.nat_to_nat2_surj    ~> surj_prod_decode
  2909   Nat_Int_Bij.i2n_n2i_id          ~> int_encode_inverse
  2910   Nat_Int_Bij.n2i_i2n_id          ~> int_decode_inverse
  2911   Nat_Int_Bij.surj_nat_to_int_bij ~> surj_int_encode
  2912   Nat_Int_Bij.surj_int_to_nat_bij ~> surj_int_decode
  2913   Nat_Int_Bij.inj_nat_to_int_bij  ~> inj_int_encode
  2914   Nat_Int_Bij.inj_int_to_nat_bij  ~> inj_int_decode
  2915   Nat_Int_Bij.bij_nat_to_int_bij  ~> bij_int_encode
  2916   Nat_Int_Bij.bij_int_to_nat_bij  ~> bij_int_decode
  2917 
  2918 * Sledgehammer:
  2919   - Renamed ATP commands:
  2920     atp_info     ~> sledgehammer running_atps
  2921     atp_kill     ~> sledgehammer kill_atps
  2922     atp_messages ~> sledgehammer messages
  2923     atp_minimize ~> sledgehammer minimize
  2924     print_atps   ~> sledgehammer available_atps
  2925     INCOMPATIBILITY.
  2926   - Added user's manual ("isabelle doc sledgehammer").
  2927   - Added option syntax and "sledgehammer_params" to customize
  2928     Sledgehammer's behavior.  See the manual for details.
  2929   - Modified the Isar proof reconstruction code so that it produces
  2930     direct proofs rather than proofs by contradiction.  (This feature
  2931     is still experimental.)
  2932   - Made Isar proof reconstruction work for SPASS, remote ATPs, and in
  2933     full-typed mode.
  2934   - Added support for TPTP syntax for SPASS via the "spass_tptp" ATP.
  2935 
  2936 * Nitpick:
  2937   - Added and implemented "binary_ints" and "bits" options.
  2938   - Added "std" option and implemented support for nonstandard models.
  2939   - Added and implemented "finitize" option to improve the precision
  2940     of infinite datatypes based on a monotonicity analysis.
  2941   - Added support for quotient types.
  2942   - Added support for "specification" and "ax_specification"
  2943     constructs.
  2944   - Added support for local definitions (for "function" and
  2945     "termination" proofs).
  2946   - Added support for term postprocessors.
  2947   - Optimized "Multiset.multiset" and "FinFun.finfun".
  2948   - Improved efficiency of "destroy_constrs" optimization.
  2949   - Fixed soundness bugs related to "destroy_constrs" optimization and
  2950     record getters.
  2951   - Fixed soundness bug related to higher-order constructors.
  2952   - Fixed soundness bug when "full_descrs" is enabled.
  2953   - Improved precision of set constructs.
  2954   - Added "atoms" option.
  2955   - Added cache to speed up repeated Kodkod invocations on the same
  2956     problems.
  2957   - Renamed "MiniSatJNI", "zChaffJNI", "BerkMinAlloy", and
  2958     "SAT4JLight" to "MiniSat_JNI", "zChaff_JNI", "BerkMin_Alloy", and
  2959     "SAT4J_Light".  INCOMPATIBILITY.
  2960   - Removed "skolemize", "uncurry", "sym_break", "flatten_prop",
  2961     "sharing_depth", and "show_skolems" options.  INCOMPATIBILITY.
  2962   - Removed "nitpick_intro" attribute.  INCOMPATIBILITY.
  2963 
  2964 * Method "induct" now takes instantiations of the form t, where t is not
  2965   a variable, as a shorthand for "x == t", where x is a fresh variable.
  2966   If this is not intended, t has to be enclosed in parentheses.
  2967   By default, the equalities generated by definitional instantiations
  2968   are pre-simplified, which may cause parameters of inductive cases
  2969   to disappear, or may even delete some of the inductive cases.
  2970   Use "induct (no_simp)" instead of "induct" to restore the old
  2971   behaviour. The (no_simp) option is also understood by the "cases"
  2972   and "nominal_induct" methods, which now perform pre-simplification, too.
  2973   INCOMPATIBILITY.
  2974 
  2975 
  2976 *** HOLCF ***
  2977 
  2978 * Variable names in lemmas generated by the domain package have
  2979 changed; the naming scheme is now consistent with the HOL datatype
  2980 package.  Some proof scripts may be affected, INCOMPATIBILITY.
  2981 
  2982 * The domain package no longer defines the function "foo_copy" for
  2983 recursive domain "foo".  The reach lemma is now stated directly in
  2984 terms of "foo_take".  Lemmas and proofs that mention "foo_copy" must
  2985 be reformulated in terms of "foo_take", INCOMPATIBILITY.
  2986 
  2987 * Most definedness lemmas generated by the domain package (previously
  2988 of the form "x ~= UU ==> foo$x ~= UU") now have an if-and-only-if form
  2989 like "foo$x = UU <-> x = UU", which works better as a simp rule.
  2990 Proofs that used definedness lemmas as intro rules may break,
  2991 potential INCOMPATIBILITY.
  2992 
  2993 * Induction and casedist rules generated by the domain package now
  2994 declare proper case_names (one called "bottom", and one named for each
  2995 constructor).  INCOMPATIBILITY.
  2996 
  2997 * For mutually-recursive domains, separate "reach" and "take_lemma"
  2998 rules are generated for each domain, INCOMPATIBILITY.
  2999 
  3000   foo_bar.reach       ~> foo.reach  bar.reach
  3001   foo_bar.take_lemmas ~> foo.take_lemma  bar.take_lemma
  3002 
  3003 * Some lemmas generated by the domain package have been renamed for
  3004 consistency with the datatype package, INCOMPATIBILITY.
  3005 
  3006   foo.ind        ~> foo.induct
  3007   foo.finite_ind ~> foo.finite_induct
  3008   foo.coind      ~> foo.coinduct
  3009   foo.casedist   ~> foo.exhaust
  3010   foo.exhaust    ~> foo.nchotomy
  3011 
  3012 * For consistency with other definition packages, the fixrec package
  3013 now generates qualified theorem names, INCOMPATIBILITY.
  3014 
  3015   foo_simps  ~> foo.simps
  3016   foo_unfold ~> foo.unfold
  3017   foo_induct ~> foo.induct
  3018 
  3019 * The "fixrec_simp" attribute has been removed.  The "fixrec_simp"
  3020 method and internal fixrec proofs now use the default simpset instead.
  3021 INCOMPATIBILITY.
  3022 
  3023 * The "contlub" predicate has been removed.  Proof scripts should use
  3024 lemma contI2 in place of monocontlub2cont, INCOMPATIBILITY.
  3025 
  3026 * The "admw" predicate has been removed, INCOMPATIBILITY.
  3027 
  3028 * The constants cpair, cfst, and csnd have been removed in favor of
  3029 Pair, fst, and snd from Isabelle/HOL, INCOMPATIBILITY.
  3030 
  3031 
  3032 *** ML ***
  3033 
  3034 * Antiquotations for basic formal entities:
  3035 
  3036     @{class NAME}         -- type class
  3037     @{class_syntax NAME}  -- syntax representation of the above
  3038 
  3039     @{type_name NAME}     -- logical type
  3040     @{type_abbrev NAME}   -- type abbreviation
  3041     @{nonterminal NAME}   -- type of concrete syntactic category
  3042     @{type_syntax NAME}   -- syntax representation of any of the above
  3043 
  3044     @{const_name NAME}    -- logical constant (INCOMPATIBILITY)
  3045     @{const_abbrev NAME}  -- abbreviated constant
  3046     @{const_syntax NAME}  -- syntax representation of any of the above
  3047 
  3048 * Antiquotation @{syntax_const NAME} ensures that NAME refers to a raw
  3049 syntax constant (cf. 'syntax' command).
  3050 
  3051 * Antiquotation @{make_string} inlines a function to print arbitrary
  3052 values similar to the ML toplevel.  The result is compiler dependent
  3053 and may fall back on "?" in certain situations.
  3054 
  3055 * Diagnostic commands 'ML_val' and 'ML_command' may refer to
  3056 antiquotations @{Isar.state} and @{Isar.goal}.  This replaces impure
  3057 Isar.state() and Isar.goal(), which belong to the old TTY loop and do
  3058 not work with the asynchronous Isar document model.
  3059 
  3060 * Configuration options now admit dynamic default values, depending on
  3061 the context or even global references.
  3062 
  3063 * SHA1.digest digests strings according to SHA-1 (see RFC 3174).  It
  3064 uses an efficient external library if available (for Poly/ML).
  3065 
  3066 * Renamed some important ML structures, while keeping the old names
  3067 for some time as aliases within the structure Legacy:
  3068 
  3069   OuterKeyword  ~>  Keyword
  3070   OuterLex      ~>  Token
  3071   OuterParse    ~>  Parse
  3072   OuterSyntax   ~>  Outer_Syntax
  3073   PrintMode     ~>  Print_Mode
  3074   SpecParse     ~>  Parse_Spec
  3075   ThyInfo       ~>  Thy_Info
  3076   ThyLoad       ~>  Thy_Load
  3077   ThyOutput     ~>  Thy_Output
  3078   TypeInfer     ~>  Type_Infer
  3079 
  3080 Note that "open Legacy" simplifies porting of sources, but forgetting
  3081 to remove it again will complicate porting again in the future.
  3082 
  3083 * Most operations that refer to a global context are named
  3084 accordingly, e.g. Simplifier.global_context or
  3085 ProofContext.init_global.  There are some situations where a global
  3086 context actually works, but under normal circumstances one needs to
  3087 pass the proper local context through the code!
  3088 
  3089 * Discontinued old TheoryDataFun with its copy/init operation -- data
  3090 needs to be pure.  Functor Theory_Data_PP retains the traditional
  3091 Pretty.pp argument to merge, which is absent in the standard
  3092 Theory_Data version.
  3093 
  3094 * Sorts.certify_sort and derived "cert" operations for types and terms
  3095 no longer minimize sorts.  Thus certification at the boundary of the
  3096 inference kernel becomes invariant under addition of class relations,
  3097 which is an important monotonicity principle.  Sorts are now minimized
  3098 in the syntax layer only, at the boundary between the end-user and the
  3099 system.  Subtle INCOMPATIBILITY, may have to use Sign.minimize_sort
  3100 explicitly in rare situations.
  3101 
  3102 * Renamed old-style Drule.standard to Drule.export_without_context, to
  3103 emphasize that this is in no way a standard operation.
  3104 INCOMPATIBILITY.
  3105 
  3106 * Subgoal.FOCUS (and variants): resulting goal state is normalized as
  3107 usual for resolution.  Rare INCOMPATIBILITY.
  3108 
  3109 * Renamed varify/unvarify operations to varify_global/unvarify_global
  3110 to emphasize that these only work in a global situation (which is
  3111 quite rare).
  3112 
  3113 * Curried take and drop in library.ML; negative length is interpreted
  3114 as infinity (as in chop).  Subtle INCOMPATIBILITY.
  3115 
  3116 * Proof terms: type substitutions on proof constants now use canonical
  3117 order of type variables.  INCOMPATIBILITY for tools working with proof
  3118 terms.
  3119 
  3120 * Raw axioms/defs may no longer carry sort constraints, and raw defs
  3121 may no longer carry premises.  User-level specifications are
  3122 transformed accordingly by Thm.add_axiom/add_def.
  3123 
  3124 
  3125 *** System ***
  3126 
  3127 * Discontinued special HOL_USEDIR_OPTIONS for the main HOL image;
  3128 ISABELLE_USEDIR_OPTIONS applies uniformly to all sessions.  Note that
  3129 proof terms are enabled unconditionally in the new HOL-Proofs image.
  3130 
  3131 * Discontinued old ISABELLE and ISATOOL environment settings (legacy
  3132 feature since Isabelle2009).  Use ISABELLE_PROCESS and ISABELLE_TOOL,
  3133 respectively.
  3134 
  3135 * Old lib/scripts/polyml-platform is superseded by the
  3136 ISABELLE_PLATFORM setting variable, which defaults to the 32 bit
  3137 variant, even on a 64 bit machine.  The following example setting
  3138 prefers 64 bit if available:
  3139 
  3140   ML_PLATFORM="${ISABELLE_PLATFORM64:-$ISABELLE_PLATFORM}"
  3141 
  3142 * The preliminary Isabelle/jEdit application demonstrates the emerging
  3143 Isabelle/Scala layer for advanced prover interaction and integration.
  3144 See src/Tools/jEdit or "isabelle jedit" provided by the properly built
  3145 component.
  3146 
  3147 * "IsabelleText" is a Unicode font derived from Bitstream Vera Mono
  3148 and Bluesky TeX fonts.  It provides the usual Isabelle symbols,
  3149 similar to the default assignment of the document preparation system
  3150 (cf. isabellesym.sty).  The Isabelle/Scala class Isabelle_System
  3151 provides some operations for direct access to the font without asking
  3152 the user for manual installation.
  3153 
  3154 
  3155 
  3156 New in Isabelle2009-1 (December 2009)
  3157 -------------------------------------
  3158 
  3159 *** General ***
  3160 
  3161 * Discontinued old form of "escaped symbols" such as \\<forall>.  Only
  3162 one backslash should be used, even in ML sources.
  3163 
  3164 
  3165 *** Pure ***
  3166 
  3167 * Locale interpretation propagates mixins along the locale hierarchy.
  3168 The currently only available mixins are the equations used to map
  3169 local definitions to terms of the target domain of an interpretation.
  3170 
  3171 * Reactivated diagnostic command 'print_interps'.  Use "print_interps
  3172 loc" to print all interpretations of locale "loc" in the theory.
  3173 Interpretations in proofs are not shown.
  3174 
  3175 * Thoroughly revised locales tutorial.  New section on conditional
  3176 interpretation.
  3177 
  3178 * On instantiation of classes, remaining undefined class parameters
  3179 are formally declared.  INCOMPATIBILITY.
  3180 
  3181 
  3182 *** Document preparation ***
  3183 
  3184 * New generalized style concept for printing terms: @{foo (style) ...}
  3185 instead of @{foo_style style ...}  (old form is still retained for
  3186 backward compatibility).  Styles can be also applied for
  3187 antiquotations prop, term_type and typeof.
  3188 
  3189 
  3190 *** HOL ***
  3191 
  3192 * New proof method "smt" for a combination of first-order logic with
  3193 equality, linear and nonlinear (natural/integer/real) arithmetic, and
  3194 fixed-size bitvectors; there is also basic support for higher-order
  3195 features (esp. lambda abstractions).  It is an incomplete decision
  3196 procedure based on external SMT solvers using the oracle mechanism;
  3197 for the SMT solver Z3, this method is proof-producing.  Certificates
  3198 are provided to avoid calling the external solvers solely for
  3199 re-checking proofs.  Due to a remote SMT service there is no need for
  3200 installing SMT solvers locally.  See src/HOL/SMT.
  3201 
  3202 * New commands to load and prove verification conditions generated by
  3203 the Boogie program verifier or derived systems (e.g. the Verifying C
  3204 Compiler (VCC) or Spec#).  See src/HOL/Boogie.
  3205 
  3206 * New counterexample generator tool 'nitpick' based on the Kodkod
  3207 relational model finder.  See src/HOL/Tools/Nitpick and
  3208 src/HOL/Nitpick_Examples.
  3209 
  3210 * New commands 'code_pred' and 'values' to invoke the predicate
  3211 compiler and to enumerate values of inductive predicates.
  3212 
  3213 * A tabled implementation of the reflexive transitive closure.
  3214 
  3215 * New implementation of quickcheck uses generic code generator;
  3216 default generators are provided for all suitable HOL types, records
  3217 and datatypes.  Old quickcheck can be re-activated importing theory
  3218 Library/SML_Quickcheck.
  3219 
  3220 * New testing tool Mirabelle for automated proof tools.  Applies
  3221 several tools and tactics like sledgehammer, metis, or quickcheck, to
  3222 every proof step in a theory.  To be used in batch mode via the
  3223 "mirabelle" utility.
  3224 
  3225 * New proof method "sos" (sum of squares) for nonlinear real
  3226 arithmetic (originally due to John Harison). It requires theory
  3227 Library/Sum_Of_Squares.  It is not a complete decision procedure but
  3228 works well in practice on quantifier-free real arithmetic with +, -,
  3229 *, ^, =, <= and <, i.e. boolean combinations of equalities and
  3230 inequalities between polynomials.  It makes use of external
  3231 semidefinite programming solvers.  Method "sos" generates a
  3232 certificate that can be pasted into the proof thus avoiding the need
  3233 to call an external tool every time the proof is checked.  See
  3234 src/HOL/Library/Sum_Of_Squares.
  3235 
  3236 * New method "linarith" invokes existing linear arithmetic decision
  3237 procedure only.
  3238 
  3239 * New command 'atp_minimal' reduces result produced by Sledgehammer.
  3240 
  3241 * New Sledgehammer option "Full Types" in Proof General settings menu.
  3242 Causes full type information to be output to the ATPs.  This slows
  3243 ATPs down considerably but eliminates a source of unsound "proofs"
  3244 that fail later.
  3245 
  3246 * New method "metisFT": A version of metis that uses full type
  3247 information in order to avoid failures of proof reconstruction.
  3248 
  3249 * New evaluator "approximate" approximates an real valued term using
  3250 the same method as the approximation method.
  3251 
  3252 * Method "approximate" now supports arithmetic expressions as
  3253 boundaries of intervals and implements interval splitting and Taylor
  3254 series expansion.
  3255 
  3256 * ML antiquotation @{code_datatype} inserts definition of a datatype
  3257 generated by the code generator; e.g. see src/HOL/Predicate.thy.
  3258 
  3259 * New theory SupInf of the supremum and infimum operators for sets of
  3260 reals.
  3261 
  3262 * New theory Probability, which contains a development of measure
  3263 theory, eventually leading to Lebesgue integration and probability.
  3264 
  3265 * Extended Multivariate Analysis to include derivation and Brouwer's
  3266 fixpoint theorem.
  3267 
  3268 * Reorganization of number theory, INCOMPATIBILITY:
  3269   - new number theory development for nat and int, in theories Divides
  3270     and GCD as well as in new session Number_Theory
  3271   - some constants and facts now suffixed with _nat and _int
  3272     accordingly
  3273   - former session NumberTheory now named Old_Number_Theory, including
  3274     theories Legacy_GCD and Primes (prefer Number_Theory if possible)
  3275   - moved theory Pocklington from src/HOL/Library to
  3276     src/HOL/Old_Number_Theory
  3277 
  3278 * Theory GCD includes functions Gcd/GCD and Lcm/LCM for the gcd and
  3279 lcm of finite and infinite sets. It is shown that they form a complete
  3280 lattice.
  3281 
  3282 * Class semiring_div requires superclass no_zero_divisors and proof of
  3283 div_mult_mult1; theorems div_mult_mult1, div_mult_mult2,
  3284 div_mult_mult1_if, div_mult_mult1 and div_mult_mult2 have been
  3285 generalized to class semiring_div, subsuming former theorems
  3286 zdiv_zmult_zmult1, zdiv_zmult_zmult1_if, zdiv_zmult_zmult1 and
  3287 zdiv_zmult_zmult2.  div_mult_mult1 is now [simp] by default.
  3288 INCOMPATIBILITY.
  3289 
  3290 * Refinements to lattice classes and sets:
  3291   - less default intro/elim rules in locale variant, more default
  3292     intro/elim rules in class variant: more uniformity
  3293   - lemma ge_sup_conv renamed to le_sup_iff, in accordance with
  3294     le_inf_iff
  3295   - dropped lemma alias inf_ACI for inf_aci (same for sup_ACI and
  3296     sup_aci)
  3297   - renamed ACI to inf_sup_aci
  3298   - new class "boolean_algebra"
  3299   - class "complete_lattice" moved to separate theory
  3300     "Complete_Lattice"; corresponding constants (and abbreviations)
  3301     renamed and with authentic syntax:
  3302     Set.Inf ~>    Complete_Lattice.Inf
  3303     Set.Sup ~>    Complete_Lattice.Sup
  3304     Set.INFI ~>   Complete_Lattice.INFI
  3305     Set.SUPR ~>   Complete_Lattice.SUPR
  3306     Set.Inter ~>  Complete_Lattice.Inter
  3307     Set.Union ~>  Complete_Lattice.Union
  3308     Set.INTER ~>  Complete_Lattice.INTER
  3309     Set.UNION ~>  Complete_Lattice.UNION
  3310   - authentic syntax for
  3311     Set.Pow
  3312     Set.image
  3313   - mere abbreviations:
  3314     Set.empty               (for bot)
  3315     Set.UNIV                (for top)
  3316     Set.inter               (for inf, formerly Set.Int)
  3317     Set.union               (for sup, formerly Set.Un)
  3318     Complete_Lattice.Inter  (for Inf)
  3319     Complete_Lattice.Union  (for Sup)
  3320     Complete_Lattice.INTER  (for INFI)
  3321     Complete_Lattice.UNION  (for SUPR)
  3322   - object-logic definitions as far as appropriate
  3323 
  3324 INCOMPATIBILITY.  Care is required when theorems Int_subset_iff or
  3325 Un_subset_iff are explicitly deleted as default simp rules; then also
  3326 their lattice counterparts le_inf_iff and le_sup_iff have to be
  3327 deleted to achieve the desired effect.
  3328 
  3329 * Rules inf_absorb1, inf_absorb2, sup_absorb1, sup_absorb2 are no simp
  3330 rules by default any longer; the same applies to min_max.inf_absorb1
  3331 etc.  INCOMPATIBILITY.
  3332 
  3333 * Rules sup_Int_eq and sup_Un_eq are no longer declared as
  3334 pred_set_conv by default.  INCOMPATIBILITY.
  3335 
  3336 * Power operations on relations and functions are now one dedicated
  3337 constant "compow" with infix syntax "^^".  Power operation on
  3338 multiplicative monoids retains syntax "^" and is now defined generic
  3339 in class power.  INCOMPATIBILITY.
  3340 
  3341 * Relation composition "R O S" now has a more standard argument order:
  3342 "R O S = {(x, z). EX y. (x, y) : R & (y, z) : S}".  INCOMPATIBILITY,
  3343 rewrite propositions with "S O R" --> "R O S". Proofs may occasionally
  3344 break, since the O_assoc rule was not rewritten like this.  Fix using
  3345 O_assoc[symmetric].  The same applies to the curried version "R OO S".
  3346 
  3347 * Function "Inv" is renamed to "inv_into" and function "inv" is now an
  3348 abbreviation for "inv_into UNIV".  Lemmas are renamed accordingly.
  3349 INCOMPATIBILITY.
  3350 
  3351 * Most rules produced by inductive and datatype package have mandatory
  3352 prefixes.  INCOMPATIBILITY.
  3353 
  3354 * Changed "DERIV_intros" to a dynamic fact, which can be augmented by
  3355 the attribute of the same name.  Each of the theorems in the list
  3356 DERIV_intros assumes composition with an additional function and
  3357 matches a variable to the derivative, which has to be solved by the
  3358 Simplifier.  Hence (auto intro!: DERIV_intros) computes the derivative
  3359 of most elementary terms.  Former Maclauren.DERIV_tac and
  3360 Maclauren.deriv_tac should be replaced by (auto intro!: DERIV_intros).
  3361 INCOMPATIBILITY.
  3362 
  3363 * Code generator attributes follow the usual underscore convention:
  3364     code_unfold     replaces    code unfold
  3365     code_post       replaces    code post
  3366     etc.
  3367   INCOMPATIBILITY.
  3368 
  3369 * Renamed methods:
  3370     sizechange -> size_change
  3371     induct_scheme -> induction_schema
  3372   INCOMPATIBILITY.
  3373 
  3374 * Discontinued abbreviation "arbitrary" of constant "undefined".
  3375 INCOMPATIBILITY, use "undefined" directly.
  3376 
  3377 * Renamed theorems:
  3378     Suc_eq_add_numeral_1 -> Suc_eq_plus1
  3379     Suc_eq_add_numeral_1_left -> Suc_eq_plus1_left
  3380     Suc_plus1 -> Suc_eq_plus1
  3381     *anti_sym -> *antisym*
  3382     vector_less_eq_def -> vector_le_def
  3383   INCOMPATIBILITY.
  3384 
  3385 * Added theorem List.map_map as [simp].  Removed List.map_compose.
  3386 INCOMPATIBILITY.
  3387 
  3388 * Removed predicate "M hassize n" (<--> card M = n & finite M).
  3389 INCOMPATIBILITY.
  3390 
  3391 
  3392 *** HOLCF ***
  3393 
  3394 * Theory Representable defines a class "rep" of domains that are
  3395 representable (via an ep-pair) in the universal domain type "udom".
  3396 Instances are provided for all type constructors defined in HOLCF.
  3397 
  3398 * The 'new_domain' command is a purely definitional version of the
  3399 domain package, for representable domains.  Syntax is identical to the
  3400 old domain package.  The 'new_domain' package also supports indirect
  3401 recursion using previously-defined type constructors.  See
  3402 src/HOLCF/ex/New_Domain.thy for examples.
  3403 
  3404 * Method "fixrec_simp" unfolds one step of a fixrec-defined constant
  3405 on the left-hand side of an equation, and then performs
  3406 simplification.  Rewriting is done using rules declared with the
  3407 "fixrec_simp" attribute.  The "fixrec_simp" method is intended as a
  3408 replacement for "fixpat"; see src/HOLCF/ex/Fixrec_ex.thy for examples.
  3409 
  3410 * The pattern-match compiler in 'fixrec' can now handle constructors
  3411 with HOL function types.  Pattern-match combinators for the Pair
  3412 constructor are pre-configured.
  3413 
  3414 * The 'fixrec' package now produces better fixed-point induction rules
  3415 for mutually-recursive definitions:  Induction rules have conclusions
  3416 of the form "P foo bar" instead of "P <foo, bar>".
  3417 
  3418 * The constant "sq_le" (with infix syntax "<<" or "\<sqsubseteq>") has
  3419 been renamed to "below".  The name "below" now replaces "less" in many
  3420 theorem names.  (Legacy theorem names using "less" are still supported
  3421 as well.)
  3422 
  3423 * The 'fixrec' package now supports "bottom patterns".  Bottom
  3424 patterns can be used to generate strictness rules, or to make
  3425 functions more strict (much like the bang-patterns supported by the
  3426 Glasgow Haskell Compiler).  See src/HOLCF/ex/Fixrec_ex.thy for
  3427 examples.
  3428 
  3429 
  3430 *** ML ***
  3431 
  3432 * Support for Poly/ML 5.3.0, with improved reporting of compiler
  3433 errors and run-time exceptions, including detailed source positions.
  3434 
  3435 * Structure Name_Space (formerly NameSpace) now manages uniquely
  3436 identified entries, with some additional information such as source
  3437 position, logical grouping etc.
  3438 
  3439 * Theory and context data is now introduced by the simplified and
  3440 modernized functors Theory_Data, Proof_Data, Generic_Data.  Data needs
  3441 to be pure, but the old TheoryDataFun for mutable data (with explicit
  3442 copy operation) is still available for some time.
  3443 
  3444 * Structure Synchronized (cf. src/Pure/Concurrent/synchronized.ML)
  3445 provides a high-level programming interface to synchronized state
  3446 variables with atomic update.  This works via pure function
  3447 application within a critical section -- its runtime should be as
  3448 short as possible; beware of deadlocks if critical code is nested,
  3449 either directly or indirectly via other synchronized variables!
  3450 
  3451 * Structure Unsynchronized (cf. src/Pure/ML-Systems/unsynchronized.ML)
  3452 wraps raw ML references, explicitly indicating their non-thread-safe
  3453 behaviour.  The Isar toplevel keeps this structure open, to
  3454 accommodate Proof General as well as quick and dirty interactive
  3455 experiments with references.
  3456 
  3457 * PARALLEL_CHOICE and PARALLEL_GOALS provide basic support for
  3458 parallel tactical reasoning.
  3459 
  3460 * Tacticals Subgoal.FOCUS, Subgoal.FOCUS_PREMS, Subgoal.FOCUS_PARAMS
  3461 are similar to SUBPROOF, but are slightly more flexible: only the
  3462 specified parts of the subgoal are imported into the context, and the
  3463 body tactic may introduce new subgoals and schematic variables.
  3464 
  3465 * Old tactical METAHYPS, which does not observe the proof context, has
  3466 been renamed to Old_Goals.METAHYPS and awaits deletion.  Use SUBPROOF
  3467 or Subgoal.FOCUS etc.
  3468 
  3469 * Renamed functor TableFun to Table, and GraphFun to Graph.  (Since
  3470 functors have their own ML name space there is no point to mark them
  3471 separately.)  Minor INCOMPATIBILITY.
  3472 
  3473 * Renamed NamedThmsFun to Named_Thms.  INCOMPATIBILITY.
  3474 
  3475 * Renamed several structures FooBar to Foo_Bar.  Occasional,
  3476 INCOMPATIBILITY.
  3477 
  3478 * Operations of structure Skip_Proof no longer require quick_and_dirty
  3479 mode, which avoids critical setmp.
  3480 
  3481 * Eliminated old Attrib.add_attributes, Method.add_methods and related
  3482 combinators for "args".  INCOMPATIBILITY, need to use simplified
  3483 Attrib/Method.setup introduced in Isabelle2009.
  3484 
  3485 * Proper context for simpset_of, claset_of, clasimpset_of.  May fall
  3486 back on global_simpset_of, global_claset_of, global_clasimpset_of as
  3487 last resort.  INCOMPATIBILITY.
  3488 
  3489 * Display.pretty_thm now requires a proper context (cf. former
  3490 ProofContext.pretty_thm).  May fall back on Display.pretty_thm_global
  3491 or even Display.pretty_thm_without_context as last resort.
  3492 INCOMPATIBILITY.
  3493 
  3494 * Discontinued Display.pretty_ctyp/cterm etc.  INCOMPATIBILITY, use
  3495 Syntax.pretty_typ/term directly, preferably with proper context
  3496 instead of global theory.
  3497 
  3498 
  3499 *** System ***
  3500 
  3501 * Further fine tuning of parallel proof checking, scales up to 8 cores
  3502 (max. speedup factor 5.0).  See also Goal.parallel_proofs in ML and
  3503 usedir option -q.
  3504 
  3505 * Support for additional "Isabelle components" via etc/components, see
  3506 also the system manual.
  3507 
  3508 * The isabelle makeall tool now operates on all components with
  3509 IsaMakefile, not just hardwired "logics".
  3510 
  3511 * Removed "compress" option from isabelle-process and isabelle usedir;
  3512 this is always enabled.
  3513 
  3514 * Discontinued support for Poly/ML 4.x versions.
  3515 
  3516 * Isabelle tool "wwwfind" provides web interface for 'find_theorems'
  3517 on a given logic image.  This requires the lighttpd webserver and is
  3518 currently supported on Linux only.
  3519 
  3520 
  3521 
  3522 New in Isabelle2009 (April 2009)
  3523 --------------------------------
  3524 
  3525 *** General ***
  3526 
  3527 * Simplified main Isabelle executables, with less surprises on
  3528 case-insensitive file-systems (such as Mac OS).
  3529 
  3530   - The main Isabelle tool wrapper is now called "isabelle" instead of
  3531     "isatool."
  3532 
  3533   - The former "isabelle" alias for "isabelle-process" has been
  3534     removed (should rarely occur to regular users).
  3535 
  3536   - The former "isabelle-interface" and its alias "Isabelle" have been
  3537     removed (interfaces are now regular Isabelle tools).
  3538 
  3539 Within scripts and make files, the Isabelle environment variables
  3540 ISABELLE_TOOL and ISABELLE_PROCESS replace old ISATOOL and ISABELLE,
  3541 respectively.  (The latter are still available as legacy feature.)
  3542 
  3543 The old isabelle-interface wrapper could react in confusing ways if
  3544 the interface was uninstalled or changed otherwise.  Individual
  3545 interface tool configuration is now more explicit, see also the
  3546 Isabelle system manual.  In particular, Proof General is now available
  3547 via "isabelle emacs".
  3548 
  3549 INCOMPATIBILITY, need to adapt derivative scripts.  Users may need to
  3550 purge installed copies of Isabelle executables and re-run "isabelle
  3551 install -p ...", or use symlinks.
  3552 
  3553 * The default for ISABELLE_HOME_USER is now ~/.isabelle instead of the
  3554 old ~/isabelle, which was slightly non-standard and apt to cause
  3555 surprises on case-insensitive file-systems (such as Mac OS).
  3556 
  3557 INCOMPATIBILITY, need to move existing ~/isabelle/etc,
  3558 ~/isabelle/heaps, ~/isabelle/browser_info to the new place.  Special
  3559 care is required when using older releases of Isabelle.  Note that
  3560 ISABELLE_HOME_USER can be changed in Isabelle/etc/settings of any
  3561 Isabelle distribution, in order to use the new ~/.isabelle uniformly.
  3562 
  3563 * Proofs of fully specified statements are run in parallel on
  3564 multi-core systems.  A speedup factor of 2.5 to 3.2 can be expected on
  3565 a regular 4-core machine, if the initial heap space is made reasonably
  3566 large (cf. Poly/ML option -H).  (Requires Poly/ML 5.2.1 or later.)
  3567 
  3568 * The main reference manuals ("isar-ref", "implementation", and
  3569 "system") have been updated and extended.  Formally checked references
  3570 as hyperlinks are now available uniformly.
  3571 
  3572 
  3573 *** Pure ***
  3574 
  3575 * Complete re-implementation of locales.  INCOMPATIBILITY in several
  3576 respects.  The most important changes are listed below.  See the
  3577 Tutorial on Locales ("locales" manual) for details.
  3578 
  3579 - In locale expressions, instantiation replaces renaming.  Parameters
  3580 must be declared in a for clause.  To aid compatibility with previous
  3581 parameter inheritance, in locale declarations, parameters that are not
  3582 'touched' (instantiation position "_" or omitted) are implicitly added
  3583 with their syntax at the beginning of the for clause.
  3584 
  3585 - Syntax from abbreviations and definitions in locales is available in
  3586 locale expressions and context elements.  The latter is particularly
  3587 useful in locale declarations.
  3588 
  3589 - More flexible mechanisms to qualify names generated by locale
  3590 expressions.  Qualifiers (prefixes) may be specified in locale
  3591 expressions, and can be marked as mandatory (syntax: "name!:") or
  3592 optional (syntax "name?:").  The default depends for plain "name:"
  3593 depends on the situation where a locale expression is used: in
  3594 commands 'locale' and 'sublocale' prefixes are optional, in
  3595 'interpretation' and 'interpret' prefixes are mandatory.  The old
  3596 implicit qualifiers derived from the parameter names of a locale are
  3597 no longer generated.
  3598 
  3599 - Command "sublocale l < e" replaces "interpretation l < e".  The
  3600 instantiation clause in "interpretation" and "interpret" (square
  3601 brackets) is no longer available.  Use locale expressions.
  3602 
  3603 - When converting proof scripts, mandatory qualifiers in
  3604 'interpretation' and 'interpret' should be retained by default, even
  3605 if this is an INCOMPATIBILITY compared to former behavior.  In the
  3606 worst case, use the "name?:" form for non-mandatory ones.  Qualifiers
  3607 in locale expressions range over a single locale instance only.
  3608 
  3609 - Dropped locale element "includes".  This is a major INCOMPATIBILITY.
  3610 In existing theorem specifications replace the includes element by the
  3611 respective context elements of the included locale, omitting those
  3612 that are already present in the theorem specification.  Multiple
  3613 assume elements of a locale should be replaced by a single one
  3614 involving the locale predicate.  In the proof body, declarations (most
  3615 notably theorems) may be regained by interpreting the respective
  3616 locales in the proof context as required (command "interpret").
  3617 
  3618 If using "includes" in replacement of a target solely because the
  3619 parameter types in the theorem are not as general as in the target,
  3620 consider declaring a new locale with additional type constraints on
  3621 the parameters (context element "constrains").
  3622 
  3623 - Discontinued "locale (open)".  INCOMPATIBILITY.
  3624 
  3625 - Locale interpretation commands no longer attempt to simplify goal.
  3626 INCOMPATIBILITY: in rare situations the generated goal differs.  Use
  3627 methods intro_locales and unfold_locales to clarify.
  3628 
  3629 - Locale interpretation commands no longer accept interpretation
  3630 attributes.  INCOMPATIBILITY.
  3631 
  3632 * Class declaration: so-called "base sort" must not be given in import
  3633 list any longer, but is inferred from the specification.  Particularly
  3634 in HOL, write
  3635 
  3636     class foo = ...
  3637 
  3638 instead of
  3639 
  3640     class foo = type + ...
  3641 
  3642 * Class target: global versions of theorems stemming do not carry a
  3643 parameter prefix any longer.  INCOMPATIBILITY.
  3644 
  3645 * Class 'instance' command no longer accepts attached definitions.
  3646 INCOMPATIBILITY, use proper 'instantiation' target instead.
  3647 
  3648 * Recovered hiding of consts, which was accidentally broken in
  3649 Isabelle2007.  Potential INCOMPATIBILITY, ``hide const c'' really
  3650 makes c inaccessible; consider using ``hide (open) const c'' instead.
  3651 
  3652 * Slightly more coherent Pure syntax, with updated documentation in
  3653 isar-ref manual.  Removed locales meta_term_syntax and
  3654 meta_conjunction_syntax: TERM and &&& (formerly &&) are now permanent,
  3655 INCOMPATIBILITY in rare situations.  Note that &&& should not be used
  3656 directly in regular applications.
  3657 
  3658 * There is a new syntactic category "float_const" for signed decimal
  3659 fractions (e.g. 123.45 or -123.45).
  3660 
  3661 * Removed exotic 'token_translation' command.  INCOMPATIBILITY, use ML
  3662 interface with 'setup' command instead.
  3663 
  3664 * Command 'local_setup' is similar to 'setup', but operates on a local
  3665 theory context.
  3666 
  3667 * The 'axiomatization' command now only works within a global theory
  3668 context.  INCOMPATIBILITY.
  3669 
  3670 * Goal-directed proof now enforces strict proof irrelevance wrt. sort
  3671 hypotheses.  Sorts required in the course of reasoning need to be
  3672 covered by the constraints in the initial statement, completed by the
  3673 type instance information of the background theory.  Non-trivial sort
  3674 hypotheses, which rarely occur in practice, may be specified via
  3675 vacuous propositions of the form SORT_CONSTRAINT('a::c).  For example:
  3676 
  3677   lemma assumes "SORT_CONSTRAINT('a::empty)" shows False ...
  3678 
  3679 The result contains an implicit sort hypotheses as before --
  3680 SORT_CONSTRAINT premises are eliminated as part of the canonical rule
  3681 normalization.
  3682 
  3683 * Generalized Isar history, with support for linear undo, direct state
  3684 addressing etc.
  3685 
  3686 * Changed defaults for unify configuration options:
  3687 
  3688   unify_trace_bound = 50 (formerly 25)
  3689   unify_search_bound = 60 (formerly 30)
  3690 
  3691 * Different bookkeeping for code equations (INCOMPATIBILITY):
  3692 
  3693   a) On theory merge, the last set of code equations for a particular
  3694      constant is taken (in accordance with the policy applied by other
  3695      parts of the code generator framework).
  3696 
  3697   b) Code equations stemming from explicit declarations (e.g. code
  3698      attribute) gain priority over default code equations stemming
  3699      from definition, primrec, fun etc.
  3700 
  3701 * Keyword 'code_exception' now named 'code_abort'.  INCOMPATIBILITY.
  3702 
  3703 * Unified theorem tables for both code generators.  Thus [code
  3704 func] has disappeared and only [code] remains.  INCOMPATIBILITY.
  3705 
  3706 * Command 'find_consts' searches for constants based on type and name
  3707 patterns, e.g.
  3708 
  3709     find_consts "_ => bool"
  3710 
  3711 By default, matching is against subtypes, but it may be restricted to
  3712 the whole type.  Searching by name is possible.  Multiple queries are
  3713 conjunctive and queries may be negated by prefixing them with a
  3714 hyphen:
  3715 
  3716     find_consts strict: "_ => bool" name: "Int" -"int => int"
  3717 
  3718 * New 'find_theorems' criterion "solves" matches theorems that
  3719 directly solve the current goal (modulo higher-order unification).
  3720 
  3721 * Auto solve feature for main theorem statements: whenever a new goal
  3722 is stated, "find_theorems solves" is called; any theorems that could
  3723 solve the lemma directly are listed as part of the goal state.
  3724 Cf. associated options in Proof General Isabelle settings menu,
  3725 enabled by default, with reasonable timeout for pathological cases of
  3726 higher-order unification.
  3727 
  3728 
  3729 *** Document preparation ***
  3730 
  3731 * Antiquotation @{lemma} now imitates a regular terminal proof,
  3732 demanding keyword 'by' and supporting the full method expression
  3733 syntax just like the Isar command 'by'.
  3734 
  3735 
  3736 *** HOL ***
  3737 
  3738 * Integrated main parts of former image HOL-Complex with HOL.  Entry
  3739 points Main and Complex_Main remain as before.
  3740 
  3741 * Logic image HOL-Plain provides a minimal HOL with the most important
  3742 tools available (inductive, datatype, primrec, ...).  This facilitates
  3743 experimentation and tool development.  Note that user applications
  3744 (and library theories) should never refer to anything below theory
  3745 Main, as before.
  3746 
  3747 * Logic image HOL-Main stops at theory Main, and thus facilitates
  3748 experimentation due to shorter build times.
  3749 
  3750 * Logic image HOL-NSA contains theories of nonstandard analysis which
  3751 were previously part of former HOL-Complex.  Entry point Hyperreal
  3752 remains valid, but theories formerly using Complex_Main should now use
  3753 new entry point Hypercomplex.
  3754 
  3755 * Generic ATP manager for Sledgehammer, based on ML threads instead of
  3756 Posix processes.  Avoids potentially expensive forking of the ML
  3757 process.  New thread-based implementation also works on non-Unix
  3758 platforms (Cygwin).  Provers are no longer hardwired, but defined
  3759 within the theory via plain ML wrapper functions.  Basic Sledgehammer
  3760 commands are covered in the isar-ref manual.
  3761 
  3762 * Wrapper scripts for remote SystemOnTPTP service allows to use
  3763 sledgehammer without local ATP installation (Vampire etc.). Other
  3764 provers may be included via suitable ML wrappers, see also
  3765 src/HOL/ATP_Linkup.thy.
  3766 
  3767 * ATP selection (E/Vampire/Spass) is now via Proof General's settings
  3768 menu.
  3769 
  3770 * The metis method no longer fails because the theorem is too trivial
  3771 (contains the empty clause).
  3772 
  3773 * The metis method now fails in the usual manner, rather than raising
  3774 an exception, if it determines that it cannot prove the theorem.
  3775 
  3776 * Method "coherent" implements a prover for coherent logic (see also
  3777 src/Tools/coherent.ML).
  3778 
  3779 * Constants "undefined" and "default" replace "arbitrary".  Usually
  3780 "undefined" is the right choice to replace "arbitrary", though
  3781 logically there is no difference.  INCOMPATIBILITY.
  3782 
  3783 * Command "value" now integrates different evaluation mechanisms.  The
  3784 result of the first successful evaluation mechanism is printed.  In
  3785 square brackets a particular named evaluation mechanisms may be
  3786 specified (currently, [SML], [code] or [nbe]).  See further
  3787 src/HOL/ex/Eval_Examples.thy.
  3788 
  3789 * Normalization by evaluation now allows non-leftlinear equations.
  3790 Declare with attribute [code nbe].
  3791 
  3792 * Methods "case_tac" and "induct_tac" now refer to the very same rules
  3793 as the structured Isar versions "cases" and "induct", cf. the
  3794 corresponding "cases" and "induct" attributes.  Mutual induction rules
  3795 are now presented as a list of individual projections
  3796 (e.g. foo_bar.inducts for types foo and bar); the old format with
  3797 explicit HOL conjunction is no longer supported.  INCOMPATIBILITY, in
  3798 rare situations a different rule is selected --- notably nested tuple
  3799 elimination instead of former prod.exhaust: use explicit (case_tac t
  3800 rule: prod.exhaust) here.
  3801 
  3802 * Attributes "cases", "induct", "coinduct" support "del" option.
  3803 
  3804 * Removed fact "case_split_thm", which duplicates "case_split".
  3805 
  3806 * The option datatype has been moved to a new theory Option.  Renamed
  3807 option_map to Option.map, and o2s to Option.set, INCOMPATIBILITY.
  3808 
  3809 * New predicate "strict_mono" classifies strict functions on partial
  3810 orders.  With strict functions on linear orders, reasoning about
  3811 (in)equalities is facilitated by theorems "strict_mono_eq",
  3812 "strict_mono_less_eq" and "strict_mono_less".
  3813 
  3814 * Some set operations are now proper qualified constants with
  3815 authentic syntax.  INCOMPATIBILITY:
  3816 
  3817     op Int ~>   Set.Int
  3818     op Un ~>    Set.Un
  3819     INTER ~>    Set.INTER
  3820     UNION ~>    Set.UNION
  3821     Inter ~>    Set.Inter
  3822     Union ~>    Set.Union
  3823     {} ~>       Set.empty
  3824     UNIV ~>     Set.UNIV
  3825 
  3826 * Class complete_lattice with operations Inf, Sup, INFI, SUPR now in
  3827 theory Set.
  3828 
  3829 * Auxiliary class "itself" has disappeared -- classes without any
  3830 parameter are treated as expected by the 'class' command.
  3831 
  3832 * Leibnitz's Series for Pi and the arcus tangens and logarithm series.
  3833 
  3834 * Common decision procedures (Cooper, MIR, Ferrack, Approximation,
  3835 Dense_Linear_Order) are now in directory HOL/Decision_Procs.
  3836 
  3837 * Theory src/HOL/Decision_Procs/Approximation provides the new proof
  3838 method "approximation".  It proves formulas on real values by using
  3839 interval arithmetic.  In the formulas are also the transcendental
  3840 functions sin, cos, tan, atan, ln, exp and the constant pi are
  3841 allowed. For examples see
  3842 src/HOL/Descision_Procs/ex/Approximation_Ex.thy.
  3843 
  3844 * Theory "Reflection" now resides in HOL/Library.
  3845 
  3846 * Entry point to Word library now simply named "Word".
  3847 INCOMPATIBILITY.
  3848 
  3849 * Made source layout more coherent with logical distribution
  3850 structure:
  3851 
  3852     src/HOL/Library/RType.thy ~> src/HOL/Typerep.thy
  3853     src/HOL/Library/Code_Message.thy ~> src/HOL/
  3854     src/HOL/Library/GCD.thy ~> src/HOL/
  3855     src/HOL/Library/Order_Relation.thy ~> src/HOL/
  3856     src/HOL/Library/Parity.thy ~> src/HOL/
  3857     src/HOL/Library/Univ_Poly.thy ~> src/HOL/
  3858     src/HOL/Real/ContNotDenum.thy ~> src/HOL/Library/
  3859     src/HOL/Real/Lubs.thy ~> src/HOL/
  3860     src/HOL/Real/PReal.thy ~> src/HOL/
  3861     src/HOL/Real/Rational.thy ~> src/HOL/
  3862     src/HOL/Real/RComplete.thy ~> src/HOL/
  3863     src/HOL/Real/RealDef.thy ~> src/HOL/
  3864     src/HOL/Real/RealPow.thy ~> src/HOL/
  3865     src/HOL/Real/Real.thy ~> src/HOL/
  3866     src/HOL/Complex/Complex_Main.thy ~> src/HOL/
  3867     src/HOL/Complex/Complex.thy ~> src/HOL/
  3868     src/HOL/Complex/FrechetDeriv.thy ~> src/HOL/Library/
  3869     src/HOL/Complex/Fundamental_Theorem_Algebra.thy ~> src/HOL/Library/
  3870     src/HOL/Hyperreal/Deriv.thy ~> src/HOL/
  3871     src/HOL/Hyperreal/Fact.thy ~> src/HOL/
  3872     src/HOL/Hyperreal/Integration.thy ~> src/HOL/
  3873     src/HOL/Hyperreal/Lim.thy ~> src/HOL/
  3874     src/HOL/Hyperreal/Ln.thy ~> src/HOL/
  3875     src/HOL/Hyperreal/Log.thy ~> src/HOL/
  3876     src/HOL/Hyperreal/MacLaurin.thy ~> src/HOL/
  3877     src/HOL/Hyperreal/NthRoot.thy ~> src/HOL/
  3878     src/HOL/Hyperreal/Series.thy ~> src/HOL/
  3879     src/HOL/Hyperreal/SEQ.thy ~> src/HOL/
  3880     src/HOL/Hyperreal/Taylor.thy ~> src/HOL/
  3881     src/HOL/Hyperreal/Transcendental.thy ~> src/HOL/
  3882     src/HOL/Real/Float ~> src/HOL/Library/
  3883     src/HOL/Real/HahnBanach ~> src/HOL/HahnBanach
  3884     src/HOL/Real/RealVector.thy ~> src/HOL/
  3885 
  3886     src/HOL/arith_data.ML ~> src/HOL/Tools
  3887     src/HOL/hologic.ML ~> src/HOL/Tools
  3888     src/HOL/simpdata.ML ~> src/HOL/Tools
  3889     src/HOL/int_arith1.ML ~> src/HOL/Tools/int_arith.ML
  3890     src/HOL/int_factor_simprocs.ML ~> src/HOL/Tools
  3891     src/HOL/nat_simprocs.ML ~> src/HOL/Tools
  3892     src/HOL/Real/float_arith.ML ~> src/HOL/Tools
  3893     src/HOL/Real/float_syntax.ML ~> src/HOL/Tools
  3894     src/HOL/Real/rat_arith.ML ~> src/HOL/Tools
  3895     src/HOL/Real/real_arith.ML ~> src/HOL/Tools
  3896 
  3897     src/HOL/Library/Array.thy ~> src/HOL/Imperative_HOL
  3898     src/HOL/Library/Heap_Monad.thy ~> src/HOL/Imperative_HOL
  3899     src/HOL/Library/Heap.thy ~> src/HOL/Imperative_HOL
  3900     src/HOL/Library/Imperative_HOL.thy ~> src/HOL/Imperative_HOL
  3901     src/HOL/Library/Ref.thy ~> src/HOL/Imperative_HOL
  3902     src/HOL/Library/Relational.thy ~> src/HOL/Imperative_HOL
  3903 
  3904 * If methods "eval" and "evaluation" encounter a structured proof
  3905 state with !!/==>, only the conclusion is evaluated to True (if
  3906 possible), avoiding strange error messages.
  3907 
  3908 * Method "sizechange" automates termination proofs using (a
  3909 modification of) the size-change principle.  Requires SAT solver.  See
  3910 src/HOL/ex/Termination.thy for examples.
  3911 
  3912 * Simplifier: simproc for let expressions now unfolds if bound
  3913 variable occurs at most once in let expression body.  INCOMPATIBILITY.
  3914 
  3915 * Method "arith": Linear arithmetic now ignores all inequalities when
  3916 fast_arith_neq_limit is exceeded, instead of giving up entirely.
  3917 
  3918 * New attribute "arith" for facts that should always be used
  3919 automatically by arithmetic. It is intended to be used locally in
  3920 proofs, e.g.
  3921 
  3922   assumes [arith]: "x > 0"
  3923 
  3924 Global usage is discouraged because of possible performance impact.
  3925 
  3926 * New classes "top" and "bot" with corresponding operations "top" and
  3927 "bot" in theory Orderings; instantiation of class "complete_lattice"
  3928 requires instantiation of classes "top" and "bot".  INCOMPATIBILITY.
  3929 
  3930 * Changed definition lemma "less_fun_def" in order to provide an
  3931 instance for preorders on functions; use lemma "less_le" instead.
  3932 INCOMPATIBILITY.
  3933 
  3934 * Theory Orderings: class "wellorder" moved here, with explicit
  3935 induction rule "less_induct" as assumption.  For instantiation of
  3936 "wellorder" by means of predicate "wf", use rule wf_wellorderI.
  3937 INCOMPATIBILITY.
  3938 
  3939 * Theory Orderings: added class "preorder" as superclass of "order".
  3940 INCOMPATIBILITY: Instantiation proofs for order, linorder
  3941 etc. slightly changed.  Some theorems named order_class.* now named
  3942 preorder_class.*.
  3943 
  3944 * Theory Relation: renamed "refl" to "refl_on", "reflexive" to "refl,
  3945 "diag" to "Id_on".
  3946 
  3947 * Theory Finite_Set: added a new fold combinator of type
  3948 
  3949   ('a => 'b => 'b) => 'b => 'a set => 'b
  3950 
  3951 Occasionally this is more convenient than the old fold combinator
  3952 which is now defined in terms of the new one and renamed to
  3953 fold_image.
  3954 
  3955 * Theories Ring_and_Field and OrderedGroup: The lemmas "group_simps"
  3956 and "ring_simps" have been replaced by "algebra_simps" (which can be
  3957 extended with further lemmas!).  At the moment both still exist but
  3958 the former will disappear at some point.
  3959 
  3960 * Theory Power: Lemma power_Suc is now declared as a simp rule in
  3961 class recpower.  Type-specific simp rules for various recpower types
  3962 have been removed.  INCOMPATIBILITY, rename old lemmas as follows:
  3963 
  3964 rat_power_0    -> power_0
  3965 rat_power_Suc  -> power_Suc
  3966 realpow_0      -> power_0
  3967 realpow_Suc    -> power_Suc
  3968 complexpow_0   -> power_0
  3969 complexpow_Suc -> power_Suc
  3970 power_poly_0   -> power_0
  3971 power_poly_Suc -> power_Suc
  3972 
  3973 * Theories Ring_and_Field and Divides: Definition of "op dvd" has been
  3974 moved to separate class dvd in Ring_and_Field; a couple of lemmas on
  3975 dvd has been generalized to class comm_semiring_1.  Likewise a bunch
  3976 of lemmas from Divides has been generalized from nat to class
  3977 semiring_div.  INCOMPATIBILITY.  This involves the following theorem
  3978 renames resulting from duplicate elimination:
  3979 
  3980     dvd_def_mod ~>          dvd_eq_mod_eq_0
  3981     zero_dvd_iff ~>         dvd_0_left_iff
  3982     dvd_0 ~>                dvd_0_right
  3983     DIVISION_BY_ZERO_DIV ~> div_by_0
  3984     DIVISION_BY_ZERO_MOD ~> mod_by_0
  3985     mult_div ~>             div_mult_self2_is_id
  3986     mult_mod ~>             mod_mult_self2_is_0
  3987 
  3988 * Theory IntDiv: removed many lemmas that are instances of class-based
  3989 generalizations (from Divides and Ring_and_Field).  INCOMPATIBILITY,
  3990 rename old lemmas as follows:
  3991 
  3992 dvd_diff               -> nat_dvd_diff
  3993 dvd_zminus_iff         -> dvd_minus_iff
  3994 mod_add1_eq            -> mod_add_eq
  3995 mod_mult1_eq           -> mod_mult_right_eq
  3996 mod_mult1_eq'          -> mod_mult_left_eq
  3997 mod_mult_distrib_mod   -> mod_mult_eq
  3998 nat_mod_add_left_eq    -> mod_add_left_eq
  3999 nat_mod_add_right_eq   -> mod_add_right_eq
  4000 nat_mod_div_trivial    -> mod_div_trivial
  4001 nat_mod_mod_trivial    -> mod_mod_trivial
  4002 zdiv_zadd_self1        -> div_add_self1
  4003 zdiv_zadd_self2        -> div_add_self2
  4004 zdiv_zmult_self1       -> div_mult_self2_is_id
  4005 zdiv_zmult_self2       -> div_mult_self1_is_id
  4006 zdvd_triv_left         -> dvd_triv_left
  4007 zdvd_triv_right        -> dvd_triv_right
  4008 zdvd_zmult_cancel_disj -> dvd_mult_cancel_left
  4009 zmod_eq0_zdvd_iff      -> dvd_eq_mod_eq_0[symmetric]
  4010 zmod_zadd_left_eq      -> mod_add_left_eq
  4011 zmod_zadd_right_eq     -> mod_add_right_eq
  4012 zmod_zadd_self1        -> mod_add_self1
  4013 zmod_zadd_self2        -> mod_add_self2
  4014 zmod_zadd1_eq          -> mod_add_eq
  4015 zmod_zdiff1_eq         -> mod_diff_eq
  4016 zmod_zdvd_zmod         -> mod_mod_cancel
  4017 zmod_zmod_cancel       -> mod_mod_cancel
  4018 zmod_zmult_self1       -> mod_mult_self2_is_0
  4019 zmod_zmult_self2       -> mod_mult_self1_is_0
  4020 zmod_1                 -> mod_by_1
  4021 zdiv_1                 -> div_by_1
  4022 zdvd_abs1              -> abs_dvd_iff
  4023 zdvd_abs2              -> dvd_abs_iff
  4024 zdvd_refl              -> dvd_refl
  4025 zdvd_trans             -> dvd_trans
  4026 zdvd_zadd              -> dvd_add
  4027 zdvd_zdiff             -> dvd_diff
  4028 zdvd_zminus_iff        -> dvd_minus_iff
  4029 zdvd_zminus2_iff       -> minus_dvd_iff
  4030 zdvd_zmultD            -> dvd_mult_right
  4031 zdvd_zmultD2           -> dvd_mult_left
  4032 zdvd_zmult_mono        -> mult_dvd_mono
  4033 zdvd_0_right           -> dvd_0_right
  4034 zdvd_0_left            -> dvd_0_left_iff
  4035 zdvd_1_left            -> one_dvd
  4036 zminus_dvd_iff         -> minus_dvd_iff
  4037 
  4038 * Theory Rational: 'Fract k 0' now equals '0'.  INCOMPATIBILITY.
  4039 
  4040 * The real numbers offer decimal input syntax: 12.34 is translated
  4041 into 1234/10^2. This translation is not reversed upon output.
  4042 
  4043 * Theory Library/Polynomial defines an abstract type 'a poly of
  4044 univariate polynomials with coefficients of type 'a.  In addition to
  4045 the standard ring operations, it also supports div and mod.  Code
  4046 generation is also supported, using list-style constructors.
  4047 
  4048 * Theory Library/Inner_Product defines a class of real_inner for real
  4049 inner product spaces, with an overloaded operation inner :: 'a => 'a
  4050 => real.  Class real_inner is a subclass of real_normed_vector from
  4051 theory RealVector.
  4052 
  4053 * Theory Library/Product_Vector provides instances for the product
  4054 type 'a * 'b of several classes from RealVector and Inner_Product.
  4055 Definitions of addition, subtraction, scalar multiplication, norms,
  4056 and inner products are included.
  4057 
  4058 * Theory Library/Bit defines the field "bit" of integers modulo 2.  In
  4059 addition to the field operations, numerals and case syntax are also
  4060 supported.
  4061 
  4062 * Theory Library/Diagonalize provides constructive version of Cantor's
  4063 first diagonalization argument.
  4064 
  4065 * Theory Library/GCD: Curried operations gcd, lcm (for nat) and zgcd,
  4066 zlcm (for int); carried together from various gcd/lcm developements in
  4067 the HOL Distribution.  Constants zgcd and zlcm replace former igcd and
  4068 ilcm; corresponding theorems renamed accordingly.  INCOMPATIBILITY,
  4069 may recover tupled syntax as follows:
  4070 
  4071     hide (open) const gcd
  4072     abbreviation gcd where
  4073       "gcd == (%(a, b). GCD.gcd a b)"
  4074     notation (output)
  4075       GCD.gcd ("gcd '(_, _')")
  4076 
  4077 The same works for lcm, zgcd, zlcm.
  4078 
  4079 * Theory Library/Nat_Infinity: added addition, numeral syntax and more
  4080 instantiations for algebraic structures.  Removed some duplicate
  4081 theorems.  Changes in simp rules.  INCOMPATIBILITY.
  4082 
  4083 * ML antiquotation @{code} takes a constant as argument and generates
  4084 corresponding code in background and inserts name of the corresponding
  4085 resulting ML value/function/datatype constructor binding in place.
  4086 All occurrences of @{code} with a single ML block are generated
  4087 simultaneously.  Provides a generic and safe interface for
  4088 instrumentalizing code generation.  See
  4089 src/HOL/Decision_Procs/Ferrack.thy for a more ambitious application.
  4090 In future you ought to refrain from ad-hoc compiling generated SML
  4091 code on the ML toplevel.  Note that (for technical reasons) @{code}
  4092 cannot refer to constants for which user-defined serializations are
  4093 set.  Refer to the corresponding ML counterpart directly in that
  4094 cases.
  4095 
  4096 * Command 'rep_datatype': instead of theorem names the command now
  4097 takes a list of terms denoting the constructors of the type to be
  4098 represented as datatype.  The characteristic theorems have to be
  4099 proven.  INCOMPATIBILITY.  Also observe that the following theorems
  4100 have disappeared in favour of existing ones:
  4101 
  4102     unit_induct                 ~> unit.induct
  4103     prod_induct                 ~> prod.induct
  4104     sum_induct                  ~> sum.induct
  4105     Suc_Suc_eq                  ~> nat.inject
  4106     Suc_not_Zero Zero_not_Suc   ~> nat.distinct
  4107 
  4108 
  4109 *** HOL-Algebra ***
  4110 
  4111 * New locales for orders and lattices where the equivalence relation
  4112 is not restricted to equality.  INCOMPATIBILITY: all order and lattice
  4113 locales use a record structure with field eq for the equivalence.
  4114 
  4115 * New theory of factorial domains.
  4116 
  4117 * Units_l_inv and Units_r_inv are now simp rules by default.
  4118 INCOMPATIBILITY.  Simplifier proof that require deletion of l_inv
  4119 and/or r_inv will now also require deletion of these lemmas.
  4120 
  4121 * Renamed the following theorems, INCOMPATIBILITY:
  4122 
  4123 UpperD ~> Upper_memD
  4124 LowerD ~> Lower_memD
  4125 least_carrier ~> least_closed
  4126 greatest_carrier ~> greatest_closed
  4127 greatest_Lower_above ~> greatest_Lower_below
  4128 one_zero ~> carrier_one_zero
  4129 one_not_zero ~> carrier_one_not_zero  (collision with assumption)
  4130 
  4131 
  4132 *** HOL-Nominal ***
  4133 
  4134 * Nominal datatypes can now contain type-variables.
  4135 
  4136 * Commands 'nominal_inductive' and 'equivariance' work with local
  4137 theory targets.
  4138 
  4139 * Nominal primrec can now works with local theory targets and its
  4140 specification syntax now conforms to the general format as seen in
  4141 'inductive' etc.
  4142 
  4143 * Method "perm_simp" honours the standard simplifier attributes
  4144 (no_asm), (no_asm_use) etc.
  4145 
  4146 * The new predicate #* is defined like freshness, except that on the
  4147 left hand side can be a set or list of atoms.
  4148 
  4149 * Experimental command 'nominal_inductive2' derives strong induction
  4150 principles for inductive definitions.  In contrast to
  4151 'nominal_inductive', which can only deal with a fixed number of
  4152 binders, it can deal with arbitrary expressions standing for sets of
  4153 atoms to be avoided.  The only inductive definition we have at the
  4154 moment that needs this generalisation is the typing rule for Lets in
  4155 the algorithm W:
  4156 
  4157  Gamma |- t1 : T1   (x,close Gamma T1)::Gamma |- t2 : T2   x#Gamma
  4158  -----------------------------------------------------------------
  4159          Gamma |- Let x be t1 in t2 : T2
  4160 
  4161 In this rule one wants to avoid all the binders that are introduced by
  4162 "close Gamma T1".  We are looking for other examples where this
  4163 feature might be useful.  Please let us know.
  4164 
  4165 
  4166 *** HOLCF ***
  4167 
  4168 * Reimplemented the simplification procedure for proving continuity
  4169 subgoals.  The new simproc is extensible; users can declare additional
  4170 continuity introduction rules with the attribute [cont2cont].
  4171 
  4172 * The continuity simproc now uses a different introduction rule for
  4173 solving continuity subgoals on terms with lambda abstractions.  In
  4174 some rare cases the new simproc may fail to solve subgoals that the
  4175 old one could solve, and "simp add: cont2cont_LAM" may be necessary.
  4176 Potential INCOMPATIBILITY.
  4177 
  4178 * Command 'fixrec': specification syntax now conforms to the general
  4179 format as seen in 'inductive' etc.  See src/HOLCF/ex/Fixrec_ex.thy for
  4180 examples.  INCOMPATIBILITY.
  4181 
  4182 
  4183 *** ZF ***
  4184 
  4185 * Proof of Zorn's Lemma for partial orders.
  4186 
  4187 
  4188 *** ML ***
  4189 
  4190 * Multithreading for Poly/ML 5.1/5.2 is no longer supported, only for
  4191 Poly/ML 5.2.1 or later.  Important note: the TimeLimit facility
  4192 depends on multithreading, so timouts will not work before Poly/ML
  4193 5.2.1!
  4194 
  4195 * High-level support for concurrent ML programming, see
  4196 src/Pure/Cuncurrent.  The data-oriented model of "future values" is
  4197 particularly convenient to organize independent functional
  4198 computations.  The concept of "synchronized variables" provides a
  4199 higher-order interface for components with shared state, avoiding the
  4200 delicate details of mutexes and condition variables.  (Requires
  4201 Poly/ML 5.2.1 or later.)
  4202 
  4203 * ML bindings produced via Isar commands are stored within the Isar
  4204 context (theory or proof).  Consequently, commands like 'use' and 'ML'
  4205 become thread-safe and work with undo as expected (concerning
  4206 top-level bindings, not side-effects on global references).
  4207 INCOMPATIBILITY, need to provide proper Isar context when invoking the
  4208 compiler at runtime; really global bindings need to be given outside a
  4209 theory.  (Requires Poly/ML 5.2 or later.)
  4210 
  4211 * Command 'ML_prf' is analogous to 'ML' but works within a proof
  4212 context.  Top-level ML bindings are stored within the proof context in
  4213 a purely sequential fashion, disregarding the nested proof structure.
  4214 ML bindings introduced by 'ML_prf' are discarded at the end of the
  4215 proof.  (Requires Poly/ML 5.2 or later.)
  4216 
  4217 * Simplified ML attribute and method setup, cf. functions Attrib.setup
  4218 and Method.setup, as well as Isar commands 'attribute_setup' and
  4219 'method_setup'.  INCOMPATIBILITY for 'method_setup', need to simplify
  4220 existing code accordingly, or use plain 'setup' together with old
  4221 Method.add_method.
  4222 
  4223 * Simplified ML oracle interface Thm.add_oracle promotes 'a -> cterm
  4224 to 'a -> thm, while results are always tagged with an authentic oracle
  4225 name.  The Isar command 'oracle' is now polymorphic, no argument type
  4226 is specified.  INCOMPATIBILITY, need to simplify existing oracle code
  4227 accordingly.  Note that extra performance may be gained by producing
  4228 the cterm carefully, avoiding slow Thm.cterm_of.
  4229 
  4230 * Simplified interface for defining document antiquotations via
  4231 ThyOutput.antiquotation, ThyOutput.output, and optionally
  4232 ThyOutput.maybe_pretty_source.  INCOMPATIBILITY, need to simplify user
  4233 antiquotations accordingly, see src/Pure/Thy/thy_output.ML for common
  4234 examples.
  4235 
  4236 * More systematic treatment of long names, abstract name bindings, and
  4237 name space operations.  Basic operations on qualified names have been
  4238 move from structure NameSpace to Long_Name, e.g. Long_Name.base_name,
  4239 Long_Name.append.  Old type bstring has been mostly replaced by
  4240 abstract type binding (see structure Binding), which supports precise
  4241 qualification by packages and local theory targets, as well as proper
  4242 tracking of source positions.  INCOMPATIBILITY, need to wrap old
  4243 bstring values into Binding.name, or better pass through abstract
  4244 bindings everywhere.  See further src/Pure/General/long_name.ML,
  4245 src/Pure/General/binding.ML and src/Pure/General/name_space.ML
  4246 
  4247 * Result facts (from PureThy.note_thms, ProofContext.note_thms,
  4248 LocalTheory.note etc.) now refer to the *full* internal name, not the
  4249 bstring as before.  INCOMPATIBILITY, not detected by ML type-checking!
  4250 
  4251 * Disposed old type and term read functions (Sign.read_def_typ,
  4252 Sign.read_typ, Sign.read_def_terms, Sign.read_term,
  4253 Thm.read_def_cterms, Thm.read_cterm etc.).  INCOMPATIBILITY, should
  4254 use regular Syntax.read_typ, Syntax.read_term, Syntax.read_typ_global,
  4255 Syntax.read_term_global etc.; see also OldGoals.read_term as last
  4256 resort for legacy applications.
  4257 
  4258 * Disposed old declarations, tactics, tactic combinators that refer to
  4259 the simpset or claset of an implicit theory (such as Addsimps,
  4260 Simp_tac, SIMPSET).  INCOMPATIBILITY, should use @{simpset} etc. in
  4261 embedded ML text, or local_simpset_of with a proper context passed as
  4262 explicit runtime argument.
  4263 
  4264 * Rules and tactics that read instantiations (read_instantiate,
  4265 res_inst_tac, thin_tac, subgoal_tac etc.) now demand a proper proof
  4266 context, which is required for parsing and type-checking.  Moreover,
  4267 the variables are specified as plain indexnames, not string encodings
  4268 thereof.  INCOMPATIBILITY.
  4269 
  4270 * Generic Toplevel.add_hook interface allows to analyze the result of
  4271 transactions.  E.g. see src/Pure/ProofGeneral/proof_general_pgip.ML
  4272 for theorem dependency output of transactions resulting in a new
  4273 theory state.
  4274 
  4275 * ML antiquotations: block-structured compilation context indicated by
  4276 \<lbrace> ... \<rbrace>; additional antiquotation forms:
  4277 
  4278   @{binding name}                         - basic name binding
  4279   @{let ?pat = term}                      - term abbreviation (HO matching)
  4280   @{note name = fact}                     - fact abbreviation
  4281   @{thm fact}                             - singleton fact (with attributes)
  4282   @{thms fact}                            - general fact (with attributes)
  4283   @{lemma prop by method}                 - singleton goal
  4284   @{lemma prop by meth1 meth2}            - singleton goal
  4285   @{lemma prop1 ... propN by method}      - general goal
  4286   @{lemma prop1 ... propN by meth1 meth2} - general goal
  4287   @{lemma (open) ...}                     - open derivation
  4288 
  4289 
  4290 *** System ***
  4291 
  4292 * The Isabelle "emacs" tool provides a specific interface to invoke
  4293 Proof General / Emacs, with more explicit failure if that is not
  4294 installed (the old isabelle-interface script silently falls back on
  4295 isabelle-process).  The PROOFGENERAL_HOME setting determines the
  4296 installation location of the Proof General distribution.
  4297 
  4298 * Isabelle/lib/classes/Pure.jar provides basic support to integrate
  4299 the Isabelle process into a JVM/Scala application.  See
  4300 Isabelle/lib/jedit/plugin for a minimal example.  (The obsolete Java
  4301 process wrapper has been discontinued.)
  4302 
  4303 * Added homegrown Isabelle font with unicode layout, see lib/fonts.
  4304 
  4305 * Various status messages (with exact source position information) are
  4306 emitted, if proper markup print mode is enabled.  This allows
  4307 user-interface components to provide detailed feedback on internal
  4308 prover operations.
  4309 
  4310 
  4311 
  4312 New in Isabelle2008 (June 2008)
  4313 -------------------------------
  4314 
  4315 *** General ***
  4316 
  4317 * The Isabelle/Isar Reference Manual (isar-ref) has been reorganized
  4318 and updated, with formally checked references as hyperlinks.
  4319 
  4320 * Theory loader: use_thy (and similar operations) no longer set the
  4321 implicit ML context, which was occasionally hard to predict and in
  4322 conflict with concurrency.  INCOMPATIBILITY, use ML within Isar which
  4323 provides a proper context already.
  4324 
  4325 * Theory loader: old-style ML proof scripts being *attached* to a thy
  4326 file are no longer supported.  INCOMPATIBILITY, regular 'uses' and
  4327 'use' within a theory file will do the job.
  4328 
  4329 * Name space merge now observes canonical order, i.e. the second space
  4330 is inserted into the first one, while existing entries in the first
  4331 space take precedence.  INCOMPATIBILITY in rare situations, may try to
  4332 swap theory imports.
  4333 
  4334 * Syntax: symbol \<chi> is now considered a letter.  Potential
  4335 INCOMPATIBILITY in identifier syntax etc.
  4336 
  4337 * Outer syntax: string tokens no longer admit escaped white space,
  4338 which was an accidental (undocumented) feature.  INCOMPATIBILITY, use
  4339 white space without escapes.
  4340 
  4341 * Outer syntax: string tokens may contain arbitrary character codes
  4342 specified via 3 decimal digits (as in SML).  E.g. "foo\095bar" for
  4343 "foo_bar".
  4344 
  4345 
  4346 *** Pure ***
  4347 
  4348 * Context-dependent token translations.  Default setup reverts locally
  4349 fixed variables, and adds hilite markup for undeclared frees.
  4350 
  4351 * Unused theorems can be found using the new command 'unused_thms'.
  4352 There are three ways of invoking it:
  4353 
  4354 (1) unused_thms
  4355      Only finds unused theorems in the current theory.
  4356 
  4357 (2) unused_thms thy_1 ... thy_n -
  4358      Finds unused theorems in the current theory and all of its ancestors,
  4359      excluding the theories thy_1 ... thy_n and all of their ancestors.
  4360 
  4361 (3) unused_thms thy_1 ... thy_n - thy'_1 ... thy'_m
  4362      Finds unused theorems in the theories thy'_1 ... thy'_m and all of
  4363      their ancestors, excluding the theories thy_1 ... thy_n and all of
  4364      their ancestors.
  4365 
  4366 In order to increase the readability of the list produced by
  4367 unused_thms, theorems that have been created by a particular instance
  4368 of a theory command such as 'inductive' or 'function' are considered
  4369 to belong to the same "group", meaning that if at least one theorem in
  4370 this group is used, the other theorems in the same group are no longer
  4371 reported as unused.  Moreover, if all theorems in the group are
  4372 unused, only one theorem in the group is displayed.
  4373 
  4374 Note that proof objects have to be switched on in order for
  4375 unused_thms to work properly (i.e. !proofs must be >= 1, which is
  4376 usually the case when using Proof General with the default settings).
  4377 
  4378 * Authentic naming of facts disallows ad-hoc overwriting of previous
  4379 theorems within the same name space.  INCOMPATIBILITY, need to remove
  4380 duplicate fact bindings, or even accidental fact duplications.  Note
  4381 that tools may maintain dynamically scoped facts systematically, using
  4382 PureThy.add_thms_dynamic.
  4383 
  4384 * Command 'hide' now allows to hide from "fact" name space as well.
  4385 
  4386 * Eliminated destructive theorem database, simpset, claset, and
  4387 clasimpset.  Potential INCOMPATIBILITY, really need to observe linear
  4388 update of theories within ML code.
  4389 
  4390 * Eliminated theory ProtoPure and CPure, leaving just one Pure theory.
  4391 INCOMPATIBILITY, object-logics depending on former Pure require
  4392 additional setup PureThy.old_appl_syntax_setup; object-logics
  4393 depending on former CPure need to refer to Pure.
  4394 
  4395 * Commands 'use' and 'ML' are now purely functional, operating on
  4396 theory/local_theory.  Removed former 'ML_setup' (on theory), use 'ML'
  4397 instead.  Added 'ML_val' as mere diagnostic replacement for 'ML'.
  4398 INCOMPATIBILITY.
  4399 
  4400 * Command 'setup': discontinued implicit version with ML reference.
  4401 
  4402 * Instantiation target allows for simultaneous specification of class
  4403 instance operations together with an instantiation proof.
  4404 Type-checking phase allows to refer to class operations uniformly.
  4405 See src/HOL/Complex/Complex.thy for an Isar example and
  4406 src/HOL/Library/Eval.thy for an ML example.
  4407 
  4408 * Indexing of literal facts: be more serious about including only
  4409 facts from the visible specification/proof context, but not the
  4410 background context (locale etc.).  Affects `prop` notation and method
  4411 "fact".  INCOMPATIBILITY: need to name facts explicitly in rare
  4412 situations.
  4413 
  4414 * Method "cases", "induct", "coinduct": removed obsolete/undocumented
  4415 "(open)" option, which used to expose internal bound variables to the
  4416 proof text.
  4417 
  4418 * Isar statements: removed obsolete case "rule_context".
  4419 INCOMPATIBILITY, better use explicit fixes/assumes.
  4420 
  4421 * Locale proofs: default proof step now includes 'unfold_locales';
  4422 hence 'proof' without argument may be used to unfold locale
  4423 predicates.
  4424 
  4425 
  4426 *** Document preparation ***
  4427 
  4428 * Simplified pdfsetup.sty: color/hyperref is used unconditionally for
  4429 both pdf and dvi (hyperlinks usually work in xdvi as well); removed
  4430 obsolete thumbpdf setup (contemporary PDF viewers do this on the
  4431 spot); renamed link color from "darkblue" to "linkcolor" (default
  4432 value unchanged, can be redefined via \definecolor); no longer sets
  4433 "a4paper" option (unnecessary or even intrusive).
  4434 
  4435 * Antiquotation @{lemma A method} proves proposition A by the given
  4436 method (either a method name or a method name plus (optional) method
  4437 arguments in parentheses) and prints A just like @{prop A}.
  4438 
  4439 
  4440 *** HOL ***
  4441 
  4442 * New primrec package.  Specification syntax conforms in style to
  4443 definition/function/....  No separate induction rule is provided.  The
  4444 "primrec" command distinguishes old-style and new-style specifications
  4445 by syntax.  The former primrec package is now named OldPrimrecPackage.
  4446 When adjusting theories, beware: constants stemming from new-style
  4447 primrec specifications have authentic syntax.
  4448 
  4449 * Metis prover is now an order of magnitude faster, and also works
  4450 with multithreading.
  4451 
  4452 * Metis: the maximum number of clauses that can be produced from a
  4453 theorem is now given by the attribute max_clauses.  Theorems that
  4454 exceed this number are ignored, with a warning printed.
  4455 
  4456 * Sledgehammer no longer produces structured proofs by default. To
  4457 enable, declare [[sledgehammer_full = true]].  Attributes
  4458 reconstruction_modulus, reconstruction_sorts renamed
  4459 sledgehammer_modulus, sledgehammer_sorts.  INCOMPATIBILITY.
  4460 
  4461 * Method "induct_scheme" derives user-specified induction rules
  4462 from well-founded induction and completeness of patterns. This factors
  4463 out some operations that are done internally by the function package
  4464 and makes them available separately.  See
  4465 src/HOL/ex/Induction_Scheme.thy for examples.
  4466 
  4467 * More flexible generation of measure functions for termination
  4468 proofs: Measure functions can be declared by proving a rule of the
  4469 form "is_measure f" and giving it the [measure_function] attribute.
  4470 The "is_measure" predicate is logically meaningless (always true), and
  4471 just guides the heuristic.  To find suitable measure functions, the
  4472 termination prover sets up the goal "is_measure ?f" of the appropriate
  4473 type and generates all solutions by prolog-style backwards proof using
  4474 the declared rules.
  4475 
  4476 This setup also deals with rules like 
  4477 
  4478   "is_measure f ==> is_measure (list_size f)"
  4479 
  4480 which accommodates nested datatypes that recurse through lists.
  4481 Similar rules are predeclared for products and option types.
  4482 
  4483 * Turned the type of sets "'a set" into an abbreviation for "'a => bool"
  4484 
  4485   INCOMPATIBILITIES:
  4486 
  4487   - Definitions of overloaded constants on sets have to be replaced by
  4488     definitions on => and bool.
  4489 
  4490   - Some definitions of overloaded operators on sets can now be proved
  4491     using the definitions of the operators on => and bool.  Therefore,
  4492     the following theorems have been renamed:
  4493 
  4494       subset_def   -> subset_eq
  4495       psubset_def  -> psubset_eq
  4496       set_diff_def -> set_diff_eq
  4497       Compl_def    -> Compl_eq
  4498       Sup_set_def  -> Sup_set_eq
  4499       Inf_set_def  -> Inf_set_eq
  4500       sup_set_def  -> sup_set_eq
  4501       inf_set_def  -> inf_set_eq
  4502 
  4503   - Due to the incompleteness of the HO unification algorithm, some
  4504     rules such as subst may require manual instantiation, if some of
  4505     the unknowns in the rule is a set.
  4506 
  4507   - Higher order unification and forward proofs:
  4508     The proof pattern
  4509 
  4510       have "P (S::'a set)" <...>
  4511       then have "EX S. P S" ..
  4512 
  4513     no longer works (due to the incompleteness of the HO unification
  4514     algorithm) and must be replaced by the pattern
  4515 
  4516       have "EX S. P S"
  4517       proof
  4518         show "P S" <...>
  4519       qed
  4520 
  4521   - Calculational reasoning with subst (or similar rules):
  4522     The proof pattern
  4523 
  4524       have "P (S::'a set)" <...>
  4525       also have "S = T" <...>
  4526       finally have "P T" .
  4527 
  4528     no longer works (for similar reasons as the previous example) and
  4529     must be replaced by something like
  4530 
  4531       have "P (S::'a set)" <...>
  4532       moreover have "S = T" <...>
  4533       ultimately have "P T" by simp
  4534 
  4535   - Tactics or packages written in ML code:
  4536     Code performing pattern matching on types via
  4537 
  4538       Type ("set", [T]) => ...
  4539 
  4540     must be rewritten. Moreover, functions like strip_type or
  4541     binder_types no longer return the right value when applied to a
  4542     type of the form
  4543 
  4544       T1 => ... => Tn => U => bool
  4545 
  4546     rather than
  4547 
  4548       T1 => ... => Tn => U set
  4549 
  4550 * Merged theories Wellfounded_Recursion, Accessible_Part and
  4551 Wellfounded_Relations to theory Wellfounded.
  4552 
  4553 * Explicit class "eq" for executable equality.  INCOMPATIBILITY.
  4554 
  4555 * Class finite no longer treats UNIV as class parameter.  Use class
  4556 enum from theory Library/Enum instead to achieve a similar effect.
  4557 INCOMPATIBILITY.
  4558 
  4559 * Theory List: rule list_induct2 now has explicitly named cases "Nil"
  4560 and "Cons".  INCOMPATIBILITY.
  4561 
  4562 * HOL (and FOL): renamed variables in rules imp_elim and swap.
  4563 Potential INCOMPATIBILITY.
  4564 
  4565 * Theory Product_Type: duplicated lemmas split_Pair_apply and
  4566 injective_fst_snd removed, use split_eta and prod_eqI instead.
  4567 Renamed upd_fst to apfst and upd_snd to apsnd.  INCOMPATIBILITY.
  4568 
  4569 * Theory Nat: removed redundant lemmas that merely duplicate lemmas of
  4570 the same name in theory Orderings:
  4571 
  4572   less_trans
  4573   less_linear
  4574   le_imp_less_or_eq
  4575   le_less_trans
  4576   less_le_trans
  4577   less_not_sym
  4578   less_asym
  4579 
  4580 Renamed less_imp_le to less_imp_le_nat, and less_irrefl to
  4581 less_irrefl_nat.  Potential INCOMPATIBILITY due to more general types
  4582 and different variable names.
  4583 
  4584 * Library/Option_ord.thy: Canonical order on option type.
  4585 
  4586 * Library/RBT.thy: Red-black trees, an efficient implementation of
  4587 finite maps.
  4588 
  4589 * Library/Countable.thy: Type class for countable types.
  4590 
  4591 * Theory Int: The representation of numerals has changed.  The infix
  4592 operator BIT and the bit datatype with constructors B0 and B1 have
  4593 disappeared.  INCOMPATIBILITY, use "Int.Bit0 x" and "Int.Bit1 y" in
  4594 place of "x BIT bit.B0" and "y BIT bit.B1", respectively.  Theorems
  4595 involving BIT, B0, or B1 have been renamed with "Bit0" or "Bit1"
  4596 accordingly.
  4597 
  4598 * Theory Nat: definition of <= and < on natural numbers no longer
  4599 depend on well-founded relations.  INCOMPATIBILITY.  Definitions
  4600 le_def and less_def have disappeared.  Consider lemmas not_less
  4601 [symmetric, where ?'a = nat] and less_eq [symmetric] instead.
  4602 
  4603 * Theory Finite_Set: locales ACf, ACe, ACIf, ACIfSL and ACIfSLlin
  4604 (whose purpose mainly is for various fold_set functionals) have been
  4605 abandoned in favor of the existing algebraic classes
  4606 ab_semigroup_mult, comm_monoid_mult, ab_semigroup_idem_mult,
  4607 lower_semilattice (resp. upper_semilattice) and linorder.
  4608 INCOMPATIBILITY.
  4609 
  4610 * Theory Transitive_Closure: induct and cases rules now declare proper
  4611 case_names ("base" and "step").  INCOMPATIBILITY.
  4612 
  4613 * Theorem Inductive.lfp_ordinal_induct generalized to complete
  4614 lattices.  The form set-specific version is available as
  4615 Inductive.lfp_ordinal_induct_set.
  4616 
  4617 * Renamed theorems "power.simps" to "power_int.simps".
  4618 INCOMPATIBILITY.
  4619 
  4620 * Class semiring_div provides basic abstract properties of semirings
  4621 with division and modulo operations.  Subsumes former class dvd_mod.
  4622 
  4623 * Merged theories IntDef, Numeral and IntArith into unified theory
  4624 Int.  INCOMPATIBILITY.
  4625 
  4626 * Theory Library/Code_Index: type "index" now represents natural
  4627 numbers rather than integers.  INCOMPATIBILITY.
  4628 
  4629 * New class "uminus" with operation "uminus" (split of from class
  4630 "minus" which now only has operation "minus", binary).
  4631 INCOMPATIBILITY.
  4632 
  4633 * Constants "card", "internal_split", "option_map" now with authentic
  4634 syntax.  INCOMPATIBILITY.
  4635 
  4636 * Definitions subset_def, psubset_def, set_diff_def, Compl_def,
  4637 le_bool_def, less_bool_def, le_fun_def, less_fun_def, inf_bool_def,
  4638 sup_bool_def, Inf_bool_def, Sup_bool_def, inf_fun_def, sup_fun_def,
  4639 Inf_fun_def, Sup_fun_def, inf_set_def, sup_set_def, Inf_set_def,
  4640 Sup_set_def, le_def, less_def, option_map_def now with object
  4641 equality.  INCOMPATIBILITY.
  4642 
  4643 * Records. Removed K_record, and replaced it by pure lambda term
  4644 %x. c. The simplifier setup is now more robust against eta expansion.
  4645 INCOMPATIBILITY: in cases explicitly referring to K_record.
  4646 
  4647 * Library/Multiset: {#a, b, c#} abbreviates {#a#} + {#b#} + {#c#}.
  4648 
  4649 * Library/ListVector: new theory of arithmetic vector operations.
  4650 
  4651 * Library/Order_Relation: new theory of various orderings as sets of
  4652 pairs.  Defines preorders, partial orders, linear orders and
  4653 well-orders on sets and on types.
  4654 
  4655 
  4656 *** ZF ***
  4657 
  4658 * Renamed some theories to allow to loading both ZF and HOL in the
  4659 same session:
  4660 
  4661   Datatype  -> Datatype_ZF
  4662   Inductive -> Inductive_ZF
  4663   Int       -> Int_ZF
  4664   IntDiv    -> IntDiv_ZF
  4665   Nat       -> Nat_ZF
  4666   List      -> List_ZF
  4667   Main      -> Main_ZF
  4668 
  4669 INCOMPATIBILITY: ZF theories that import individual theories below
  4670 Main might need to be adapted.  Regular theory Main is still
  4671 available, as trivial extension of Main_ZF.
  4672 
  4673 
  4674 *** ML ***
  4675 
  4676 * ML within Isar: antiquotation @{const name} or @{const
  4677 name(typargs)} produces statically-checked Const term.
  4678 
  4679 * Functor NamedThmsFun: data is available to the user as dynamic fact
  4680 (of the same name).  Removed obsolete print command.
  4681 
  4682 * Removed obsolete "use_legacy_bindings" function.
  4683 
  4684 * The ``print mode'' is now a thread-local value derived from a global
  4685 template (the former print_mode reference), thus access becomes
  4686 non-critical.  The global print_mode reference is for session
  4687 management only; user-code should use print_mode_value,
  4688 print_mode_active, PrintMode.setmp etc.  INCOMPATIBILITY.
  4689 
  4690 * Functions system/system_out provide a robust way to invoke external
  4691 shell commands, with propagation of interrupts (requires Poly/ML
  4692 5.2.1).  Do not use OS.Process.system etc. from the basis library!
  4693 
  4694 
  4695 *** System ***
  4696 
  4697 * Default settings: PROOFGENERAL_OPTIONS no longer impose xemacs ---
  4698 in accordance with Proof General 3.7, which prefers GNU emacs.
  4699 
  4700 * isatool tty runs Isabelle process with plain tty interaction;
  4701 optional line editor may be specified via ISABELLE_LINE_EDITOR
  4702 setting, the default settings attempt to locate "ledit" and "rlwrap".
  4703 
  4704 * isatool browser now works with Cygwin as well, using general
  4705 "javapath" function defined in Isabelle process environment.
  4706 
  4707 * YXML notation provides a simple and efficient alternative to
  4708 standard XML transfer syntax.  See src/Pure/General/yxml.ML and
  4709 isatool yxml as described in the Isabelle system manual.
  4710 
  4711 * JVM class isabelle.IsabelleProcess (located in Isabelle/lib/classes)
  4712 provides general wrapper for managing an Isabelle process in a robust
  4713 fashion, with ``cooked'' output from stdin/stderr.
  4714 
  4715 * Rudimentary Isabelle plugin for jEdit (see Isabelle/lib/jedit),
  4716 based on Isabelle/JVM process wrapper (see Isabelle/lib/classes).
  4717 
  4718 * Removed obsolete THIS_IS_ISABELLE_BUILD feature.  NB: the documented
  4719 way of changing the user's settings is via
  4720 ISABELLE_HOME_USER/etc/settings, which is a fully featured bash
  4721 script.
  4722 
  4723 * Multithreading.max_threads := 0 refers to the number of actual CPU
  4724 cores of the underlying machine, which is a good starting point for
  4725 optimal performance tuning.  The corresponding usedir option -M allows
  4726 "max" as an alias for "0".  WARNING: does not work on certain versions
  4727 of Mac OS (with Poly/ML 5.1).
  4728 
  4729 * isabelle-process: non-ML sessions are run with "nice", to reduce the
  4730 adverse effect of Isabelle flooding interactive front-ends (notably
  4731 ProofGeneral / XEmacs).
  4732 
  4733 
  4734 
  4735 New in Isabelle2007 (November 2007)
  4736 -----------------------------------
  4737 
  4738 *** General ***
  4739 
  4740 * More uniform information about legacy features, notably a
  4741 warning/error of "Legacy feature: ...", depending on the state of the
  4742 tolerate_legacy_features flag (default true). FUTURE INCOMPATIBILITY:
  4743 legacy features will disappear eventually.
  4744 
  4745 * Theory syntax: the header format ``theory A = B + C:'' has been
  4746 discontinued in favour of ``theory A imports B C begin''.  Use isatool
  4747 fixheaders to convert existing theory files.  INCOMPATIBILITY.
  4748 
  4749 * Theory syntax: the old non-Isar theory file format has been
  4750 discontinued altogether.  Note that ML proof scripts may still be used
  4751 with Isar theories; migration is usually quite simple with the ML
  4752 function use_legacy_bindings.  INCOMPATIBILITY.
  4753 
  4754 * Theory syntax: some popular names (e.g. 'class', 'declaration',
  4755 'fun', 'help', 'if') are now keywords.  INCOMPATIBILITY, use double
  4756 quotes.
  4757 
  4758 * Theory loader: be more serious about observing the static theory
  4759 header specifications (including optional directories), but not the
  4760 accidental file locations of previously successful loads.  The strict
  4761 update policy of former update_thy is now already performed by
  4762 use_thy, so the former has been removed; use_thys updates several
  4763 theories simultaneously, just as 'imports' within a theory header
  4764 specification, but without merging the results.  Potential
  4765 INCOMPATIBILITY: may need to refine theory headers and commands
  4766 ROOT.ML which depend on load order.
  4767 
  4768 * Theory loader: optional support for content-based file
  4769 identification, instead of the traditional scheme of full physical
  4770 path plus date stamp; configured by the ISABELLE_FILE_IDENT setting
  4771 (cf. the system manual).  The new scheme allows to work with
  4772 non-finished theories in persistent session images, such that source
  4773 files may be moved later on without requiring reloads.
  4774 
  4775 * Theory loader: old-style ML proof scripts being *attached* to a thy
  4776 file (with the same base name as the theory) are considered a legacy
  4777 feature, which will disappear eventually. Even now, the theory loader
  4778 no longer maintains dependencies on such files.
  4779 
  4780 * Syntax: the scope for resolving ambiguities via type-inference is
  4781 now limited to individual terms, instead of whole simultaneous
  4782 specifications as before. This greatly reduces the complexity of the
  4783 syntax module and improves flexibility by separating parsing and
  4784 type-checking. INCOMPATIBILITY: additional type-constraints (explicit
  4785 'fixes' etc.) are required in rare situations.
  4786 
  4787 * Syntax: constants introduced by new-style packages ('definition',
  4788 'abbreviation' etc.) are passed through the syntax module in
  4789 ``authentic mode''. This means that associated mixfix annotations
  4790 really stick to such constants, independently of potential name space
  4791 ambiguities introduced later on. INCOMPATIBILITY: constants in parse
  4792 trees are represented slightly differently, may need to adapt syntax
  4793 translations accordingly. Use CONST marker in 'translations' and
  4794 @{const_syntax} antiquotation in 'parse_translation' etc.
  4795 
  4796 * Legacy goal package: reduced interface to the bare minimum required
  4797 to keep existing proof scripts running.  Most other user-level
  4798 functions are now part of the OldGoals structure, which is *not* open
  4799 by default (consider isatool expandshort before open OldGoals).
  4800 Removed top_sg, prin, printyp, pprint_term/typ altogether, because
  4801 these tend to cause confusion about the actual goal (!) context being
  4802 used here, which is not necessarily the same as the_context().
  4803 
  4804 * Command 'find_theorems': supports "*" wild-card in "name:"
  4805 criterion; "with_dups" option.  Certain ProofGeneral versions might
  4806 support a specific search form (see ProofGeneral/CHANGES).
  4807 
  4808 * The ``prems limit'' option (cf. ProofContext.prems_limit) is now -1
  4809 by default, which means that "prems" (and also "fixed variables") are
  4810 suppressed from proof state output.  Note that the ProofGeneral
  4811 settings mechanism allows to change and save options persistently, but
  4812 older versions of Isabelle will fail to start up if a negative prems
  4813 limit is imposed.
  4814 
  4815 * Local theory targets may be specified by non-nested blocks of
  4816 ``context/locale/class ... begin'' followed by ``end''.  The body may
  4817 contain definitions, theorems etc., including any derived mechanism
  4818 that has been implemented on top of these primitives.  This concept
  4819 generalizes the existing ``theorem (in ...)'' towards more versatility
  4820 and scalability.
  4821 
  4822 * Proof General interface: proper undo of final 'end' command;
  4823 discontinued Isabelle/classic mode (ML proof scripts).
  4824 
  4825 
  4826 *** Document preparation ***
  4827 
  4828 * Added antiquotation @{theory name} which prints the given name,
  4829 after checking that it refers to a valid ancestor theory in the
  4830 current context.
  4831 
  4832 * Added antiquotations @{ML_type text} and @{ML_struct text} which
  4833 check the given source text as ML type/structure, printing verbatim.
  4834 
  4835 * Added antiquotation @{abbrev "c args"} which prints the abbreviation
  4836 "c args == rhs" given in the current context.  (Any number of
  4837 arguments may be given on the LHS.)
  4838 
  4839 
  4840 *** Pure ***
  4841 
  4842 * The 'class' package offers a combination of axclass and locale to
  4843 achieve Haskell-like type classes in Isabelle.  Definitions and
  4844 theorems within a class context produce both relative results (with
  4845 implicit parameters according to the locale context), and polymorphic
  4846 constants with qualified polymorphism (according to the class
  4847 context).  Within the body context of a 'class' target, a separate
  4848 syntax layer ("user space type system") takes care of converting
  4849 between global polymorphic consts and internal locale representation.
  4850 See src/HOL/ex/Classpackage.thy for examples (as well as main HOL).
  4851 "isatool doc classes" provides a tutorial.
  4852 
  4853 * Generic code generator framework allows to generate executable
  4854 code for ML and Haskell (including Isabelle classes).  A short usage
  4855 sketch:
  4856 
  4857     internal compilation:
  4858         export_code <list of constants (term syntax)> in SML
  4859     writing SML code to a file:
  4860         export_code <list of constants (term syntax)> in SML <filename>
  4861     writing OCaml code to a file:
  4862         export_code <list of constants (term syntax)> in OCaml <filename>
  4863     writing Haskell code to a bunch of files:
  4864         export_code <list of constants (term syntax)> in Haskell <filename>
  4865 
  4866     evaluating closed propositions to True/False using code generation:
  4867         method ``eval''
  4868 
  4869 Reasonable default setup of framework in HOL.
  4870 
  4871 Theorem attributs for selecting and transforming function equations theorems:
  4872 
  4873     [code fun]:        select a theorem as function equation for a specific constant
  4874     [code fun del]:    deselect a theorem as function equation for a specific constant
  4875     [code inline]:     select an equation theorem for unfolding (inlining) in place
  4876     [code inline del]: deselect an equation theorem for unfolding (inlining) in place
  4877 
  4878 User-defined serializations (target in {SML, OCaml, Haskell}):
  4879 
  4880     code_const <and-list of constants (term syntax)>
  4881       {(target) <and-list of const target syntax>}+
  4882 
  4883     code_type <and-list of type constructors>
  4884       {(target) <and-list of type target syntax>}+
  4885 
  4886     code_instance <and-list of instances>
  4887       {(target)}+
  4888         where instance ::= <type constructor> :: <class>
  4889 
  4890     code_class <and_list of classes>
  4891       {(target) <and-list of class target syntax>}+
  4892         where class target syntax ::= <class name> {where {<classop> == <target syntax>}+}?
  4893 
  4894 code_instance and code_class only are effective to target Haskell.
  4895 
  4896 For example usage see src/HOL/ex/Codegenerator.thy and
  4897 src/HOL/ex/Codegenerator_Pretty.thy.  A separate tutorial on code
  4898 generation from Isabelle/HOL theories is available via "isatool doc
  4899 codegen".
  4900 
  4901 * Code generator: consts in 'consts_code' Isar commands are now
  4902 referred to by usual term syntax (including optional type
  4903 annotations).
  4904 
  4905 * Command 'no_translations' removes translation rules from theory
  4906 syntax.
  4907 
  4908 * Overloaded definitions are now actually checked for acyclic
  4909 dependencies.  The overloading scheme is slightly more general than
  4910 that of Haskell98, although Isabelle does not demand an exact
  4911 correspondence to type class and instance declarations.
  4912 INCOMPATIBILITY, use ``defs (unchecked overloaded)'' to admit more
  4913 exotic versions of overloading -- at the discretion of the user!
  4914 
  4915 Polymorphic constants are represented via type arguments, i.e. the
  4916 instantiation that matches an instance against the most general
  4917 declaration given in the signature.  For example, with the declaration
  4918 c :: 'a => 'a => 'a, an instance c :: nat => nat => nat is represented
  4919 as c(nat).  Overloading is essentially simultaneous structural
  4920 recursion over such type arguments.  Incomplete specification patterns
  4921 impose global constraints on all occurrences, e.g. c('a * 'a) on the
  4922 LHS means that more general c('a * 'b) will be disallowed on any RHS.
  4923 Command 'print_theory' outputs the normalized system of recursive
  4924 equations, see section "definitions".
  4925 
  4926 * Configuration options are maintained within the theory or proof
  4927 context (with name and type bool/int/string), providing a very simple
  4928 interface to a poor-man's version of general context data.  Tools may
  4929 declare options in ML (e.g. using Attrib.config_int) and then refer to
  4930 these values using Config.get etc.  Users may change options via an
  4931 associated attribute of the same name.  This form of context
  4932 declaration works particularly well with commands 'declare' or
  4933 'using', for example ``declare [[foo = 42]]''.  Thus it has become
  4934 very easy to avoid global references, which would not observe Isar
  4935 toplevel undo/redo and fail to work with multithreading.
  4936 
  4937 Various global ML references of Pure and HOL have been turned into
  4938 configuration options:
  4939 
  4940   Unify.search_bound		unify_search_bound
  4941   Unify.trace_bound		unify_trace_bound
  4942   Unify.trace_simp		unify_trace_simp
  4943   Unify.trace_types		unify_trace_types
  4944   Simplifier.simp_depth_limit	simp_depth_limit
  4945   Blast.depth_limit		blast_depth_limit
  4946   DatatypeProp.dtK		datatype_distinctness_limit
  4947   fast_arith_neq_limit  	fast_arith_neq_limit
  4948   fast_arith_split_limit	fast_arith_split_limit
  4949 
  4950 * Named collections of theorems may be easily installed as context
  4951 data using the functor NamedThmsFun (see also
  4952 src/Pure/Tools/named_thms.ML).  The user may add or delete facts via
  4953 attributes; there is also a toplevel print command.  This facility is
  4954 just a common case of general context data, which is the preferred way
  4955 for anything more complex than just a list of facts in canonical
  4956 order.
  4957 
  4958 * Isar: command 'declaration' augments a local theory by generic
  4959 declaration functions written in ML.  This enables arbitrary content
  4960 being added to the context, depending on a morphism that tells the
  4961 difference of the original declaration context wrt. the application
  4962 context encountered later on.
  4963 
  4964 * Isar: proper interfaces for simplification procedures.  Command
  4965 'simproc_setup' declares named simprocs (with match patterns, and body
  4966 text in ML).  Attribute "simproc" adds/deletes simprocs in the current
  4967 context.  ML antiquotation @{simproc name} retrieves named simprocs.
  4968 
  4969 * Isar: an extra pair of brackets around attribute declarations
  4970 abbreviates a theorem reference involving an internal dummy fact,
  4971 which will be ignored later --- only the effect of the attribute on
  4972 the background context will persist.  This form of in-place
  4973 declarations is particularly useful with commands like 'declare' and
  4974 'using', for example ``have A using [[simproc a]] by simp''.
  4975 
  4976 * Isar: method "assumption" (and implicit closing of subproofs) now
  4977 takes simple non-atomic goal assumptions into account: after applying
  4978 an assumption as a rule the resulting subgoals are solved by atomic
  4979 assumption steps.  This is particularly useful to finish 'obtain'
  4980 goals, such as "!!x. (!!x. P x ==> thesis) ==> P x ==> thesis",
  4981 without referring to the original premise "!!x. P x ==> thesis" in the
  4982 Isar proof context.  POTENTIAL INCOMPATIBILITY: method "assumption" is
  4983 more permissive.
  4984 
  4985 * Isar: implicit use of prems from the Isar proof context is
  4986 considered a legacy feature.  Common applications like ``have A .''
  4987 may be replaced by ``have A by fact'' or ``note `A`''.  In general,
  4988 referencing facts explicitly here improves readability and
  4989 maintainability of proof texts.
  4990 
  4991 * Isar: improper proof element 'guess' is like 'obtain', but derives
  4992 the obtained context from the course of reasoning!  For example:
  4993 
  4994   assume "EX x y. A x & B y"   -- "any previous fact"
  4995   then guess x and y by clarify
  4996 
  4997 This technique is potentially adventurous, depending on the facts and
  4998 proof tools being involved here.
  4999 
  5000 * Isar: known facts from the proof context may be specified as literal
  5001 propositions, using ASCII back-quote syntax.  This works wherever
  5002 named facts used to be allowed so far, in proof commands, proof
  5003 methods, attributes etc.  Literal facts are retrieved from the context
  5004 according to unification of type and term parameters.  For example,
  5005 provided that "A" and "A ==> B" and "!!x. P x ==> Q x" are known
  5006 theorems in the current context, then these are valid literal facts:
  5007 `A` and `A ==> B` and `!!x. P x ==> Q x" as well as `P a ==> Q a` etc.
  5008 
  5009 There is also a proof method "fact" which does the same composition
  5010 for explicit goal states, e.g. the following proof texts coincide with
  5011 certain special cases of literal facts:
  5012 
  5013   have "A" by fact                 ==  note `A`
  5014   have "A ==> B" by fact           ==  note `A ==> B`
  5015   have "!!x. P x ==> Q x" by fact  ==  note `!!x. P x ==> Q x`
  5016   have "P a ==> Q a" by fact       ==  note `P a ==> Q a`
  5017 
  5018 * Isar: ":" (colon) is no longer a symbolic identifier character in
  5019 outer syntax.  Thus symbolic identifiers may be used without
  5020 additional white space in declarations like this: ``assume *: A''.
  5021 
  5022 * Isar: 'print_facts' prints all local facts of the current context,
  5023 both named and unnamed ones.
  5024 
  5025 * Isar: 'def' now admits simultaneous definitions, e.g.:
  5026 
  5027   def x == "t" and y == "u"
  5028 
  5029 * Isar: added command 'unfolding', which is structurally similar to
  5030 'using', but affects both the goal state and facts by unfolding given
  5031 rewrite rules.  Thus many occurrences of the 'unfold' method or
  5032 'unfolded' attribute may be replaced by first-class proof text.
  5033 
  5034 * Isar: methods 'unfold' / 'fold', attributes 'unfolded' / 'folded',
  5035 and command 'unfolding' now all support object-level equalities
  5036 (potentially conditional).  The underlying notion of rewrite rule is
  5037 analogous to the 'rule_format' attribute, but *not* that of the
  5038 Simplifier (which is usually more generous).
  5039 
  5040 * Isar: the new attribute [rotated n] (default n = 1) rotates the
  5041 premises of a theorem by n. Useful in conjunction with drule.
  5042 
  5043 * Isar: the goal restriction operator [N] (default N = 1) evaluates a
  5044 method expression within a sandbox consisting of the first N
  5045 sub-goals, which need to exist.  For example, ``simp_all [3]''
  5046 simplifies the first three sub-goals, while (rule foo, simp_all)[]
  5047 simplifies all new goals that emerge from applying rule foo to the
  5048 originally first one.
  5049 
  5050 * Isar: schematic goals are no longer restricted to higher-order
  5051 patterns; e.g. ``lemma "?P(?x)" by (rule TrueI)'' now works as
  5052 expected.
  5053 
  5054 * Isar: the conclusion of a long theorem statement is now either
  5055 'shows' (a simultaneous conjunction, as before), or 'obtains'
  5056 (essentially a disjunction of cases with local parameters and
  5057 assumptions).  The latter allows to express general elimination rules
  5058 adequately; in this notation common elimination rules look like this:
  5059 
  5060   lemma exE:    -- "EX x. P x ==> (!!x. P x ==> thesis) ==> thesis"
  5061     assumes "EX x. P x"
  5062     obtains x where "P x"
  5063 
  5064   lemma conjE:  -- "A & B ==> (A ==> B ==> thesis) ==> thesis"
  5065     assumes "A & B"
  5066     obtains A and B
  5067 
  5068   lemma disjE:  -- "A | B ==> (A ==> thesis) ==> (B ==> thesis) ==> thesis"
  5069     assumes "A | B"
  5070     obtains
  5071       A
  5072     | B
  5073 
  5074 The subsequent classical rules even refer to the formal "thesis"
  5075 explicitly:
  5076 
  5077   lemma classical:     -- "(~ thesis ==> thesis) ==> thesis"
  5078     obtains "~ thesis"
  5079 
  5080   lemma Peirce's_Law:  -- "((thesis ==> something) ==> thesis) ==> thesis"
  5081     obtains "thesis ==> something"
  5082 
  5083 The actual proof of an 'obtains' statement is analogous to that of the
  5084 Isar proof element 'obtain', only that there may be several cases.
  5085 Optional case names may be specified in parentheses; these will be
  5086 available both in the present proof and as annotations in the
  5087 resulting rule, for later use with the 'cases' method (cf. attribute
  5088 case_names).
  5089 
  5090 * Isar: the assumptions of a long theorem statement are available as
  5091 "assms" fact in the proof context.  This is more appropriate than the
  5092 (historical) "prems", which refers to all assumptions of the current
  5093 context, including those from the target locale, proof body etc.
  5094 
  5095 * Isar: 'print_statement' prints theorems from the current theory or
  5096 proof context in long statement form, according to the syntax of a
  5097 top-level lemma.
  5098 
  5099 * Isar: 'obtain' takes an optional case name for the local context
  5100 introduction rule (default "that").
  5101 
  5102 * Isar: removed obsolete 'concl is' patterns.  INCOMPATIBILITY, use
  5103 explicit (is "_ ==> ?foo") in the rare cases where this still happens
  5104 to occur.
  5105 
  5106 * Pure: syntax "CONST name" produces a fully internalized constant
  5107 according to the current context.  This is particularly useful for
  5108 syntax translations that should refer to internal constant
  5109 representations independently of name spaces.
  5110 
  5111 * Pure: syntax constant for foo (binder "FOO ") is called "foo_binder"
  5112 instead of "FOO ". This allows multiple binder declarations to coexist
  5113 in the same context.  INCOMPATIBILITY.
  5114 
  5115 * Isar/locales: 'notation' provides a robust interface to the 'syntax'
  5116 primitive that also works in a locale context (both for constants and
  5117 fixed variables). Type declaration and internal syntactic representation
  5118 of given constants retrieved from the context. Likewise, the
  5119 'no_notation' command allows to remove given syntax annotations from the
  5120 current context.
  5121 
  5122 * Isar/locales: new derived specification elements 'axiomatization',
  5123 'definition', 'abbreviation', which support type-inference, admit
  5124 object-level specifications (equality, equivalence).  See also the
  5125 isar-ref manual.  Examples:
  5126 
  5127   axiomatization
  5128     eq  (infix "===" 50) where
  5129     eq_refl: "x === x" and eq_subst: "x === y ==> P x ==> P y"
  5130 
  5131   definition "f x y = x + y + 1"
  5132   definition g where "g x = f x x"
  5133 
  5134   abbreviation
  5135     neq  (infix "=!=" 50) where
  5136     "x =!= y == ~ (x === y)"
  5137 
  5138 These specifications may be also used in a locale context.  Then the
  5139 constants being introduced depend on certain fixed parameters, and the
  5140 constant name is qualified by the locale base name.  An internal
  5141 abbreviation takes care for convenient input and output, making the
  5142 parameters implicit and using the original short name.  See also
  5143 src/HOL/ex/Abstract_NAT.thy for an example of deriving polymorphic
  5144 entities from a monomorphic theory.
  5145 
  5146 Presently, abbreviations are only available 'in' a target locale, but
  5147 not inherited by general import expressions.  Also note that
  5148 'abbreviation' may be used as a type-safe replacement for 'syntax' +
  5149 'translations' in common applications.  The "no_abbrevs" print mode
  5150 prevents folding of abbreviations in term output.
  5151 
  5152 Concrete syntax is attached to specified constants in internal form,
  5153 independently of name spaces.  The parse tree representation is
  5154 slightly different -- use 'notation' instead of raw 'syntax', and
  5155 'translations' with explicit "CONST" markup to accommodate this.
  5156 
  5157 * Pure/Isar: unified syntax for new-style specification mechanisms
  5158 (e.g.  'definition', 'abbreviation', or 'inductive' in HOL) admits
  5159 full type inference and dummy patterns ("_").  For example:
  5160 
  5161   definition "K x _ = x"
  5162 
  5163   inductive conj for A B
  5164   where "A ==> B ==> conj A B"
  5165 
  5166 * Pure: command 'print_abbrevs' prints all constant abbreviations of
  5167 the current context.  Print mode "no_abbrevs" prevents inversion of
  5168 abbreviations on output.
  5169 
  5170 * Isar/locales: improved parameter handling: use of locales "var" and
  5171 "struct" no longer necessary; - parameter renamings are no longer
  5172 required to be injective.  For example, this allows to define
  5173 endomorphisms as locale endom = homom mult mult h.
  5174 
  5175 * Isar/locales: changed the way locales with predicates are defined.
  5176 Instead of accumulating the specification, the imported expression is
  5177 now an interpretation.  INCOMPATIBILITY: different normal form of
  5178 locale expressions.  In particular, in interpretations of locales with
  5179 predicates, goals repesenting already interpreted fragments are not
  5180 removed automatically.  Use methods `intro_locales' and
  5181 `unfold_locales'; see below.
  5182 
  5183 * Isar/locales: new methods `intro_locales' and `unfold_locales'
  5184 provide backward reasoning on locales predicates.  The methods are
  5185 aware of interpretations and discharge corresponding goals.
  5186 `intro_locales' is less aggressive then `unfold_locales' and does not
  5187 unfold predicates to assumptions.
  5188 
  5189 * Isar/locales: the order in which locale fragments are accumulated
  5190 has changed.  This enables to override declarations from fragments due
  5191 to interpretations -- for example, unwanted simp rules.
  5192 
  5193 * Isar/locales: interpretation in theories and proof contexts has been
  5194 extended.  One may now specify (and prove) equations, which are
  5195 unfolded in interpreted theorems.  This is useful for replacing
  5196 defined concepts (constants depending on locale parameters) by
  5197 concepts already existing in the target context.  Example:
  5198 
  5199   interpretation partial_order ["op <= :: [int, int] => bool"]
  5200     where "partial_order.less (op <=) (x::int) y = (x < y)"
  5201 
  5202 Typically, the constant `partial_order.less' is created by a
  5203 definition specification element in the context of locale
  5204 partial_order.
  5205 
  5206 * Method "induct": improved internal context management to support
  5207 local fixes and defines on-the-fly. Thus explicit meta-level
  5208 connectives !!  and ==> are rarely required anymore in inductive goals
  5209 (using object-logic connectives for this purpose has been long
  5210 obsolete anyway). Common proof patterns are explained in
  5211 src/HOL/Induct/Common_Patterns.thy, see also
  5212 src/HOL/Isar_examples/Puzzle.thy and src/HOL/Lambda for realistic
  5213 examples.
  5214 
  5215 * Method "induct": improved handling of simultaneous goals. Instead of
  5216 introducing object-level conjunction, the statement is now split into
  5217 several conclusions, while the corresponding symbolic cases are nested
  5218 accordingly. INCOMPATIBILITY, proofs need to be structured explicitly,
  5219 see src/HOL/Induct/Common_Patterns.thy, for example.
  5220 
  5221 * Method "induct": mutual induction rules are now specified as a list
  5222 of rule sharing the same induction cases. HOL packages usually provide
  5223 foo_bar.inducts for mutually defined items foo and bar (e.g. inductive
  5224 predicates/sets or datatypes). INCOMPATIBILITY, users need to specify
  5225 mutual induction rules differently, i.e. like this:
  5226 
  5227   (induct rule: foo_bar.inducts)
  5228   (induct set: foo bar)
  5229   (induct pred: foo bar)
  5230   (induct type: foo bar)
  5231 
  5232 The ML function ProjectRule.projections turns old-style rules into the
  5233 new format.
  5234 
  5235 * Method "coinduct": dual of induction, see
  5236 src/HOL/Library/Coinductive_List.thy for various examples.
  5237 
  5238 * Method "cases", "induct", "coinduct": the ``(open)'' option is
  5239 considered a legacy feature.
  5240 
  5241 * Attribute "symmetric" produces result with standardized schematic
  5242 variables (index 0).  Potential INCOMPATIBILITY.
  5243 
  5244 * Simplifier: by default the simplifier trace only shows top level
  5245 rewrites now. That is, trace_simp_depth_limit is set to 1 by
  5246 default. Thus there is less danger of being flooded by the trace. The
  5247 trace indicates where parts have been suppressed.
  5248   
  5249 * Provers/classical: removed obsolete classical version of elim_format
  5250 attribute; classical elim/dest rules are now treated uniformly when
  5251 manipulating the claset.
  5252 
  5253 * Provers/classical: stricter checks to ensure that supplied intro,
  5254 dest and elim rules are well-formed; dest and elim rules must have at
  5255 least one premise.
  5256 
  5257 * Provers/classical: attributes dest/elim/intro take an optional
  5258 weight argument for the rule (just as the Pure versions).  Weights are
  5259 ignored by automated tools, but determine the search order of single
  5260 rule steps.
  5261 
  5262 * Syntax: input syntax now supports dummy variable binding "%_. b",
  5263 where the body does not mention the bound variable.  Note that dummy
  5264 patterns implicitly depend on their context of bounds, which makes
  5265 "{_. _}" match any set comprehension as expected.  Potential
  5266 INCOMPATIBILITY -- parse translations need to cope with syntactic
  5267 constant "_idtdummy" in the binding position.
  5268 
  5269 * Syntax: removed obsolete syntactic constant "_K" and its associated
  5270 parse translation.  INCOMPATIBILITY -- use dummy abstraction instead,
  5271 for example "A -> B" => "Pi A (%_. B)".
  5272 
  5273 * Pure: 'class_deps' command visualizes the subclass relation, using
  5274 the graph browser tool.
  5275 
  5276 * Pure: 'print_theory' now suppresses certain internal declarations by
  5277 default; use '!' option for full details.
  5278 
  5279 
  5280 *** HOL ***
  5281 
  5282 * Method "metis" proves goals by applying the Metis general-purpose
  5283 resolution prover (see also http://gilith.com/software/metis/).
  5284 Examples are in the directory MetisExamples.  WARNING: the
  5285 Isabelle/HOL-Metis integration does not yet work properly with
  5286 multi-threading.
  5287   
  5288 * Command 'sledgehammer' invokes external automatic theorem provers as
  5289 background processes.  It generates calls to the "metis" method if
  5290 successful. These can be pasted into the proof.  Users do not have to
  5291 wait for the automatic provers to return.  WARNING: does not really
  5292 work with multi-threading.
  5293 
  5294 * New "auto_quickcheck" feature tests outermost goal statements for
  5295 potential counter-examples.  Controlled by ML references
  5296 auto_quickcheck (default true) and auto_quickcheck_time_limit (default
  5297 5000 milliseconds).  Fails silently if statements is outside of
  5298 executable fragment, or any other codgenerator problem occurs.
  5299 
  5300 * New constant "undefined" with axiom "undefined x = undefined".
  5301 
  5302 * Added class "HOL.eq", allowing for code generation with polymorphic
  5303 equality.
  5304 
  5305 * Some renaming of class constants due to canonical name prefixing in
  5306 the new 'class' package:
  5307 
  5308     HOL.abs ~> HOL.abs_class.abs
  5309     HOL.divide ~> HOL.divide_class.divide
  5310     0 ~> HOL.zero_class.zero
  5311     1 ~> HOL.one_class.one
  5312     op + ~> HOL.plus_class.plus
  5313     op - ~> HOL.minus_class.minus
  5314     uminus ~> HOL.minus_class.uminus
  5315     op * ~> HOL.times_class.times
  5316     op < ~> HOL.ord_class.less
  5317     op <= > HOL.ord_class.less_eq
  5318     Nat.power ~> Power.power_class.power
  5319     Nat.size ~> Nat.size_class.size
  5320     Numeral.number_of ~> Numeral.number_class.number_of
  5321     FixedPoint.Inf ~> Lattices.complete_lattice_class.Inf
  5322     FixedPoint.Sup ~> Lattices.complete_lattice_class.Sup
  5323     Orderings.min ~> Orderings.ord_class.min
  5324     Orderings.max ~> Orderings.ord_class.max
  5325     Divides.op div ~> Divides.div_class.div
  5326     Divides.op mod ~> Divides.div_class.mod
  5327     Divides.op dvd ~> Divides.div_class.dvd
  5328 
  5329 INCOMPATIBILITY.  Adaptions may be required in the following cases:
  5330 
  5331 a) User-defined constants using any of the names "plus", "minus",
  5332 "times", "less" or "less_eq". The standard syntax translations for
  5333 "+", "-" and "*" may go wrong.  INCOMPATIBILITY: use more specific
  5334 names.
  5335 
  5336 b) Variables named "plus", "minus", "times", "less", "less_eq"
  5337 INCOMPATIBILITY: use more specific names.
  5338 
  5339 c) Permutative equations (e.g. "a + b = b + a")
  5340 Since the change of names also changes the order of terms, permutative
  5341 rewrite rules may get applied in a different order. Experience shows
  5342 that this is rarely the case (only two adaptions in the whole Isabelle
  5343 distribution).  INCOMPATIBILITY: rewrite proofs
  5344 
  5345 d) ML code directly refering to constant names
  5346 This in general only affects hand-written proof tactics, simprocs and
  5347 so on.  INCOMPATIBILITY: grep your sourcecode and replace names.
  5348 Consider using @{const_name} antiquotation.
  5349 
  5350 * New class "default" with associated constant "default".
  5351 
  5352 * Function "sgn" is now overloaded and available on int, real, complex
  5353 (and other numeric types), using class "sgn".  Two possible defs of
  5354 sgn are given as equational assumptions in the classes sgn_if and
  5355 sgn_div_norm; ordered_idom now also inherits from sgn_if.
  5356 INCOMPATIBILITY.
  5357 
  5358 * Locale "partial_order" now unified with class "order" (cf. theory
  5359 Orderings), added parameter "less".  INCOMPATIBILITY.
  5360 
  5361 * Renamings in classes "order" and "linorder": facts "refl", "trans" and
  5362 "cases" to "order_refl", "order_trans" and "linorder_cases", to avoid
  5363 clashes with HOL "refl" and "trans".  INCOMPATIBILITY.
  5364 
  5365 * Classes "order" and "linorder": potential INCOMPATIBILITY due to
  5366 changed order of proof goals in instance proofs.
  5367 
  5368 * The transitivity reasoner for partial and linear orders is set up
  5369 for classes "order" and "linorder".  Instances of the reasoner are available
  5370 in all contexts importing or interpreting the corresponding locales.
  5371 Method "order" invokes the reasoner separately; the reasoner
  5372 is also integrated with the Simplifier as a solver.  Diagnostic
  5373 command 'print_orders' shows the available instances of the reasoner
  5374 in the current context.
  5375 
  5376 * Localized monotonicity predicate in theory "Orderings"; integrated
  5377 lemmas max_of_mono and min_of_mono with this predicate.
  5378 INCOMPATIBILITY.
  5379 
  5380 * Formulation of theorem "dense" changed slightly due to integration
  5381 with new class dense_linear_order.
  5382 
  5383 * Uniform lattice theory development in HOL.
  5384 
  5385     constants "meet" and "join" now named "inf" and "sup"
  5386     constant "Meet" now named "Inf"
  5387 
  5388     classes "meet_semilorder" and "join_semilorder" now named
  5389       "lower_semilattice" and "upper_semilattice"
  5390     class "lorder" now named "lattice"
  5391     class "comp_lat" now named "complete_lattice"
  5392 
  5393     Instantiation of lattice classes allows explicit definitions
  5394     for "inf" and "sup" operations (or "Inf" and "Sup" for complete lattices).
  5395 
  5396   INCOMPATIBILITY.  Theorem renames:
  5397 
  5398     meet_left_le            ~> inf_le1
  5399     meet_right_le           ~> inf_le2
  5400     join_left_le            ~> sup_ge1
  5401     join_right_le           ~> sup_ge2
  5402     meet_join_le            ~> inf_sup_ord
  5403     le_meetI                ~> le_infI
  5404     join_leI                ~> le_supI
  5405     le_meet                 ~> le_inf_iff
  5406     le_join                 ~> ge_sup_conv
  5407     meet_idempotent         ~> inf_idem
  5408     join_idempotent         ~> sup_idem
  5409     meet_comm               ~> inf_commute
  5410     join_comm               ~> sup_commute
  5411     meet_leI1               ~> le_infI1
  5412     meet_leI2               ~> le_infI2
  5413     le_joinI1               ~> le_supI1
  5414     le_joinI2               ~> le_supI2
  5415     meet_assoc              ~> inf_assoc
  5416     join_assoc              ~> sup_assoc
  5417     meet_left_comm          ~> inf_left_commute
  5418     meet_left_idempotent    ~> inf_left_idem
  5419     join_left_comm          ~> sup_left_commute
  5420     join_left_idempotent    ~> sup_left_idem
  5421     meet_aci                ~> inf_aci
  5422     join_aci                ~> sup_aci
  5423     le_def_meet             ~> le_iff_inf
  5424     le_def_join             ~> le_iff_sup
  5425     join_absorp2            ~> sup_absorb2
  5426     join_absorp1            ~> sup_absorb1
  5427     meet_absorp1            ~> inf_absorb1
  5428     meet_absorp2            ~> inf_absorb2
  5429     meet_join_absorp        ~> inf_sup_absorb
  5430     join_meet_absorp        ~> sup_inf_absorb
  5431     distrib_join_le         ~> distrib_sup_le
  5432     distrib_meet_le         ~> distrib_inf_le
  5433 
  5434     add_meet_distrib_left   ~> add_inf_distrib_left
  5435     add_join_distrib_left   ~> add_sup_distrib_left
  5436     is_join_neg_meet        ~> is_join_neg_inf
  5437     is_meet_neg_join        ~> is_meet_neg_sup
  5438     add_meet_distrib_right  ~> add_inf_distrib_right
  5439     add_join_distrib_right  ~> add_sup_distrib_right
  5440     add_meet_join_distribs  ~> add_sup_inf_distribs
  5441     join_eq_neg_meet        ~> sup_eq_neg_inf
  5442     meet_eq_neg_join        ~> inf_eq_neg_sup
  5443     add_eq_meet_join        ~> add_eq_inf_sup
  5444     meet_0_imp_0            ~> inf_0_imp_0
  5445     join_0_imp_0            ~> sup_0_imp_0
  5446     meet_0_eq_0             ~> inf_0_eq_0
  5447     join_0_eq_0             ~> sup_0_eq_0
  5448     neg_meet_eq_join        ~> neg_inf_eq_sup
  5449     neg_join_eq_meet        ~> neg_sup_eq_inf
  5450     join_eq_if              ~> sup_eq_if
  5451 
  5452     mono_meet               ~> mono_inf
  5453     mono_join               ~> mono_sup
  5454     meet_bool_eq            ~> inf_bool_eq
  5455     join_bool_eq            ~> sup_bool_eq
  5456     meet_fun_eq             ~> inf_fun_eq
  5457     join_fun_eq             ~> sup_fun_eq
  5458     meet_set_eq             ~> inf_set_eq
  5459     join_set_eq             ~> sup_set_eq
  5460     meet1_iff               ~> inf1_iff
  5461     meet2_iff               ~> inf2_iff
  5462     meet1I                  ~> inf1I
  5463     meet2I                  ~> inf2I
  5464     meet1D1                 ~> inf1D1
  5465     meet2D1                 ~> inf2D1
  5466     meet1D2                 ~> inf1D2
  5467     meet2D2                 ~> inf2D2
  5468     meet1E                  ~> inf1E
  5469     meet2E                  ~> inf2E
  5470     join1_iff               ~> sup1_iff
  5471     join2_iff               ~> sup2_iff
  5472     join1I1                 ~> sup1I1
  5473     join2I1                 ~> sup2I1
  5474     join1I1                 ~> sup1I1
  5475     join2I2                 ~> sup1I2
  5476     join1CI                 ~> sup1CI
  5477     join2CI                 ~> sup2CI
  5478     join1E                  ~> sup1E
  5479     join2E                  ~> sup2E
  5480 
  5481     is_meet_Meet            ~> is_meet_Inf
  5482     Meet_bool_def           ~> Inf_bool_def
  5483     Meet_fun_def            ~> Inf_fun_def
  5484     Meet_greatest           ~> Inf_greatest
  5485     Meet_lower              ~> Inf_lower
  5486     Meet_set_def            ~> Inf_set_def
  5487 
  5488     Sup_def                 ~> Sup_Inf
  5489     Sup_bool_eq             ~> Sup_bool_def
  5490     Sup_fun_eq              ~> Sup_fun_def
  5491     Sup_set_eq              ~> Sup_set_def
  5492 
  5493     listsp_meetI            ~> listsp_infI
  5494     listsp_meet_eq          ~> listsp_inf_eq
  5495 
  5496     meet_min                ~> inf_min
  5497     join_max                ~> sup_max
  5498 
  5499 * Added syntactic class "size"; overloaded constant "size" now has
  5500 type "'a::size ==> bool"
  5501 
  5502 * Internal reorganisation of `size' of datatypes: size theorems
  5503 "foo.size" are no longer subsumed by "foo.simps" (but are still
  5504 simplification rules by default!); theorems "prod.size" now named
  5505 "*.size".
  5506 
  5507 * Class "div" now inherits from class "times" rather than "type".
  5508 INCOMPATIBILITY.
  5509 
  5510 * HOL/Finite_Set: "name-space" locales Lattice, Distrib_lattice,
  5511 Linorder etc.  have disappeared; operations defined in terms of
  5512 fold_set now are named Inf_fin, Sup_fin.  INCOMPATIBILITY.
  5513 
  5514 * HOL/Nat: neq0_conv no longer declared as iff.  INCOMPATIBILITY.
  5515 
  5516 * HOL-Word: New extensive library and type for generic, fixed size
  5517 machine words, with arithmetic, bit-wise, shifting and rotating
  5518 operations, reflection into int, nat, and bool lists, automation for
  5519 linear arithmetic (by automatic reflection into nat or int), including
  5520 lemmas on overflow and monotonicity.  Instantiated to all appropriate
  5521 arithmetic type classes, supporting automatic simplification of
  5522 numerals on all operations.
  5523 
  5524 * Library/Boolean_Algebra: locales for abstract boolean algebras.
  5525 
  5526 * Library/Numeral_Type: numbers as types, e.g. TYPE(32).
  5527 
  5528 * Code generator library theories:
  5529   - Code_Integer represents HOL integers by big integer literals in target
  5530     languages.
  5531   - Code_Char represents HOL characters by character literals in target
  5532     languages.
  5533   - Code_Char_chr like Code_Char, but also offers treatment of character
  5534     codes; includes Code_Integer.
  5535   - Executable_Set allows to generate code for finite sets using lists.
  5536   - Executable_Rat implements rational numbers as triples (sign, enumerator,
  5537     denominator).
  5538   - Executable_Real implements a subset of real numbers, namly those
  5539     representable by rational numbers.
  5540   - Efficient_Nat implements natural numbers by integers, which in general will
  5541     result in higher efficency; pattern matching with 0/Suc is eliminated;
  5542     includes Code_Integer.
  5543   - Code_Index provides an additional datatype index which is mapped to
  5544     target-language built-in integers.
  5545   - Code_Message provides an additional datatype message_string which is isomorphic to
  5546     strings; messages are mapped to target-language strings.
  5547 
  5548 * New package for inductive predicates
  5549 
  5550   An n-ary predicate p with m parameters z_1, ..., z_m can now be defined via
  5551 
  5552     inductive
  5553       p :: "U_1 => ... => U_m => T_1 => ... => T_n => bool"
  5554       for z_1 :: U_1 and ... and z_n :: U_m
  5555     where
  5556       rule_1: "... ==> p z_1 ... z_m t_1_1 ... t_1_n"
  5557     | ...
  5558 
  5559   with full support for type-inference, rather than
  5560 
  5561     consts s :: "U_1 => ... => U_m => (T_1 * ... * T_n) set"
  5562 
  5563     abbreviation p :: "U_1 => ... => U_m => T_1 => ... => T_n => bool"
  5564     where "p z_1 ... z_m x_1 ... x_n == (x_1, ..., x_n) : s z_1 ... z_m"
  5565 
  5566     inductive "s z_1 ... z_m"
  5567     intros
  5568       rule_1: "... ==> (t_1_1, ..., t_1_n) : s z_1 ... z_m"
  5569       ...
  5570 
  5571   For backward compatibility, there is a wrapper allowing inductive
  5572   sets to be defined with the new package via
  5573 
  5574     inductive_set
  5575       s :: "U_1 => ... => U_m => (T_1 * ... * T_n) set"
  5576       for z_1 :: U_1 and ... and z_n :: U_m
  5577     where
  5578       rule_1: "... ==> (t_1_1, ..., t_1_n) : s z_1 ... z_m"
  5579     | ...
  5580 
  5581   or
  5582 
  5583     inductive_set
  5584       s :: "U_1 => ... => U_m => (T_1 * ... * T_n) set"
  5585       and p :: "U_1 => ... => U_m => T_1 => ... => T_n => bool"
  5586       for z_1 :: U_1 and ... and z_n :: U_m
  5587     where
  5588       "p z_1 ... z_m x_1 ... x_n == (x_1, ..., x_n) : s z_1 ... z_m"
  5589     | rule_1: "... ==> p z_1 ... z_m t_1_1 ... t_1_n"
  5590     | ...
  5591 
  5592   if the additional syntax "p ..." is required.
  5593 
  5594   Numerous examples can be found in the subdirectories src/HOL/Auth,
  5595   src/HOL/Bali, src/HOL/Induct, and src/HOL/MicroJava.
  5596 
  5597   INCOMPATIBILITIES:
  5598 
  5599   - Since declaration and definition of inductive sets or predicates
  5600     is no longer separated, abbreviations involving the newly
  5601     introduced sets or predicates must be specified together with the
  5602     introduction rules after the 'where' keyword (see above), rather
  5603     than before the actual inductive definition.
  5604 
  5605   - The variables in induction and elimination rules are now
  5606     quantified in the order of their occurrence in the introduction
  5607     rules, rather than in alphabetical order. Since this may break
  5608     some proofs, these proofs either have to be repaired, e.g. by
  5609     reordering the variables a_i_1 ... a_i_{k_i} in Isar 'case'
  5610     statements of the form
  5611 
  5612       case (rule_i a_i_1 ... a_i_{k_i})
  5613 
  5614     or the old order of quantification has to be restored by explicitly adding
  5615     meta-level quantifiers in the introduction rules, i.e.
  5616 
  5617       | rule_i: "!!a_i_1 ... a_i_{k_i}. ... ==> p z_1 ... z_m t_i_1 ... t_i_n"
  5618 
  5619   - The format of the elimination rules is now
  5620 
  5621       p z_1 ... z_m x_1 ... x_n ==>
  5622         (!!a_1_1 ... a_1_{k_1}. x_1 = t_1_1 ==> ... ==> x_n = t_1_n ==> ... ==> P)
  5623         ==> ... ==> P
  5624 
  5625     for predicates and
  5626 
  5627       (x_1, ..., x_n) : s z_1 ... z_m ==>
  5628         (!!a_1_1 ... a_1_{k_1}. x_1 = t_1_1 ==> ... ==> x_n = t_1_n ==> ... ==> P)
  5629         ==> ... ==> P
  5630 
  5631     for sets rather than
  5632 
  5633       x : s z_1 ... z_m ==>
  5634         (!!a_1_1 ... a_1_{k_1}. x = (t_1_1, ..., t_1_n) ==> ... ==> P)
  5635         ==> ... ==> P
  5636 
  5637     This may require terms in goals to be expanded to n-tuples
  5638     (e.g. using case_tac or simplification with the split_paired_all
  5639     rule) before the above elimination rule is applicable.
  5640 
  5641   - The elimination or case analysis rules for (mutually) inductive
  5642     sets or predicates are now called "p_1.cases" ... "p_k.cases". The
  5643     list of rules "p_1_..._p_k.elims" is no longer available.
  5644 
  5645 * New package "function"/"fun" for general recursive functions,
  5646 supporting mutual and nested recursion, definitions in local contexts,
  5647 more general pattern matching and partiality. See HOL/ex/Fundefs.thy
  5648 for small examples, and the separate tutorial on the function
  5649 package. The old recdef "package" is still available as before, but
  5650 users are encouraged to use the new package.
  5651 
  5652 * Method "lexicographic_order" automatically synthesizes termination
  5653 relations as lexicographic combinations of size measures. 
  5654 
  5655 * Case-expressions allow arbitrary constructor-patterns (including
  5656 "_") and take their order into account, like in functional
  5657 programming.  Internally, this is translated into nested
  5658 case-expressions; missing cases are added and mapped to the predefined
  5659 constant "undefined". In complicated cases printing may no longer show
  5660 the original input but the internal form. Lambda-abstractions allow
  5661 the same form of pattern matching: "% pat1 => e1 | ..." is an
  5662 abbreviation for "%x. case x of pat1 => e1 | ..." where x is a new
  5663 variable.
  5664 
  5665 * IntDef: The constant "int :: nat => int" has been removed; now "int"
  5666 is an abbreviation for "of_nat :: nat => int". The simplification
  5667 rules for "of_nat" have been changed to work like "int" did
  5668 previously.  Potential INCOMPATIBILITY:
  5669   - "of_nat (Suc m)" simplifies to "1 + of_nat m" instead of "of_nat m + 1"
  5670   - of_nat_diff and of_nat_mult are no longer default simp rules
  5671 
  5672 * Method "algebra" solves polynomial equations over (semi)rings using
  5673 Groebner bases. The (semi)ring structure is defined by locales and the
  5674 tool setup depends on that generic context. Installing the method for
  5675 a specific type involves instantiating the locale and possibly adding
  5676 declarations for computation on the coefficients.  The method is
  5677 already instantiated for natural numbers and for the axiomatic class
  5678 of idoms with numerals.  See also the paper by Chaieb and Wenzel at
  5679 CALCULEMUS 2007 for the general principles underlying this
  5680 architecture of context-aware proof-tools.
  5681 
  5682 * Method "ferrack" implements quantifier elimination over
  5683 special-purpose dense linear orders using locales (analogous to
  5684 "algebra"). The method is already installed for class
  5685 {ordered_field,recpower,number_ring} which subsumes real, hyperreal,
  5686 rat, etc.
  5687 
  5688 * Former constant "List.op @" now named "List.append".  Use ML
  5689 antiquotations @{const_name List.append} or @{term " ... @ ... "} to
  5690 circumvent possible incompatibilities when working on ML level.
  5691 
  5692 * primrec: missing cases mapped to "undefined" instead of "arbitrary".
  5693 
  5694 * New function listsum :: 'a list => 'a for arbitrary monoids.
  5695 Special syntax: "SUM x <- xs. f x" (and latex variants)
  5696 
  5697 * New syntax for Haskell-like list comprehension (input only), eg.
  5698 [(x,y). x <- xs, y <- ys, x ~= y], see also src/HOL/List.thy.
  5699 
  5700 * The special syntax for function "filter" has changed from [x :
  5701 xs. P] to [x <- xs. P] to avoid an ambiguity caused by list
  5702 comprehension syntax, and for uniformity.  INCOMPATIBILITY.
  5703 
  5704 * [a..b] is now defined for arbitrary linear orders.  It used to be
  5705 defined on nat only, as an abbreviation for [a..<Suc b]
  5706 INCOMPATIBILITY.
  5707 
  5708 * Renamed lemma "set_take_whileD"  to "set_takeWhileD".
  5709 
  5710 * New functions "sorted" and "sort" in src/HOL/List.thy.
  5711 
  5712 * New lemma collection field_simps (an extension of ring_simps) for
  5713 manipulating (in)equations involving division. Multiplies with all
  5714 denominators that can be proved to be non-zero (in equations) or
  5715 positive/negative (in inequations).
  5716 
  5717 * Lemma collections ring_eq_simps, group_eq_simps and ring_distrib
  5718 have been improved and renamed to ring_simps, group_simps and
  5719 ring_distribs.  Removed lemmas field_xyz in theory Ring_and_Field
  5720 because they were subsumed by lemmas xyz.  INCOMPATIBILITY.
  5721 
  5722 * Theory Library/Commutative_Ring: switched from recdef to function
  5723 package; constants add, mul, pow now curried.  Infix syntax for
  5724 algebraic operations.
  5725 
  5726 * Dropped redundant lemma def_imp_eq in favor of meta_eq_to_obj_eq.
  5727 INCOMPATIBILITY.
  5728 
  5729 * Dropped redundant lemma if_def2 in favor of if_bool_eq_conj.
  5730 INCOMPATIBILITY.
  5731 
  5732 * HOL/records: generalised field-update to take a function on the
  5733 field rather than the new value: r(|A := x|) is translated to A_update
  5734 (K x) r The K-combinator that is internally used is called K_record.
  5735 INCOMPATIBILITY: Usage of the plain update functions has to be
  5736 adapted.
  5737  
  5738 * Class "semiring_0" now contains annihilation axioms x * 0 = 0 and 0
  5739 * x = 0, which are required for a semiring.  Richer structures do not
  5740 inherit from semiring_0 anymore, because this property is a theorem
  5741 there, not an axiom.  INCOMPATIBILITY: In instances of semiring_0,
  5742 there is more to prove, but this is mostly trivial.
  5743 
  5744 * Class "recpower" is generalized to arbitrary monoids, not just
  5745 commutative semirings.  INCOMPATIBILITY: may need to incorporate
  5746 commutativity or semiring properties additionally.
  5747 
  5748 * Constant "List.list_all2" in List.thy now uses authentic syntax.
  5749 INCOMPATIBILITY: translations containing list_all2 may go wrong,
  5750 better use 'abbreviation'.
  5751 
  5752 * Renamed constant "List.op mem" to "List.member".  INCOMPATIBILITY.
  5753 
  5754 * Numeral syntax: type 'bin' which was a mere type copy of 'int' has
  5755 been abandoned in favour of plain 'int'.  INCOMPATIBILITY --
  5756 significant changes for setting up numeral syntax for types:
  5757   - New constants Numeral.pred and Numeral.succ instead
  5758       of former Numeral.bin_pred and Numeral.bin_succ.
  5759   - Use integer operations instead of bin_add, bin_mult and so on.
  5760   - Numeral simplification theorems named Numeral.numeral_simps instead of Bin_simps.
  5761   - ML structure Bin_Simprocs now named Int_Numeral_Base_Simprocs.
  5762 
  5763 See src/HOL/Integ/IntArith.thy for an example setup.
  5764 
  5765 * Command 'normal_form' computes the normal form of a term that may
  5766 contain free variables.  For example ``normal_form "rev [a, b, c]"''
  5767 produces ``[b, c, a]'' (without proof).  This command is suitable for
  5768 heavy-duty computations because the functions are compiled to ML
  5769 first.  Correspondingly, a method "normalization" is provided.  See
  5770 further src/HOL/ex/NormalForm.thy and src/Tools/nbe.ML.
  5771 
  5772 * Alternative iff syntax "A <-> B" for equality on bool (with priority
  5773 25 like -->); output depends on the "iff" print_mode, the default is
  5774 "A = B" (with priority 50).
  5775 
  5776 * Relations less (<) and less_eq (<=) are also available on type bool.
  5777 Modified syntax to disallow nesting without explicit parentheses,
  5778 e.g. "(x < y) < z" or "x < (y < z)", but NOT "x < y < z".  Potential
  5779 INCOMPATIBILITY.
  5780 
  5781 * "LEAST x:A. P" expands to "LEAST x. x:A & P" (input only).
  5782 
  5783 * Relation composition operator "op O" now has precedence 75 and binds
  5784 stronger than union and intersection. INCOMPATIBILITY.
  5785 
  5786 * The old set interval syntax "{m..n(}" (and relatives) has been
  5787 removed.  Use "{m..<n}" (and relatives) instead.
  5788 
  5789 * In the context of the assumption "~(s = t)" the Simplifier rewrites
  5790 "t = s" to False (by simproc "neq").  INCOMPATIBILITY, consider using
  5791 ``declare [[simproc del: neq]]''.
  5792 
  5793 * Simplifier: "m dvd n" where m and n are numbers is evaluated to
  5794 True/False.
  5795 
  5796 * Theorem Cons_eq_map_conv no longer declared as "simp".
  5797 
  5798 * Theorem setsum_mult renamed to setsum_right_distrib.
  5799 
  5800 * Prefer ex1I over ex_ex1I in single-step reasoning, e.g. by the
  5801 ``rule'' method.
  5802 
  5803 * Reimplemented methods "sat" and "satx", with several improvements:
  5804 goals no longer need to be stated as "<prems> ==> False", equivalences
  5805 (i.e. "=" on type bool) are handled, variable names of the form
  5806 "lit_<n>" are no longer reserved, significant speedup.
  5807 
  5808 * Methods "sat" and "satx" can now replay MiniSat proof traces.
  5809 zChaff is still supported as well.
  5810 
  5811 * 'inductive' and 'datatype': provide projections of mutual rules,
  5812 bundled as foo_bar.inducts;
  5813 
  5814 * Library: moved theories Parity, GCD, Binomial, Infinite_Set to
  5815 Library.
  5816 
  5817 * Library: moved theory Accessible_Part to main HOL.
  5818 
  5819 * Library: added theory Coinductive_List of potentially infinite lists
  5820 as greatest fixed-point.
  5821 
  5822 * Library: added theory AssocList which implements (finite) maps as
  5823 association lists.
  5824 
  5825 * Method "evaluation" solves goals (i.e. a boolean expression)
  5826 efficiently by compiling it to ML.  The goal is "proved" (via an
  5827 oracle) if it evaluates to True.
  5828 
  5829 * Linear arithmetic now splits certain operators (e.g. min, max, abs)
  5830 also when invoked by the simplifier.  This results in the Simplifier
  5831 being more powerful on arithmetic goals.  INCOMPATIBILITY.
  5832 Configuration option fast_arith_split_limit=0 recovers the old
  5833 behavior.
  5834 
  5835 * Support for hex (0x20) and binary (0b1001) numerals.
  5836 
  5837 * New method: reify eqs (t), where eqs are equations for an
  5838 interpretation I :: 'a list => 'b => 'c and t::'c is an optional
  5839 parameter, computes a term s::'b and a list xs::'a list and proves the
  5840 theorem I xs s = t. This is also known as reification or quoting. The
  5841 resulting theorem is applied to the subgoal to substitute t with I xs
  5842 s.  If t is omitted, the subgoal itself is reified.
  5843 
  5844 * New method: reflection corr_thm eqs (t). The parameters eqs and (t)
  5845 are as explained above. corr_thm is a theorem for I vs (f t) = I vs t,
  5846 where f is supposed to be a computable function (in the sense of code
  5847 generattion). The method uses reify to compute s and xs as above then
  5848 applies corr_thm and uses normalization by evaluation to "prove" f s =
  5849 r and finally gets the theorem t = r, which is again applied to the
  5850 subgoal. An Example is available in src/HOL/ex/ReflectionEx.thy.
  5851 
  5852 * Reflection: Automatic reification now handels binding, an example is
  5853 available in src/HOL/ex/ReflectionEx.thy
  5854 
  5855 * HOL-Statespace: ``State Spaces: The Locale Way'' introduces a
  5856 command 'statespace' that is similar to 'record', but introduces an
  5857 abstract specification based on the locale infrastructure instead of
  5858 HOL types.  This leads to extra flexibility in composing state spaces,
  5859 in particular multiple inheritance and renaming of components.
  5860 
  5861 
  5862 *** HOL-Complex ***
  5863 
  5864 * Hyperreal: Functions root and sqrt are now defined on negative real
  5865 inputs so that root n (- x) = - root n x and sqrt (- x) = - sqrt x.
  5866 Nonnegativity side conditions have been removed from many lemmas, so
  5867 that more subgoals may now be solved by simplification; potential
  5868 INCOMPATIBILITY.
  5869 
  5870 * Real: new type classes formalize real normed vector spaces and
  5871 algebras, using new overloaded constants scaleR :: real => 'a => 'a
  5872 and norm :: 'a => real.
  5873 
  5874 * Real: constant of_real :: real => 'a::real_algebra_1 injects from
  5875 reals into other types. The overloaded constant Reals :: 'a set is now
  5876 defined as range of_real; potential INCOMPATIBILITY.
  5877 
  5878 * Real: proper support for ML code generation, including 'quickcheck'.
  5879 Reals are implemented as arbitrary precision rationals.
  5880 
  5881 * Hyperreal: Several constants that previously worked only for the
  5882 reals have been generalized, so they now work over arbitrary vector
  5883 spaces. Type annotations may need to be added in some cases; potential
  5884 INCOMPATIBILITY.
  5885 
  5886   Infinitesimal  :: ('a::real_normed_vector) star set
  5887   HFinite        :: ('a::real_normed_vector) star set
  5888   HInfinite      :: ('a::real_normed_vector) star set
  5889   approx         :: ('a::real_normed_vector) star => 'a star => bool
  5890   monad          :: ('a::real_normed_vector) star => 'a star set
  5891   galaxy         :: ('a::real_normed_vector) star => 'a star set
  5892   (NS)LIMSEQ     :: [nat => 'a::real_normed_vector, 'a] => bool
  5893   (NS)convergent :: (nat => 'a::real_normed_vector) => bool
  5894   (NS)Bseq       :: (nat => 'a::real_normed_vector) => bool
  5895   (NS)Cauchy     :: (nat => 'a::real_normed_vector) => bool
  5896   (NS)LIM        :: ['a::real_normed_vector => 'b::real_normed_vector, 'a, 'b] => bool
  5897   is(NS)Cont     :: ['a::real_normed_vector => 'b::real_normed_vector, 'a] => bool
  5898   deriv          :: ['a::real_normed_field => 'a, 'a, 'a] => bool
  5899   sgn            :: 'a::real_normed_vector => 'a
  5900   exp            :: 'a::{recpower,real_normed_field,banach} => 'a
  5901 
  5902 * Complex: Some complex-specific constants are now abbreviations for
  5903 overloaded ones: complex_of_real = of_real, cmod = norm, hcmod =
  5904 hnorm.  Other constants have been entirely removed in favor of the
  5905 polymorphic versions (INCOMPATIBILITY):
  5906 
  5907   approx        <-- capprox
  5908   HFinite       <-- CFinite
  5909   HInfinite     <-- CInfinite
  5910   Infinitesimal <-- CInfinitesimal
  5911   monad         <-- cmonad
  5912   galaxy        <-- cgalaxy
  5913   (NS)LIM       <-- (NS)CLIM, (NS)CRLIM
  5914   is(NS)Cont    <-- is(NS)Contc, is(NS)contCR
  5915   (ns)deriv     <-- (ns)cderiv
  5916 
  5917 
  5918 *** HOL-Algebra ***
  5919 
  5920 * Formalisation of ideals and the quotient construction over rings.
  5921 
  5922 * Order and lattice theory no longer based on records.
  5923 INCOMPATIBILITY.
  5924 
  5925 * Renamed lemmas least_carrier -> least_closed and greatest_carrier ->
  5926 greatest_closed.  INCOMPATIBILITY.
  5927 
  5928 * Method algebra is now set up via an attribute.  For examples see
  5929 Ring.thy.  INCOMPATIBILITY: the method is now weaker on combinations
  5930 of algebraic structures.
  5931 
  5932 * Renamed theory CRing to Ring.
  5933 
  5934 
  5935 *** HOL-Nominal ***
  5936 
  5937 * Substantial, yet incomplete support for nominal datatypes (binding
  5938 structures) based on HOL-Nominal logic.  See src/HOL/Nominal and
  5939 src/HOL/Nominal/Examples.  Prospective users should consult
  5940 http://isabelle.in.tum.de/nominal/
  5941 
  5942 
  5943 *** ML ***
  5944 
  5945 * ML basics: just one true type int, which coincides with IntInf.int
  5946 (even on SML/NJ).
  5947 
  5948 * ML within Isar: antiquotations allow to embed statically-checked
  5949 formal entities in the source, referring to the context available at
  5950 compile-time.  For example:
  5951 
  5952 ML {* @{sort "{zero,one}"} *}
  5953 ML {* @{typ "'a => 'b"} *}
  5954 ML {* @{term "%x. x"} *}
  5955 ML {* @{prop "x == y"} *}
  5956 ML {* @{ctyp "'a => 'b"} *}
  5957 ML {* @{cterm "%x. x"} *}
  5958 ML {* @{cprop "x == y"} *}
  5959 ML {* @{thm asm_rl} *}
  5960 ML {* @{thms asm_rl} *}
  5961 ML {* @{type_name c} *}
  5962 ML {* @{type_syntax c} *}
  5963 ML {* @{const_name c} *}
  5964 ML {* @{const_syntax c} *}
  5965 ML {* @{context} *}
  5966 ML {* @{theory} *}
  5967 ML {* @{theory Pure} *}
  5968 ML {* @{theory_ref} *}
  5969 ML {* @{theory_ref Pure} *}
  5970 ML {* @{simpset} *}
  5971 ML {* @{claset} *}
  5972 ML {* @{clasimpset} *}
  5973 
  5974 The same works for sources being ``used'' within an Isar context.
  5975 
  5976 * ML in Isar: improved error reporting; extra verbosity with
  5977 ML_Context.trace enabled.
  5978 
  5979 * Pure/General/table.ML: the join operations now works via exceptions
  5980 DUP/SAME instead of type option. This is simpler in simple cases, and
  5981 admits slightly more efficient complex applications.
  5982 
  5983 * Pure: 'advanced' translation functions (parse_translation etc.) now
  5984 use Context.generic instead of just theory.
  5985 
  5986 * Pure: datatype Context.generic joins theory/Proof.context and
  5987 provides some facilities for code that works in either kind of
  5988 context, notably GenericDataFun for uniform theory and proof data.
  5989 
  5990 * Pure: simplified internal attribute type, which is now always
  5991 Context.generic * thm -> Context.generic * thm. Global (theory) vs.
  5992 local (Proof.context) attributes have been discontinued, while
  5993 minimizing code duplication. Thm.rule_attribute and
  5994 Thm.declaration_attribute build canonical attributes; see also structure
  5995 Context for further operations on Context.generic, notably
  5996 GenericDataFun. INCOMPATIBILITY, need to adapt attribute type
  5997 declarations and definitions.
  5998 
  5999 * Context data interfaces (Theory/Proof/GenericDataFun): removed
  6000 name/print, uninitialized data defaults to ad-hoc copy of empty value,
  6001 init only required for impure data. INCOMPATIBILITY: empty really need
  6002 to be empty (no dependencies on theory content!)
  6003 
  6004 * Pure/kernel: consts certification ignores sort constraints given in
  6005 signature declarations. (This information is not relevant to the
  6006 logic, but only for type inference.) SIGNIFICANT INTERNAL CHANGE,
  6007 potential INCOMPATIBILITY.
  6008 
  6009 * Pure: axiomatic type classes are now purely definitional, with
  6010 explicit proofs of class axioms and super class relations performed
  6011 internally. See Pure/axclass.ML for the main internal interfaces --
  6012 notably AxClass.define_class supercedes AxClass.add_axclass, and
  6013 AxClass.axiomatize_class/classrel/arity supersede
  6014 Sign.add_classes/classrel/arities.
  6015 
  6016 * Pure/Isar: Args/Attrib parsers operate on Context.generic --
  6017 global/local versions on theory vs. Proof.context have been
  6018 discontinued; Attrib.syntax and Method.syntax have been adapted
  6019 accordingly.  INCOMPATIBILITY, need to adapt parser expressions for
  6020 attributes, methods, etc.
  6021 
  6022 * Pure: several functions of signature "... -> theory -> theory * ..."
  6023 have been reoriented to "... -> theory -> ... * theory" in order to
  6024 allow natural usage in combination with the ||>, ||>>, |-> and
  6025 fold_map combinators.
  6026 
  6027 * Pure: official theorem names (closed derivations) and additional
  6028 comments (tags) are now strictly separate.  Name hints -- which are
  6029 maintained as tags -- may be attached any time without affecting the
  6030 derivation.
  6031 
  6032 * Pure: primitive rule lift_rule now takes goal cterm instead of an
  6033 actual goal state (thm).  Use Thm.lift_rule (Thm.cprem_of st i) to
  6034 achieve the old behaviour.
  6035 
  6036 * Pure: the "Goal" constant is now called "prop", supporting a
  6037 slightly more general idea of ``protecting'' meta-level rule
  6038 statements.
  6039 
  6040 * Pure: Logic.(un)varify only works in a global context, which is now
  6041 enforced instead of silently assumed.  INCOMPATIBILITY, may use
  6042 Logic.legacy_(un)varify as temporary workaround.
  6043 
  6044 * Pure: structure Name provides scalable operations for generating
  6045 internal variable names, notably Name.variants etc.  This replaces
  6046 some popular functions from term.ML:
  6047 
  6048   Term.variant		->  Name.variant
  6049   Term.variantlist	->  Name.variant_list
  6050   Term.invent_names	->  Name.invent_list
  6051 
  6052 Note that low-level renaming rarely occurs in new code -- operations
  6053 from structure Variable are used instead (see below).
  6054 
  6055 * Pure: structure Variable provides fundamental operations for proper
  6056 treatment of fixed/schematic variables in a context.  For example,
  6057 Variable.import introduces fixes for schematics of given facts and
  6058 Variable.export reverses the effect (up to renaming) -- this replaces
  6059 various freeze_thaw operations.
  6060 
  6061 * Pure: structure Goal provides simple interfaces for
  6062 init/conclude/finish and tactical prove operations (replacing former
  6063 Tactic.prove).  Goal.prove is the canonical way to prove results
  6064 within a given context; Goal.prove_global is a degraded version for
  6065 theory level goals, including a global Drule.standard.  Note that
  6066 OldGoals.prove_goalw_cterm has long been obsolete, since it is
  6067 ill-behaved in a local proof context (e.g. with local fixes/assumes or
  6068 in a locale context).
  6069 
  6070 * Pure/Syntax: generic interfaces for parsing (Syntax.parse_term etc.)
  6071 and type checking (Syntax.check_term etc.), with common combinations
  6072 (Syntax.read_term etc.). These supersede former Sign.read_term etc.
  6073 which are considered legacy and await removal.
  6074 
  6075 * Pure/Syntax: generic interfaces for type unchecking
  6076 (Syntax.uncheck_terms etc.) and unparsing (Syntax.unparse_term etc.),
  6077 with common combinations (Syntax.pretty_term, Syntax.string_of_term
  6078 etc.).  Former Sign.pretty_term, Sign.string_of_term etc. are still
  6079 available for convenience, but refer to the very same operations using
  6080 a mere theory instead of a full context.
  6081 
  6082 * Isar: simplified treatment of user-level errors, using exception
  6083 ERROR of string uniformly.  Function error now merely raises ERROR,
  6084 without any side effect on output channels.  The Isar toplevel takes
  6085 care of proper display of ERROR exceptions.  ML code may use plain
  6086 handle/can/try; cat_error may be used to concatenate errors like this:
  6087 
  6088   ... handle ERROR msg => cat_error msg "..."
  6089 
  6090 Toplevel ML code (run directly or through the Isar toplevel) may be
  6091 embedded into the Isar toplevel with exception display/debug like
  6092 this:
  6093 
  6094   Isar.toplevel (fn () => ...)
  6095 
  6096 INCOMPATIBILITY, removed special transform_error facilities, removed
  6097 obsolete variants of user-level exceptions (ERROR_MESSAGE,
  6098 Context.PROOF, ProofContext.CONTEXT, Proof.STATE, ProofHistory.FAIL)
  6099 -- use plain ERROR instead.
  6100 
  6101 * Isar: theory setup now has type (theory -> theory), instead of a
  6102 list.  INCOMPATIBILITY, may use #> to compose setup functions.
  6103 
  6104 * Isar: ML toplevel pretty printer for type Proof.context, subject to
  6105 ProofContext.debug/verbose flags.
  6106 
  6107 * Isar: Toplevel.theory_to_proof admits transactions that modify the
  6108 theory before entering a proof state.  Transactions now always see a
  6109 quasi-functional intermediate checkpoint, both in interactive and
  6110 batch mode.
  6111 
  6112 * Isar: simplified interfaces for outer syntax.  Renamed
  6113 OuterSyntax.add_keywords to OuterSyntax.keywords.  Removed
  6114 OuterSyntax.add_parsers -- this functionality is now included in
  6115 OuterSyntax.command etc.  INCOMPATIBILITY.
  6116 
  6117 * Simplifier: the simpset of a running simplification process now
  6118 contains a proof context (cf. Simplifier.the_context), which is the
  6119 very context that the initial simpset has been retrieved from (by
  6120 simpset_of/local_simpset_of).  Consequently, all plug-in components
  6121 (solver, looper etc.) may depend on arbitrary proof data.
  6122 
  6123 * Simplifier.inherit_context inherits the proof context (plus the
  6124 local bounds) of the current simplification process; any simproc
  6125 etc. that calls the Simplifier recursively should do this!  Removed
  6126 former Simplifier.inherit_bounds, which is already included here --
  6127 INCOMPATIBILITY.  Tools based on low-level rewriting may even have to
  6128 specify an explicit context using Simplifier.context/theory_context.
  6129 
  6130 * Simplifier/Classical Reasoner: more abstract interfaces
  6131 change_simpset/claset for modifying the simpset/claset reference of a
  6132 theory; raw versions simpset/claset_ref etc. have been discontinued --
  6133 INCOMPATIBILITY.
  6134 
  6135 * Provers: more generic wrt. syntax of object-logics, avoid hardwired
  6136 "Trueprop" etc.
  6137 
  6138 
  6139 *** System ***
  6140 
  6141 * settings: the default heap location within ISABELLE_HOME_USER now
  6142 includes ISABELLE_IDENTIFIER.  This simplifies use of multiple
  6143 Isabelle installations.
  6144 
  6145 * isabelle-process: option -S (secure mode) disables some critical
  6146 operations, notably runtime compilation and evaluation of ML source
  6147 code.
  6148 
  6149 * Basic Isabelle mode for jEdit, see Isabelle/lib/jedit/.
  6150 
  6151 * Support for parallel execution, using native multicore support of
  6152 Poly/ML 5.1.  The theory loader exploits parallelism when processing
  6153 independent theories, according to the given theory header
  6154 specifications. The maximum number of worker threads is specified via
  6155 usedir option -M or the "max-threads" setting in Proof General. A
  6156 speedup factor of 1.5--3.5 can be expected on a 4-core machine, and up
  6157 to 6 on a 8-core machine.  User-code needs to observe certain
  6158 guidelines for thread-safe programming, see appendix A in the Isar
  6159 Implementation manual.
  6160 
  6161 
  6162 
  6163 New in Isabelle2005 (October 2005)
  6164 ----------------------------------
  6165 
  6166 *** General ***
  6167 
  6168 * Theory headers: the new header syntax for Isar theories is
  6169 
  6170   theory <name>
  6171   imports <theory1> ... <theoryN>
  6172   uses <file1> ... <fileM>
  6173   begin
  6174 
  6175 where the 'uses' part is optional.  The previous syntax
  6176 
  6177   theory <name> = <theory1> + ... + <theoryN>:
  6178 
  6179 will disappear in the next release.  Use isatool fixheaders to convert
  6180 existing theory files.  Note that there is no change in ancient
  6181 non-Isar theories now, but these will disappear soon.
  6182 
  6183 * Theory loader: parent theories can now also be referred to via
  6184 relative and absolute paths.
  6185 
  6186 * Command 'find_theorems' searches for a list of criteria instead of a
  6187 list of constants. Known criteria are: intro, elim, dest, name:string,
  6188 simp:term, and any term. Criteria can be preceded by '-' to select
  6189 theorems that do not match. Intro, elim, dest select theorems that
  6190 match the current goal, name:s selects theorems whose fully qualified
  6191 name contain s, and simp:term selects all simplification rules whose
  6192 lhs match term.  Any other term is interpreted as pattern and selects
  6193 all theorems matching the pattern. Available in ProofGeneral under
  6194 'ProofGeneral -> Find Theorems' or C-c C-f.  Example:
  6195 
  6196   C-c C-f (100) "(_::nat) + _ + _" intro -name: "HOL."
  6197 
  6198 prints the last 100 theorems matching the pattern "(_::nat) + _ + _",
  6199 matching the current goal as introduction rule and not having "HOL."
  6200 in their name (i.e. not being defined in theory HOL).
  6201 
  6202 * Command 'thms_containing' has been discontinued in favour of
  6203 'find_theorems'; INCOMPATIBILITY.
  6204 
  6205 * Communication with Proof General is now 8bit clean, which means that
  6206 Unicode text in UTF-8 encoding may be used within theory texts (both
  6207 formal and informal parts).  Cf. option -U of the Isabelle Proof
  6208 General interface.  Here are some simple examples (cf. src/HOL/ex):
  6209 
  6210   http://isabelle.in.tum.de/library/HOL/ex/Hebrew.html
  6211   http://isabelle.in.tum.de/library/HOL/ex/Chinese.html
  6212 
  6213 * Improved efficiency of the Simplifier and, to a lesser degree, the
  6214 Classical Reasoner.  Typical big applications run around 2 times
  6215 faster.
  6216 
  6217 
  6218 *** Document preparation ***
  6219 
  6220 * Commands 'display_drafts' and 'print_drafts' perform simple output
  6221 of raw sources.  Only those symbols that do not require additional
  6222 LaTeX packages (depending on comments in isabellesym.sty) are
  6223 displayed properly, everything else is left verbatim.  isatool display
  6224 and isatool print are used as front ends (these are subject to the
  6225 DVI/PDF_VIEWER and PRINT_COMMAND settings, respectively).
  6226 
  6227 * Command tags control specific markup of certain regions of text,
  6228 notably folding and hiding.  Predefined tags include "theory" (for
  6229 theory begin and end), "proof" for proof commands, and "ML" for
  6230 commands involving ML code; the additional tags "visible" and
  6231 "invisible" are unused by default.  Users may give explicit tag
  6232 specifications in the text, e.g. ''by %invisible (auto)''.  The
  6233 interpretation of tags is determined by the LaTeX job during document
  6234 preparation: see option -V of isatool usedir, or options -n and -t of
  6235 isatool document, or even the LaTeX macros \isakeeptag, \isafoldtag,
  6236 \isadroptag.
  6237