NEWS
author blanchet
Wed Jul 18 08:44:03 2012 +0200 (2012-07-18)
changeset 48294 2b0c5553dc46
parent 48206 937b53a339f0
child 48317 e5420161d11d
permissions -rw-r--r--
doc updates
     1 Isabelle NEWS -- history user-relevant changes
     2 ==============================================
     3 
     4 New in this Isabelle version
     5 ----------------------------
     6 
     7 *** General ***
     8 
     9 * Discontinued obsolete method fastsimp / tactic fast_simp_tac, which
    10 is called fastforce / fast_force_tac already since Isabelle2011-1.
    11 
    12 * Updated and extended "isar-ref" manual, reduced remaining material
    13 in old "ref" manual.
    14 
    15 
    16 *** Pure ***
    17 
    18 * Discontinued obsolete attribute "COMP".  Potential INCOMPATIBILITY,
    19 use regular rule composition via "OF" / "THEN", or explicit proof
    20 structure instead.  Note that Isabelle/ML provides a variety of
    21 operators like COMP, INCR_COMP, COMP_INCR, which need to be applied
    22 with some care where this is really required.
    23 
    24 
    25 *** HOL ***
    26 
    27 * Simproc "finite_Collect" rewrites set comprehensions into pointfree
    28 expressions.
    29 
    30 * Quickcheck:
    31 
    32   - added an optimisation for equality premises.
    33     It is switched on by default, and can be switched off by setting
    34     the configuration quickcheck_optimise_equality to false.    
    35 
    36 * The SMT solver Z3 has now by default a restricted set of directly
    37 supported features. For the full set of features (div/mod, nonlinear
    38 arithmetic, datatypes/records) with potential proof reconstruction
    39 failures, enable the configuration option "z3_with_extensions".
    40 Minor INCOMPATIBILITY.
    41 
    42 * Sledgehammer:
    43 
    44   - Rationalized type encodings ("type_enc" option).
    45   - Renamed options:
    46       max_relevant ~> max_facts
    47       relevance_thresholds ~> fact_thresholds
    48 
    49 
    50 *** Document preparation ***
    51 
    52 * Default for \<euro> is now based on eurosym package, instead of
    53 slightly exotic babel/greek.
    54 
    55 
    56 *** System ***
    57 
    58 * Discontinued support for Poly/ML 5.2.1, which was the last version
    59 without exception positions and advanced ML compiler/toplevel
    60 configuration.
    61 
    62 
    63 
    64 New in Isabelle2012 (May 2012)
    65 ------------------------------
    66 
    67 *** General ***
    68 
    69 * Prover IDE (PIDE) improvements:
    70 
    71   - more robust Sledgehammer integration (as before the sledgehammer
    72     command-line needs to be typed into the source buffer)
    73   - markup for bound variables
    74   - markup for types of term variables (displayed as tooltips)
    75   - support for user-defined Isar commands within the running session
    76   - improved support for Unicode outside original 16bit range
    77     e.g. glyph for \<A> (thanks to jEdit 4.5.1)
    78 
    79 * Forward declaration of outer syntax keywords within the theory
    80 header -- minor INCOMPATIBILITY for user-defined commands.  Allow new
    81 commands to be used in the same theory where defined.
    82 
    83 * Auxiliary contexts indicate block structure for specifications with
    84 additional parameters and assumptions.  Such unnamed contexts may be
    85 nested within other targets, like 'theory', 'locale', 'class',
    86 'instantiation' etc.  Results from the local context are generalized
    87 accordingly and applied to the enclosing target context.  Example:
    88 
    89   context
    90     fixes x y z :: 'a
    91     assumes xy: "x = y" and yz: "y = z"
    92   begin
    93 
    94   lemma my_trans: "x = z" using xy yz by simp
    95 
    96   end
    97 
    98   thm my_trans
    99 
   100 The most basic application is to factor-out context elements of
   101 several fixes/assumes/shows theorem statements, e.g. see
   102 ~~/src/HOL/Isar_Examples/Group_Context.thy
   103 
   104 Any other local theory specification element works within the "context
   105 ... begin ... end" block as well.
   106 
   107 * Bundled declarations associate attributed fact expressions with a
   108 given name in the context.  These may be later included in other
   109 contexts.  This allows to manage context extensions casually, without
   110 the logical dependencies of locales and locale interpretation.  See
   111 commands 'bundle', 'include', 'including' etc. in the isar-ref manual.
   112 
   113 * Commands 'lemmas' and 'theorems' allow local variables using 'for'
   114 declaration, and results are standardized before being stored.  Thus
   115 old-style "standard" after instantiation or composition of facts
   116 becomes obsolete.  Minor INCOMPATIBILITY, due to potential change of
   117 indices of schematic variables.
   118 
   119 * Rule attributes in local theory declarations (e.g. locale or class)
   120 are now statically evaluated: the resulting theorem is stored instead
   121 of the original expression.  INCOMPATIBILITY in rare situations, where
   122 the historic accident of dynamic re-evaluation in interpretations
   123 etc. was exploited.
   124 
   125 * New tutorial "Programming and Proving in Isabelle/HOL"
   126 ("prog-prove").  It completely supersedes "A Tutorial Introduction to
   127 Structured Isar Proofs" ("isar-overview"), which has been removed.  It
   128 also supersedes "Isabelle/HOL, A Proof Assistant for Higher-Order
   129 Logic" as the recommended beginners tutorial, but does not cover all
   130 of the material of that old tutorial.
   131 
   132 * Updated and extended reference manuals: "isar-ref",
   133 "implementation", "system"; reduced remaining material in old "ref"
   134 manual.
   135 
   136 
   137 *** Pure ***
   138 
   139 * Command 'definition' no longer exports the foundational "raw_def"
   140 into the user context.  Minor INCOMPATIBILITY, may use the regular
   141 "def" result with attribute "abs_def" to imitate the old version.
   142 
   143 * Attribute "abs_def" turns an equation of the form "f x y == t" into
   144 "f == %x y. t", which ensures that "simp" or "unfold" steps always
   145 expand it.  This also works for object-logic equality.  (Formerly
   146 undocumented feature.)
   147 
   148 * Sort constraints are now propagated in simultaneous statements, just
   149 like type constraints.  INCOMPATIBILITY in rare situations, where
   150 distinct sorts used to be assigned accidentally.  For example:
   151 
   152   lemma "P (x::'a::foo)" and "Q (y::'a::bar)"  -- "now illegal"
   153 
   154   lemma "P (x::'a)" and "Q (y::'a::bar)"
   155     -- "now uniform 'a::bar instead of default sort for first occurrence (!)"
   156 
   157 * Rule composition via attribute "OF" (or ML functions OF/MRS) is more
   158 tolerant against multiple unifiers, as long as the final result is
   159 unique.  (As before, rules are composed in canonical right-to-left
   160 order to accommodate newly introduced premises.)
   161 
   162 * Renamed some inner syntax categories:
   163 
   164     num ~> num_token
   165     xnum ~> xnum_token
   166     xstr ~> str_token
   167 
   168 Minor INCOMPATIBILITY.  Note that in practice "num_const" or
   169 "num_position" etc. are mainly used instead (which also include
   170 position information via constraints).
   171 
   172 * Simplified configuration options for syntax ambiguity: see
   173 "syntax_ambiguity_warning" and "syntax_ambiguity_limit" in isar-ref
   174 manual.  Minor INCOMPATIBILITY.
   175 
   176 * Discontinued configuration option "syntax_positions": atomic terms
   177 in parse trees are always annotated by position constraints.
   178 
   179 * Old code generator for SML and its commands 'code_module',
   180 'code_library', 'consts_code', 'types_code' have been discontinued.
   181 Use commands of the generic code generator instead.  INCOMPATIBILITY.
   182 
   183 * Redundant attribute "code_inline" has been discontinued. Use
   184 "code_unfold" instead.  INCOMPATIBILITY.
   185 
   186 * Dropped attribute "code_unfold_post" in favor of the its dual
   187 "code_abbrev", which yields a common pattern in definitions like
   188 
   189   definition [code_abbrev]: "f = t"
   190 
   191 INCOMPATIBILITY.
   192 
   193 * Obsolete 'types' command has been discontinued.  Use 'type_synonym'
   194 instead.  INCOMPATIBILITY.
   195 
   196 * Discontinued old "prems" fact, which used to refer to the accidental
   197 collection of foundational premises in the context (already marked as
   198 legacy since Isabelle2011).
   199 
   200 
   201 *** HOL ***
   202 
   203 * Type 'a set is now a proper type constructor (just as before
   204 Isabelle2008).  Definitions mem_def and Collect_def have disappeared.
   205 Non-trivial INCOMPATIBILITY.  For developments keeping predicates and
   206 sets separate, it is often sufficient to rephrase some set S that has
   207 been accidentally used as predicates by "%x. x : S", and some
   208 predicate P that has been accidentally used as set by "{x. P x}".
   209 Corresponding proofs in a first step should be pruned from any
   210 tinkering with former theorems mem_def and Collect_def as far as
   211 possible.
   212 
   213 For developments which deliberately mix predicates and sets, a
   214 planning step is necessary to determine what should become a predicate
   215 and what a set.  It can be helpful to carry out that step in
   216 Isabelle2011-1 before jumping right into the current release.
   217 
   218 * Code generation by default implements sets as container type rather
   219 than predicates.  INCOMPATIBILITY.
   220 
   221 * New type synonym 'a rel = ('a * 'a) set
   222 
   223 * The representation of numerals has changed.  Datatype "num"
   224 represents strictly positive binary numerals, along with functions
   225 "numeral :: num => 'a" and "neg_numeral :: num => 'a" to represent
   226 positive and negated numeric literals, respectively.  See also
   227 definitions in ~~/src/HOL/Num.thy.  Potential INCOMPATIBILITY, some
   228 user theories may require adaptations as follows:
   229 
   230   - Theorems with number_ring or number_semiring constraints: These
   231     classes are gone; use comm_ring_1 or comm_semiring_1 instead.
   232 
   233   - Theories defining numeric types: Remove number, number_semiring,
   234     and number_ring instances. Defer all theorems about numerals until
   235     after classes one and semigroup_add have been instantiated.
   236 
   237   - Numeral-only simp rules: Replace each rule having a "number_of v"
   238     pattern with two copies, one for numeral and one for neg_numeral.
   239 
   240   - Theorems about subclasses of semiring_1 or ring_1: These classes
   241     automatically support numerals now, so more simp rules and
   242     simprocs may now apply within the proof.
   243 
   244   - Definitions and theorems using old constructors Pls/Min/Bit0/Bit1:
   245     Redefine using other integer operations.
   246 
   247 * Transfer: New package intended to generalize the existing
   248 "descending" method and related theorem attributes from the Quotient
   249 package.  (Not all functionality is implemented yet, but future
   250 development will focus on Transfer as an eventual replacement for the
   251 corresponding parts of the Quotient package.)
   252 
   253   - transfer_rule attribute: Maintains a collection of transfer rules,
   254     which relate constants at two different types. Transfer rules may
   255     relate different type instances of the same polymorphic constant,
   256     or they may relate an operation on a raw type to a corresponding
   257     operation on an abstract type (quotient or subtype). For example:
   258 
   259     ((A ===> B) ===> list_all2 A ===> list_all2 B) map map
   260     (cr_int ===> cr_int ===> cr_int) (%(x,y) (u,v). (x+u, y+v)) plus_int
   261 
   262   - transfer method: Replaces a subgoal on abstract types with an
   263     equivalent subgoal on the corresponding raw types. Constants are
   264     replaced with corresponding ones according to the transfer rules.
   265     Goals are generalized over all free variables by default; this is
   266     necessary for variables whose types change, but can be overridden
   267     for specific variables with e.g. "transfer fixing: x y z".  The
   268     variant transfer' method allows replacing a subgoal with one that
   269     is logically stronger (rather than equivalent).
   270 
   271   - relator_eq attribute: Collects identity laws for relators of
   272     various type constructors, e.g. "list_all2 (op =) = (op =)".  The
   273     transfer method uses these lemmas to infer transfer rules for
   274     non-polymorphic constants on the fly.
   275 
   276   - transfer_prover method: Assists with proving a transfer rule for a
   277     new constant, provided the constant is defined in terms of other
   278     constants that already have transfer rules. It should be applied
   279     after unfolding the constant definitions.
   280 
   281   - HOL/ex/Transfer_Int_Nat.thy: Example theory demonstrating transfer
   282     from type nat to type int.
   283 
   284 * Lifting: New package intended to generalize the quotient_definition
   285 facility of the Quotient package; designed to work with Transfer.
   286 
   287   - lift_definition command: Defines operations on an abstract type in
   288     terms of a corresponding operation on a representation
   289     type.  Example syntax:
   290 
   291     lift_definition dlist_insert :: "'a => 'a dlist => 'a dlist"
   292       is List.insert
   293 
   294     Users must discharge a respectfulness proof obligation when each
   295     constant is defined. (For a type copy, i.e. a typedef with UNIV,
   296     the proof is discharged automatically.) The obligation is
   297     presented in a user-friendly, readable form; a respectfulness
   298     theorem in the standard format and a transfer rule are generated
   299     by the package.
   300 
   301   - Integration with code_abstype: For typedefs (e.g. subtypes
   302     corresponding to a datatype invariant, such as dlist),
   303     lift_definition generates a code certificate theorem and sets up
   304     code generation for each constant.
   305 
   306   - setup_lifting command: Sets up the Lifting package to work with a
   307     user-defined type. The user must provide either a quotient theorem
   308     or a type_definition theorem.  The package configures transfer
   309     rules for equality and quantifiers on the type, and sets up the
   310     lift_definition command to work with the type.
   311 
   312   - Usage examples: See Quotient_Examples/Lift_DList.thy,
   313     Quotient_Examples/Lift_RBT.thy, Quotient_Examples/Lift_FSet.thy,
   314     Word/Word.thy and Library/Float.thy.
   315 
   316 * Quotient package:
   317 
   318   - The 'quotient_type' command now supports a 'morphisms' option with
   319     rep and abs functions, similar to typedef.
   320 
   321   - 'quotient_type' sets up new types to work with the Lifting and
   322     Transfer packages, as with 'setup_lifting'.
   323 
   324   - The 'quotient_definition' command now requires the user to prove a
   325     respectfulness property at the point where the constant is
   326     defined, similar to lift_definition; INCOMPATIBILITY.
   327 
   328   - Renamed predicate 'Quotient' to 'Quotient3', and renamed theorems
   329     accordingly, INCOMPATIBILITY.
   330 
   331 * New diagnostic command 'find_unused_assms' to find potentially
   332 superfluous assumptions in theorems using Quickcheck.
   333 
   334 * Quickcheck:
   335 
   336   - Quickcheck returns variable assignments as counterexamples, which
   337     allows to reveal the underspecification of functions under test.
   338     For example, refuting "hd xs = x", it presents the variable
   339     assignment xs = [] and x = a1 as a counterexample, assuming that
   340     any property is false whenever "hd []" occurs in it.
   341 
   342     These counterexample are marked as potentially spurious, as
   343     Quickcheck also returns "xs = []" as a counterexample to the
   344     obvious theorem "hd xs = hd xs".
   345 
   346     After finding a potentially spurious counterexample, Quickcheck
   347     continues searching for genuine ones.
   348 
   349     By default, Quickcheck shows potentially spurious and genuine
   350     counterexamples. The option "genuine_only" sets quickcheck to only
   351     show genuine counterexamples.
   352 
   353   - The command 'quickcheck_generator' creates random and exhaustive
   354     value generators for a given type and operations.
   355 
   356     It generates values by using the operations as if they were
   357     constructors of that type.
   358 
   359   - Support for multisets.
   360 
   361   - Added "use_subtype" options.
   362 
   363   - Added "quickcheck_locale" configuration to specify how to process
   364     conjectures in a locale context.
   365 
   366 * Nitpick: Fixed infinite loop caused by the 'peephole_optim' option
   367 and affecting 'rat' and 'real'.
   368 
   369 * Sledgehammer:
   370   - Integrated more tightly with SPASS, as described in the ITP 2012
   371     paper "More SPASS with Isabelle".
   372   - Made it try "smt" as a fallback if "metis" fails or times out.
   373   - Added support for the following provers: Alt-Ergo (via Why3 and
   374     TFF1), iProver, iProver-Eq.
   375   - Replaced remote E-SInE with remote Satallax in the default setup.
   376   - Sped up the minimizer.
   377   - Added "lam_trans", "uncurry_aliases", and "minimize" options.
   378   - Renamed "slicing" ("no_slicing") option to "slice" ("dont_slice").
   379   - Renamed "sound" option to "strict".
   380 
   381 * Metis: Added possibility to specify lambda translations scheme as a
   382 parenthesized argument (e.g., "by (metis (lifting) ...)").
   383 
   384 * SMT: Renamed "smt_fixed" option to "smt_read_only_certificates".
   385 
   386 * Command 'try0': Renamed from 'try_methods'. INCOMPATIBILITY.
   387 
   388 * New "case_product" attribute to generate a case rule doing multiple
   389 case distinctions at the same time.  E.g.
   390 
   391   list.exhaust [case_product nat.exhaust]
   392 
   393 produces a rule which can be used to perform case distinction on both
   394 a list and a nat.
   395 
   396 * New "eventually_elim" method as a generalized variant of the
   397 eventually_elim* rules.  Supports structured proofs.
   398 
   399 * Typedef with implicit set definition is considered legacy.  Use
   400 "typedef (open)" form instead, which will eventually become the
   401 default.
   402 
   403 * Record: code generation can be switched off manually with
   404 
   405   declare [[record_coden = false]]  -- "default true"
   406 
   407 * Datatype: type parameters allow explicit sort constraints.
   408 
   409 * Concrete syntax for case expressions includes constraints for source
   410 positions, and thus produces Prover IDE markup for its bindings.
   411 INCOMPATIBILITY for old-style syntax translations that augment the
   412 pattern notation; e.g. see src/HOL/HOLCF/One.thy for translations of
   413 one_case.
   414 
   415 * Clarified attribute "mono_set": pure declaration without modifying
   416 the result of the fact expression.
   417 
   418 * More default pred/set conversions on a couple of relation operations
   419 and predicates.  Added powers of predicate relations.  Consolidation
   420 of some relation theorems:
   421 
   422   converse_def ~> converse_unfold
   423   rel_comp_def ~> relcomp_unfold
   424   symp_def ~> (modified, use symp_def and sym_def instead)
   425   transp_def ~> transp_trans
   426   Domain_def ~> Domain_unfold
   427   Range_def ~> Domain_converse [symmetric]
   428 
   429 Generalized theorems INF_INT_eq, INF_INT_eq2, SUP_UN_eq, SUP_UN_eq2.
   430 
   431 See theory "Relation" for examples for making use of pred/set
   432 conversions by means of attributes "to_set" and "to_pred".
   433 
   434 INCOMPATIBILITY.
   435 
   436 * Renamed facts about the power operation on relations, i.e., relpow
   437 to match the constant's name:
   438 
   439   rel_pow_1 ~> relpow_1
   440   rel_pow_0_I ~> relpow_0_I
   441   rel_pow_Suc_I ~> relpow_Suc_I
   442   rel_pow_Suc_I2 ~> relpow_Suc_I2
   443   rel_pow_0_E ~> relpow_0_E
   444   rel_pow_Suc_E ~> relpow_Suc_E
   445   rel_pow_E ~> relpow_E
   446   rel_pow_Suc_D2 ~> relpow_Suc_D2
   447   rel_pow_Suc_E2 ~> relpow_Suc_E2
   448   rel_pow_Suc_D2' ~> relpow_Suc_D2'
   449   rel_pow_E2 ~> relpow_E2
   450   rel_pow_add ~> relpow_add
   451   rel_pow_commute ~> relpow
   452   rel_pow_empty ~> relpow_empty:
   453   rtrancl_imp_UN_rel_pow ~> rtrancl_imp_UN_relpow
   454   rel_pow_imp_rtrancl ~> relpow_imp_rtrancl
   455   rtrancl_is_UN_rel_pow ~> rtrancl_is_UN_relpow
   456   rtrancl_imp_rel_pow ~> rtrancl_imp_relpow
   457   rel_pow_fun_conv ~> relpow_fun_conv
   458   rel_pow_finite_bounded1 ~> relpow_finite_bounded1
   459   rel_pow_finite_bounded ~> relpow_finite_bounded
   460   rtrancl_finite_eq_rel_pow ~> rtrancl_finite_eq_relpow
   461   trancl_finite_eq_rel_pow ~> trancl_finite_eq_relpow
   462   single_valued_rel_pow ~> single_valued_relpow
   463 
   464 INCOMPATIBILITY.
   465 
   466 * Theory Relation: Consolidated constant name for relation composition
   467 and corresponding theorem names:
   468 
   469   - Renamed constant rel_comp to relcomp.
   470 
   471   - Dropped abbreviation pred_comp. Use relcompp instead.
   472 
   473   - Renamed theorems:
   474 
   475     rel_compI ~> relcompI
   476     rel_compEpair ~> relcompEpair
   477     rel_compE ~> relcompE
   478     pred_comp_rel_comp_eq ~> relcompp_relcomp_eq
   479     rel_comp_empty1 ~> relcomp_empty1
   480     rel_comp_mono ~> relcomp_mono
   481     rel_comp_subset_Sigma ~> relcomp_subset_Sigma
   482     rel_comp_distrib ~> relcomp_distrib
   483     rel_comp_distrib2 ~> relcomp_distrib2
   484     rel_comp_UNION_distrib ~> relcomp_UNION_distrib
   485     rel_comp_UNION_distrib2 ~> relcomp_UNION_distrib2
   486     single_valued_rel_comp ~> single_valued_relcomp
   487     rel_comp_def ~> relcomp_unfold
   488     converse_rel_comp ~> converse_relcomp
   489     pred_compI ~> relcomppI
   490     pred_compE ~> relcomppE
   491     pred_comp_bot1 ~> relcompp_bot1
   492     pred_comp_bot2 ~> relcompp_bot2
   493     transp_pred_comp_less_eq ~> transp_relcompp_less_eq
   494     pred_comp_mono ~> relcompp_mono
   495     pred_comp_distrib ~> relcompp_distrib
   496     pred_comp_distrib2 ~> relcompp_distrib2
   497     converse_pred_comp ~> converse_relcompp
   498 
   499     finite_rel_comp ~> finite_relcomp
   500 
   501     set_rel_comp ~> set_relcomp
   502 
   503 INCOMPATIBILITY.
   504 
   505 * Theory Divides: Discontinued redundant theorems about div and mod.
   506 INCOMPATIBILITY, use the corresponding generic theorems instead.
   507 
   508   DIVISION_BY_ZERO ~> div_by_0, mod_by_0
   509   zdiv_self ~> div_self
   510   zmod_self ~> mod_self
   511   zdiv_zero ~> div_0
   512   zmod_zero ~> mod_0
   513   zdiv_zmod_equality ~> div_mod_equality2
   514   zdiv_zmod_equality2 ~> div_mod_equality
   515   zmod_zdiv_trivial ~> mod_div_trivial
   516   zdiv_zminus_zminus ~> div_minus_minus
   517   zmod_zminus_zminus ~> mod_minus_minus
   518   zdiv_zminus2 ~> div_minus_right
   519   zmod_zminus2 ~> mod_minus_right
   520   zdiv_minus1_right ~> div_minus1_right
   521   zmod_minus1_right ~> mod_minus1_right
   522   zdvd_mult_div_cancel ~> dvd_mult_div_cancel
   523   zmod_zmult1_eq ~> mod_mult_right_eq
   524   zpower_zmod ~> power_mod
   525   zdvd_zmod ~> dvd_mod
   526   zdvd_zmod_imp_zdvd ~> dvd_mod_imp_dvd
   527   mod_mult_distrib ~> mult_mod_left
   528   mod_mult_distrib2 ~> mult_mod_right
   529 
   530 * Removed redundant theorems nat_mult_2 and nat_mult_2_right; use
   531 generic mult_2 and mult_2_right instead. INCOMPATIBILITY.
   532 
   533 * Finite_Set.fold now qualified.  INCOMPATIBILITY.
   534 
   535 * Consolidated theorem names concerning fold combinators:
   536 
   537   inf_INFI_fold_inf ~> inf_INF_fold_inf
   538   sup_SUPR_fold_sup ~> sup_SUP_fold_sup
   539   INFI_fold_inf ~> INF_fold_inf
   540   SUPR_fold_sup ~> SUP_fold_sup
   541   union_set ~> union_set_fold
   542   minus_set ~> minus_set_fold
   543   INFI_set_fold ~> INF_set_fold
   544   SUPR_set_fold ~> SUP_set_fold
   545   INF_code ~> INF_set_foldr
   546   SUP_code ~> SUP_set_foldr
   547   foldr.simps ~> foldr.simps (in point-free formulation)
   548   foldr_fold_rev ~> foldr_conv_fold
   549   foldl_fold ~> foldl_conv_fold
   550   foldr_foldr ~> foldr_conv_foldl
   551   foldl_foldr ~> foldl_conv_foldr
   552   fold_set_remdups ~> fold_set_fold_remdups
   553   fold_set ~> fold_set_fold
   554   fold1_set ~> fold1_set_fold
   555 
   556 INCOMPATIBILITY.
   557 
   558 * Dropped rarely useful theorems concerning fold combinators:
   559 foldl_apply, foldl_fun_comm, foldl_rev, fold_weak_invariant,
   560 rev_foldl_cons, fold_set_remdups, fold_set, fold_set1,
   561 concat_conv_foldl, foldl_weak_invariant, foldl_invariant,
   562 foldr_invariant, foldl_absorb0, foldl_foldr1_lemma, foldl_foldr1,
   563 listsum_conv_fold, listsum_foldl, sort_foldl_insort, foldl_assoc,
   564 foldr_conv_foldl, start_le_sum, elem_le_sum, sum_eq_0_conv.
   565 INCOMPATIBILITY.  For the common phrases "%xs. List.foldr plus xs 0"
   566 and "List.foldl plus 0", prefer "List.listsum".  Otherwise it can be
   567 useful to boil down "List.foldr" and "List.foldl" to "List.fold" by
   568 unfolding "foldr_conv_fold" and "foldl_conv_fold".
   569 
   570 * Dropped lemmas minus_set_foldr, union_set_foldr, union_coset_foldr,
   571 inter_coset_foldr, Inf_fin_set_foldr, Sup_fin_set_foldr,
   572 Min_fin_set_foldr, Max_fin_set_foldr, Inf_set_foldr, Sup_set_foldr,
   573 INF_set_foldr, SUP_set_foldr.  INCOMPATIBILITY.  Prefer corresponding
   574 lemmas over fold rather than foldr, or make use of lemmas
   575 fold_conv_foldr and fold_rev.
   576 
   577 * Congruence rules Option.map_cong and Option.bind_cong for recursion
   578 through option types.
   579 
   580 * "Transitive_Closure.ntrancl": bounded transitive closure on
   581 relations.
   582 
   583 * Constant "Set.not_member" now qualified.  INCOMPATIBILITY.
   584 
   585 * Theory Int: Discontinued many legacy theorems specific to type int.
   586 INCOMPATIBILITY, use the corresponding generic theorems instead.
   587 
   588   zminus_zminus ~> minus_minus
   589   zminus_0 ~> minus_zero
   590   zminus_zadd_distrib ~> minus_add_distrib
   591   zadd_commute ~> add_commute
   592   zadd_assoc ~> add_assoc
   593   zadd_left_commute ~> add_left_commute
   594   zadd_ac ~> add_ac
   595   zmult_ac ~> mult_ac
   596   zadd_0 ~> add_0_left
   597   zadd_0_right ~> add_0_right
   598   zadd_zminus_inverse2 ~> left_minus
   599   zmult_zminus ~> mult_minus_left
   600   zmult_commute ~> mult_commute
   601   zmult_assoc ~> mult_assoc
   602   zadd_zmult_distrib ~> left_distrib
   603   zadd_zmult_distrib2 ~> right_distrib
   604   zdiff_zmult_distrib ~> left_diff_distrib
   605   zdiff_zmult_distrib2 ~> right_diff_distrib
   606   zmult_1 ~> mult_1_left
   607   zmult_1_right ~> mult_1_right
   608   zle_refl ~> order_refl
   609   zle_trans ~> order_trans
   610   zle_antisym ~> order_antisym
   611   zle_linear ~> linorder_linear
   612   zless_linear ~> linorder_less_linear
   613   zadd_left_mono ~> add_left_mono
   614   zadd_strict_right_mono ~> add_strict_right_mono
   615   zadd_zless_mono ~> add_less_le_mono
   616   int_0_less_1 ~> zero_less_one
   617   int_0_neq_1 ~> zero_neq_one
   618   zless_le ~> less_le
   619   zpower_zadd_distrib ~> power_add
   620   zero_less_zpower_abs_iff ~> zero_less_power_abs_iff
   621   zero_le_zpower_abs ~> zero_le_power_abs
   622 
   623 * Theory Deriv: Renamed
   624 
   625   DERIV_nonneg_imp_nonincreasing ~> DERIV_nonneg_imp_nondecreasing
   626 
   627 * Theory Library/Multiset: Improved code generation of multisets.
   628 
   629 * Theory HOL/Library/Set_Algebras: Addition and multiplication on sets
   630 are expressed via type classes again. The special syntax
   631 \<oplus>/\<otimes> has been replaced by plain +/*. Removed constant
   632 setsum_set, which is now subsumed by Big_Operators.setsum.
   633 INCOMPATIBILITY.
   634 
   635 * Theory HOL/Library/Diagonalize has been removed. INCOMPATIBILITY,
   636 use theory HOL/Library/Nat_Bijection instead.
   637 
   638 * Theory HOL/Library/RBT_Impl: Backing implementation of red-black
   639 trees is now inside a type class context.  Names of affected
   640 operations and lemmas have been prefixed by rbt_.  INCOMPATIBILITY for
   641 theories working directly with raw red-black trees, adapt the names as
   642 follows:
   643 
   644   Operations:
   645   bulkload -> rbt_bulkload
   646   del_from_left -> rbt_del_from_left
   647   del_from_right -> rbt_del_from_right
   648   del -> rbt_del
   649   delete -> rbt_delete
   650   ins -> rbt_ins
   651   insert -> rbt_insert
   652   insertw -> rbt_insert_with
   653   insert_with_key -> rbt_insert_with_key
   654   map_entry -> rbt_map_entry
   655   lookup -> rbt_lookup
   656   sorted -> rbt_sorted
   657   tree_greater -> rbt_greater
   658   tree_less -> rbt_less
   659   tree_less_symbol -> rbt_less_symbol
   660   union -> rbt_union
   661   union_with -> rbt_union_with
   662   union_with_key -> rbt_union_with_key
   663 
   664   Lemmas:
   665   balance_left_sorted -> balance_left_rbt_sorted
   666   balance_left_tree_greater -> balance_left_rbt_greater
   667   balance_left_tree_less -> balance_left_rbt_less
   668   balance_right_sorted -> balance_right_rbt_sorted
   669   balance_right_tree_greater -> balance_right_rbt_greater
   670   balance_right_tree_less -> balance_right_rbt_less
   671   balance_sorted -> balance_rbt_sorted
   672   balance_tree_greater -> balance_rbt_greater
   673   balance_tree_less -> balance_rbt_less
   674   bulkload_is_rbt -> rbt_bulkload_is_rbt
   675   combine_sorted -> combine_rbt_sorted
   676   combine_tree_greater -> combine_rbt_greater
   677   combine_tree_less -> combine_rbt_less
   678   delete_in_tree -> rbt_delete_in_tree
   679   delete_is_rbt -> rbt_delete_is_rbt
   680   del_from_left_tree_greater -> rbt_del_from_left_rbt_greater
   681   del_from_left_tree_less -> rbt_del_from_left_rbt_less
   682   del_from_right_tree_greater -> rbt_del_from_right_rbt_greater
   683   del_from_right_tree_less -> rbt_del_from_right_rbt_less
   684   del_in_tree -> rbt_del_in_tree
   685   del_inv1_inv2 -> rbt_del_inv1_inv2
   686   del_sorted -> rbt_del_rbt_sorted
   687   del_tree_greater -> rbt_del_rbt_greater
   688   del_tree_less -> rbt_del_rbt_less
   689   dom_lookup_Branch -> dom_rbt_lookup_Branch
   690   entries_lookup -> entries_rbt_lookup
   691   finite_dom_lookup -> finite_dom_rbt_lookup
   692   insert_sorted -> rbt_insert_rbt_sorted
   693   insertw_is_rbt -> rbt_insertw_is_rbt
   694   insertwk_is_rbt -> rbt_insertwk_is_rbt
   695   insertwk_sorted -> rbt_insertwk_rbt_sorted
   696   insertw_sorted -> rbt_insertw_rbt_sorted
   697   ins_sorted -> ins_rbt_sorted
   698   ins_tree_greater -> ins_rbt_greater
   699   ins_tree_less -> ins_rbt_less
   700   is_rbt_sorted -> is_rbt_rbt_sorted
   701   lookup_balance -> rbt_lookup_balance
   702   lookup_bulkload -> rbt_lookup_rbt_bulkload
   703   lookup_delete -> rbt_lookup_rbt_delete
   704   lookup_Empty -> rbt_lookup_Empty
   705   lookup_from_in_tree -> rbt_lookup_from_in_tree
   706   lookup_in_tree -> rbt_lookup_in_tree
   707   lookup_ins -> rbt_lookup_ins
   708   lookup_insert -> rbt_lookup_rbt_insert
   709   lookup_insertw -> rbt_lookup_rbt_insertw
   710   lookup_insertwk -> rbt_lookup_rbt_insertwk
   711   lookup_keys -> rbt_lookup_keys
   712   lookup_map -> rbt_lookup_map
   713   lookup_map_entry -> rbt_lookup_rbt_map_entry
   714   lookup_tree_greater -> rbt_lookup_rbt_greater
   715   lookup_tree_less -> rbt_lookup_rbt_less
   716   lookup_union -> rbt_lookup_rbt_union
   717   map_entry_color_of -> rbt_map_entry_color_of
   718   map_entry_inv1 -> rbt_map_entry_inv1
   719   map_entry_inv2 -> rbt_map_entry_inv2
   720   map_entry_is_rbt -> rbt_map_entry_is_rbt
   721   map_entry_sorted -> rbt_map_entry_rbt_sorted
   722   map_entry_tree_greater -> rbt_map_entry_rbt_greater
   723   map_entry_tree_less -> rbt_map_entry_rbt_less
   724   map_tree_greater -> map_rbt_greater
   725   map_tree_less -> map_rbt_less
   726   map_sorted -> map_rbt_sorted
   727   paint_sorted -> paint_rbt_sorted
   728   paint_lookup -> paint_rbt_lookup
   729   paint_tree_greater -> paint_rbt_greater
   730   paint_tree_less -> paint_rbt_less
   731   sorted_entries -> rbt_sorted_entries
   732   tree_greater_eq_trans -> rbt_greater_eq_trans
   733   tree_greater_nit -> rbt_greater_nit
   734   tree_greater_prop -> rbt_greater_prop
   735   tree_greater_simps -> rbt_greater_simps
   736   tree_greater_trans -> rbt_greater_trans
   737   tree_less_eq_trans -> rbt_less_eq_trans
   738   tree_less_nit -> rbt_less_nit
   739   tree_less_prop -> rbt_less_prop
   740   tree_less_simps -> rbt_less_simps
   741   tree_less_trans -> rbt_less_trans
   742   tree_ord_props -> rbt_ord_props
   743   union_Branch -> rbt_union_Branch
   744   union_is_rbt -> rbt_union_is_rbt
   745   unionw_is_rbt -> rbt_unionw_is_rbt
   746   unionwk_is_rbt -> rbt_unionwk_is_rbt
   747   unionwk_sorted -> rbt_unionwk_rbt_sorted
   748 
   749 * Theory HOL/Library/Float: Floating point numbers are now defined as
   750 a subset of the real numbers.  All operations are defined using the
   751 lifing-framework and proofs use the transfer method.  INCOMPATIBILITY.
   752 
   753   Changed Operations:
   754   float_abs -> abs
   755   float_nprt -> nprt
   756   float_pprt -> pprt
   757   pow2 -> use powr
   758   round_down -> float_round_down
   759   round_up -> float_round_up
   760   scale -> exponent
   761 
   762   Removed Operations:
   763   ceiling_fl, lb_mult, lb_mod, ub_mult, ub_mod
   764 
   765   Renamed Lemmas:
   766   abs_float_def -> Float.compute_float_abs
   767   bitlen_ge0 -> bitlen_nonneg
   768   bitlen.simps -> Float.compute_bitlen
   769   float_components -> Float_mantissa_exponent
   770   float_divl.simps -> Float.compute_float_divl
   771   float_divr.simps -> Float.compute_float_divr
   772   float_eq_odd -> mult_powr_eq_mult_powr_iff
   773   float_power -> real_of_float_power
   774   lapprox_posrat_def -> Float.compute_lapprox_posrat
   775   lapprox_rat.simps -> Float.compute_lapprox_rat
   776   le_float_def' -> Float.compute_float_le
   777   le_float_def -> less_eq_float.rep_eq
   778   less_float_def' -> Float.compute_float_less
   779   less_float_def -> less_float.rep_eq
   780   normfloat_def -> Float.compute_normfloat
   781   normfloat_imp_odd_or_zero -> mantissa_not_dvd and mantissa_noteq_0
   782   normfloat -> normfloat_def
   783   normfloat_unique -> use normfloat_def
   784   number_of_float_Float -> Float.compute_float_numeral, Float.compute_float_neg_numeral
   785   one_float_def -> Float.compute_float_one
   786   plus_float_def -> Float.compute_float_plus
   787   rapprox_posrat_def -> Float.compute_rapprox_posrat
   788   rapprox_rat.simps -> Float.compute_rapprox_rat
   789   real_of_float_0 -> zero_float.rep_eq
   790   real_of_float_1 -> one_float.rep_eq
   791   real_of_float_abs -> abs_float.rep_eq
   792   real_of_float_add -> plus_float.rep_eq
   793   real_of_float_minus -> uminus_float.rep_eq
   794   real_of_float_mult -> times_float.rep_eq
   795   real_of_float_simp -> Float.rep_eq
   796   real_of_float_sub -> minus_float.rep_eq
   797   round_down.simps -> Float.compute_float_round_down
   798   round_up.simps -> Float.compute_float_round_up
   799   times_float_def -> Float.compute_float_times
   800   uminus_float_def -> Float.compute_float_uminus
   801   zero_float_def -> Float.compute_float_zero
   802 
   803   Lemmas not necessary anymore, use the transfer method:
   804   bitlen_B0, bitlen_B1, bitlen_ge1, bitlen_Min, bitlen_Pls, float_divl,
   805   float_divr, float_le_simp, float_less1_mantissa_bound,
   806   float_less_simp, float_less_zero, float_le_zero,
   807   float_pos_less1_e_neg, float_pos_m_pos, float_split, float_split2,
   808   floor_pos_exp, lapprox_posrat, lapprox_posrat_bottom, lapprox_rat,
   809   lapprox_rat_bottom, normalized_float, rapprox_posrat,
   810   rapprox_posrat_le1, rapprox_rat, real_of_float_ge0_exp,
   811   real_of_float_neg_exp, real_of_float_nge0_exp, round_down floor_fl,
   812   round_up, zero_le_float, zero_less_float
   813 
   814 * New theory HOL/Library/DAList provides an abstract type for
   815 association lists with distinct keys.
   816 
   817 * Session HOL/IMP: Added new theory of abstract interpretation of
   818 annotated commands.
   819 
   820 * Session HOL-Import: Re-implementation from scratch is faster,
   821 simpler, and more scalable.  Requires a proof bundle, which is
   822 available as an external component.  Discontinued old (and mostly
   823 dead) Importer for HOL4 and HOL Light.  INCOMPATIBILITY.
   824 
   825 * Session HOL-Word: Discontinued many redundant theorems specific to
   826 type 'a word. INCOMPATIBILITY, use the corresponding generic theorems
   827 instead.
   828 
   829   word_sub_alt ~> word_sub_wi
   830   word_add_alt ~> word_add_def
   831   word_mult_alt ~> word_mult_def
   832   word_minus_alt ~> word_minus_def
   833   word_0_alt ~> word_0_wi
   834   word_1_alt ~> word_1_wi
   835   word_add_0 ~> add_0_left
   836   word_add_0_right ~> add_0_right
   837   word_mult_1 ~> mult_1_left
   838   word_mult_1_right ~> mult_1_right
   839   word_add_commute ~> add_commute
   840   word_add_assoc ~> add_assoc
   841   word_add_left_commute ~> add_left_commute
   842   word_mult_commute ~> mult_commute
   843   word_mult_assoc ~> mult_assoc
   844   word_mult_left_commute ~> mult_left_commute
   845   word_left_distrib ~> left_distrib
   846   word_right_distrib ~> right_distrib
   847   word_left_minus ~> left_minus
   848   word_diff_0_right ~> diff_0_right
   849   word_diff_self ~> diff_self
   850   word_sub_def ~> diff_minus
   851   word_diff_minus ~> diff_minus
   852   word_add_ac ~> add_ac
   853   word_mult_ac ~> mult_ac
   854   word_plus_ac0 ~> add_0_left add_0_right add_ac
   855   word_times_ac1 ~> mult_1_left mult_1_right mult_ac
   856   word_order_trans ~> order_trans
   857   word_order_refl ~> order_refl
   858   word_order_antisym ~> order_antisym
   859   word_order_linear ~> linorder_linear
   860   lenw1_zero_neq_one ~> zero_neq_one
   861   word_number_of_eq ~> number_of_eq
   862   word_of_int_add_hom ~> wi_hom_add
   863   word_of_int_sub_hom ~> wi_hom_sub
   864   word_of_int_mult_hom ~> wi_hom_mult
   865   word_of_int_minus_hom ~> wi_hom_neg
   866   word_of_int_succ_hom ~> wi_hom_succ
   867   word_of_int_pred_hom ~> wi_hom_pred
   868   word_of_int_0_hom ~> word_0_wi
   869   word_of_int_1_hom ~> word_1_wi
   870 
   871 * Session HOL-Word: New proof method "word_bitwise" for splitting
   872 machine word equalities and inequalities into logical circuits,
   873 defined in HOL/Word/WordBitwise.thy.  Supports addition, subtraction,
   874 multiplication, shifting by constants, bitwise operators and numeric
   875 constants.  Requires fixed-length word types, not 'a word.  Solves
   876 many standard word identities outright and converts more into first
   877 order problems amenable to blast or similar.  See also examples in
   878 HOL/Word/Examples/WordExamples.thy.
   879 
   880 * Session HOL-Probability: Introduced the type "'a measure" to
   881 represent measures, this replaces the records 'a algebra and 'a
   882 measure_space.  The locales based on subset_class now have two
   883 locale-parameters the space \<Omega> and the set of measurable sets M.
   884 The product of probability spaces uses now the same constant as the
   885 finite product of sigma-finite measure spaces "PiM :: ('i => 'a)
   886 measure".  Most constants are defined now outside of locales and gain
   887 an additional parameter, like null_sets, almost_eventually or \<mu>'.
   888 Measure space constructions for distributions and densities now got
   889 their own constants distr and density.  Instead of using locales to
   890 describe measure spaces with a finite space, the measure count_space
   891 and point_measure is introduced.  INCOMPATIBILITY.
   892 
   893   Renamed constants:
   894   measure -> emeasure
   895   finite_measure.\<mu>' -> measure
   896   product_algebra_generator -> prod_algebra
   897   product_prob_space.emb -> prod_emb
   898   product_prob_space.infprod_algebra -> PiM
   899 
   900   Removed locales:
   901   completeable_measure_space
   902   finite_measure_space
   903   finite_prob_space
   904   finite_product_finite_prob_space
   905   finite_product_sigma_algebra
   906   finite_sigma_algebra
   907   measure_space
   908   pair_finite_prob_space
   909   pair_finite_sigma_algebra
   910   pair_finite_space
   911   pair_sigma_algebra
   912   product_sigma_algebra
   913 
   914   Removed constants:
   915   conditional_space
   916   distribution -> use distr measure, or distributed predicate
   917   image_space
   918   joint_distribution -> use distr measure, or distributed predicate
   919   pair_measure_generator
   920   product_prob_space.infprod_algebra -> use PiM
   921   subvimage
   922 
   923   Replacement theorems:
   924   finite_additivity_sufficient -> ring_of_sets.countably_additiveI_finite
   925   finite_measure.empty_measure -> measure_empty
   926   finite_measure.finite_continuity_from_above -> finite_measure.finite_Lim_measure_decseq
   927   finite_measure.finite_continuity_from_below -> finite_measure.finite_Lim_measure_incseq
   928   finite_measure.finite_measure_countably_subadditive -> finite_measure.finite_measure_subadditive_countably
   929   finite_measure.finite_measure_eq -> finite_measure.emeasure_eq_measure
   930   finite_measure.finite_measure -> finite_measure.emeasure_finite
   931   finite_measure.finite_measure_finite_singleton -> finite_measure.finite_measure_eq_setsum_singleton
   932   finite_measure.positive_measure' -> measure_nonneg
   933   finite_measure.real_measure -> finite_measure.emeasure_real
   934   finite_product_prob_space.finite_measure_times -> finite_product_prob_space.finite_measure_PiM_emb
   935   finite_product_sigma_algebra.in_P -> sets_PiM_I_finite
   936   finite_product_sigma_algebra.P_empty -> space_PiM_empty, sets_PiM_empty
   937   information_space.conditional_entropy_eq -> information_space.conditional_entropy_simple_distributed
   938   information_space.conditional_entropy_positive -> information_space.conditional_entropy_nonneg_simple
   939   information_space.conditional_mutual_information_eq_mutual_information -> information_space.conditional_mutual_information_eq_mutual_information_simple
   940   information_space.conditional_mutual_information_generic_positive -> information_space.conditional_mutual_information_nonneg_simple
   941   information_space.conditional_mutual_information_positive -> information_space.conditional_mutual_information_nonneg_simple
   942   information_space.entropy_commute -> information_space.entropy_commute_simple
   943   information_space.entropy_eq -> information_space.entropy_simple_distributed
   944   information_space.entropy_generic_eq -> information_space.entropy_simple_distributed
   945   information_space.entropy_positive -> information_space.entropy_nonneg_simple
   946   information_space.entropy_uniform_max -> information_space.entropy_uniform
   947   information_space.KL_eq_0_imp -> information_space.KL_eq_0_iff_eq
   948   information_space.KL_eq_0 -> information_space.KL_same_eq_0
   949   information_space.KL_ge_0 -> information_space.KL_nonneg
   950   information_space.mutual_information_eq -> information_space.mutual_information_simple_distributed
   951   information_space.mutual_information_positive -> information_space.mutual_information_nonneg_simple
   952   Int_stable_cuboids -> Int_stable_atLeastAtMost
   953   Int_stable_product_algebra_generator -> positive_integral
   954   measure_preserving -> equality "distr M N f = N" "f : measurable M N"
   955   measure_space.additive -> emeasure_additive
   956   measure_space.AE_iff_null_set -> AE_iff_null
   957   measure_space.almost_everywhere_def -> eventually_ae_filter
   958   measure_space.almost_everywhere_vimage -> AE_distrD
   959   measure_space.continuity_from_above -> INF_emeasure_decseq
   960   measure_space.continuity_from_above_Lim -> Lim_emeasure_decseq
   961   measure_space.continuity_from_below_Lim -> Lim_emeasure_incseq
   962   measure_space.continuity_from_below -> SUP_emeasure_incseq
   963   measure_space_density -> emeasure_density
   964   measure_space.density_is_absolutely_continuous -> absolutely_continuousI_density
   965   measure_space.integrable_vimage -> integrable_distr
   966   measure_space.integral_translated_density -> integral_density
   967   measure_space.integral_vimage -> integral_distr
   968   measure_space.measure_additive -> plus_emeasure
   969   measure_space.measure_compl -> emeasure_compl
   970   measure_space.measure_countable_increasing -> emeasure_countable_increasing
   971   measure_space.measure_countably_subadditive -> emeasure_subadditive_countably
   972   measure_space.measure_decseq -> decseq_emeasure
   973   measure_space.measure_Diff -> emeasure_Diff
   974   measure_space.measure_Diff_null_set -> emeasure_Diff_null_set
   975   measure_space.measure_eq_0 -> emeasure_eq_0
   976   measure_space.measure_finitely_subadditive -> emeasure_subadditive_finite
   977   measure_space.measure_finite_singleton -> emeasure_eq_setsum_singleton
   978   measure_space.measure_incseq -> incseq_emeasure
   979   measure_space.measure_insert -> emeasure_insert
   980   measure_space.measure_mono -> emeasure_mono
   981   measure_space.measure_not_negative -> emeasure_not_MInf
   982   measure_space.measure_preserving_Int_stable -> measure_eqI_generator_eq
   983   measure_space.measure_setsum -> setsum_emeasure
   984   measure_space.measure_setsum_split -> setsum_emeasure_cover
   985   measure_space.measure_space_vimage -> emeasure_distr
   986   measure_space.measure_subadditive_finite -> emeasure_subadditive_finite
   987   measure_space.measure_subadditive -> subadditive
   988   measure_space.measure_top -> emeasure_space
   989   measure_space.measure_UN_eq_0 -> emeasure_UN_eq_0
   990   measure_space.measure_Un_null_set -> emeasure_Un_null_set
   991   measure_space.positive_integral_translated_density -> positive_integral_density
   992   measure_space.positive_integral_vimage -> positive_integral_distr
   993   measure_space.real_continuity_from_above -> Lim_measure_decseq
   994   measure_space.real_continuity_from_below -> Lim_measure_incseq
   995   measure_space.real_measure_countably_subadditive -> measure_subadditive_countably
   996   measure_space.real_measure_Diff -> measure_Diff
   997   measure_space.real_measure_finite_Union -> measure_finite_Union
   998   measure_space.real_measure_setsum_singleton -> measure_eq_setsum_singleton
   999   measure_space.real_measure_subadditive -> measure_subadditive
  1000   measure_space.real_measure_Union -> measure_Union
  1001   measure_space.real_measure_UNION -> measure_UNION
  1002   measure_space.simple_function_vimage -> simple_function_comp
  1003   measure_space.simple_integral_vimage -> simple_integral_distr
  1004   measure_space.simple_integral_vimage -> simple_integral_distr
  1005   measure_unique_Int_stable -> measure_eqI_generator_eq
  1006   measure_unique_Int_stable_vimage -> measure_eqI_generator_eq
  1007   pair_sigma_algebra.measurable_cut_fst -> sets_Pair1
  1008   pair_sigma_algebra.measurable_cut_snd -> sets_Pair2
  1009   pair_sigma_algebra.measurable_pair_image_fst -> measurable_Pair1
  1010   pair_sigma_algebra.measurable_pair_image_snd -> measurable_Pair2
  1011   pair_sigma_algebra.measurable_product_swap -> measurable_pair_swap_iff
  1012   pair_sigma_algebra.pair_sigma_algebra_measurable -> measurable_pair_swap
  1013   pair_sigma_algebra.pair_sigma_algebra_swap_measurable -> measurable_pair_swap'
  1014   pair_sigma_algebra.sets_swap -> sets_pair_swap
  1015   pair_sigma_finite.measure_cut_measurable_fst -> pair_sigma_finite.measurable_emeasure_Pair1
  1016   pair_sigma_finite.measure_cut_measurable_snd -> pair_sigma_finite.measurable_emeasure_Pair2
  1017   pair_sigma_finite.measure_preserving_swap -> pair_sigma_finite.distr_pair_swap
  1018   pair_sigma_finite.pair_measure_alt2 -> pair_sigma_finite.emeasure_pair_measure_alt2
  1019   pair_sigma_finite.pair_measure_alt -> pair_sigma_finite.emeasure_pair_measure_alt
  1020   pair_sigma_finite.pair_measure_times -> pair_sigma_finite.emeasure_pair_measure_Times
  1021   prob_space.indep_distribution_eq_measure -> prob_space.indep_vars_iff_distr_eq_PiM
  1022   prob_space.indep_var_distributionD -> prob_space.indep_var_distribution_eq
  1023   prob_space.measure_space_1 -> prob_space.emeasure_space_1
  1024   prob_space.prob_space_vimage -> prob_space_distr
  1025   prob_space.random_variable_restrict -> measurable_restrict
  1026   prob_space_unique_Int_stable -> measure_eqI_prob_space
  1027   product_algebraE -> prod_algebraE_all
  1028   product_algebra_generator_der -> prod_algebra_eq_finite
  1029   product_algebra_generator_into_space -> prod_algebra_sets_into_space
  1030   product_algebraI -> sets_PiM_I_finite
  1031   product_measure_exists -> product_sigma_finite.sigma_finite
  1032   product_prob_space.finite_index_eq_finite_product -> product_prob_space.sets_PiM_generator
  1033   product_prob_space.finite_measure_infprod_emb_Pi -> product_prob_space.measure_PiM_emb
  1034   product_prob_space.infprod_spec -> product_prob_space.emeasure_PiM_emb_not_empty
  1035   product_prob_space.measurable_component -> measurable_component_singleton
  1036   product_prob_space.measurable_emb -> measurable_prod_emb
  1037   product_prob_space.measurable_into_infprod_algebra -> measurable_PiM_single
  1038   product_prob_space.measurable_singleton_infprod -> measurable_component_singleton
  1039   product_prob_space.measure_emb -> emeasure_prod_emb
  1040   product_prob_space.measure_preserving_restrict -> product_prob_space.distr_restrict
  1041   product_sigma_algebra.product_algebra_into_space -> space_closed
  1042   product_sigma_finite.measure_fold -> product_sigma_finite.distr_merge
  1043   product_sigma_finite.measure_preserving_component_singelton -> product_sigma_finite.distr_singleton
  1044   product_sigma_finite.measure_preserving_merge -> product_sigma_finite.distr_merge
  1045   sequence_space.measure_infprod -> sequence_space.measure_PiM_countable
  1046   sets_product_algebra -> sets_PiM
  1047   sigma_algebra.measurable_sigma -> measurable_measure_of
  1048   sigma_finite_measure.disjoint_sigma_finite -> sigma_finite_disjoint
  1049   sigma_finite_measure.RN_deriv_vimage -> sigma_finite_measure.RN_deriv_distr
  1050   sigma_product_algebra_sigma_eq -> sigma_prod_algebra_sigma_eq
  1051   space_product_algebra -> space_PiM
  1052 
  1053 * Session HOL-TPTP: support to parse and import TPTP problems (all
  1054 languages) into Isabelle/HOL.
  1055 
  1056 
  1057 *** FOL ***
  1058 
  1059 * New "case_product" attribute (see HOL).
  1060 
  1061 
  1062 *** ZF ***
  1063 
  1064 * Greater support for structured proofs involving induction or case
  1065 analysis.
  1066 
  1067 * Much greater use of mathematical symbols.
  1068 
  1069 * Removal of many ML theorem bindings.  INCOMPATIBILITY.
  1070 
  1071 
  1072 *** ML ***
  1073 
  1074 * Antiquotation @{keyword "name"} produces a parser for outer syntax
  1075 from a minor keyword introduced via theory header declaration.
  1076 
  1077 * Antiquotation @{command_spec "name"} produces the
  1078 Outer_Syntax.command_spec from a major keyword introduced via theory
  1079 header declaration; it can be passed to Outer_Syntax.command etc.
  1080 
  1081 * Local_Theory.define no longer hard-wires default theorem name
  1082 "foo_def", but retains the binding as given.  If that is Binding.empty
  1083 / Attrib.empty_binding, the result is not registered as user-level
  1084 fact.  The Local_Theory.define_internal variant allows to specify a
  1085 non-empty name (used for the foundation in the background theory),
  1086 while omitting the fact binding in the user-context.  Potential
  1087 INCOMPATIBILITY for derived definitional packages: need to specify
  1088 naming policy for primitive definitions more explicitly.
  1089 
  1090 * Renamed Thm.capply to Thm.apply, and Thm.cabs to Thm.lambda in
  1091 conformance with similar operations in structure Term and Logic.
  1092 
  1093 * Antiquotation @{attributes [...]} embeds attribute source
  1094 representation into the ML text, which is particularly useful with
  1095 declarations like Local_Theory.note.
  1096 
  1097 * Structure Proof_Context follows standard naming scheme.  Old
  1098 ProofContext has been discontinued.  INCOMPATIBILITY.
  1099 
  1100 * Refined Local_Theory.declaration {syntax, pervasive}, with subtle
  1101 change of semantics: update is applied to auxiliary local theory
  1102 context as well.
  1103 
  1104 * Modernized some old-style infix operations:
  1105 
  1106   addeqcongs    ~> Simplifier.add_eqcong
  1107   deleqcongs    ~> Simplifier.del_eqcong
  1108   addcongs      ~> Simplifier.add_cong
  1109   delcongs      ~> Simplifier.del_cong
  1110   setmksimps    ~> Simplifier.set_mksimps
  1111   setmkcong     ~> Simplifier.set_mkcong
  1112   setmksym      ~> Simplifier.set_mksym
  1113   setmkeqTrue   ~> Simplifier.set_mkeqTrue
  1114   settermless   ~> Simplifier.set_termless
  1115   setsubgoaler  ~> Simplifier.set_subgoaler
  1116   addsplits     ~> Splitter.add_split
  1117   delsplits     ~> Splitter.del_split
  1118 
  1119 
  1120 *** System ***
  1121 
  1122 * USER_HOME settings variable points to cross-platform user home
  1123 directory, which coincides with HOME on POSIX systems only.  Likewise,
  1124 the Isabelle path specification "~" now expands to $USER_HOME, instead
  1125 of former $HOME.  A different default for USER_HOME may be set
  1126 explicitly in shell environment, before Isabelle settings are
  1127 evaluated.  Minor INCOMPATIBILITY: need to adapt Isabelle path where
  1128 the generic user home was intended.
  1129 
  1130 * ISABELLE_HOME_WINDOWS refers to ISABELLE_HOME in windows file name
  1131 notation, which is useful for the jEdit file browser, for example.
  1132 
  1133 * ISABELLE_JDK_HOME settings variable points to JDK with javac and jar
  1134 (not just JRE).
  1135 
  1136 
  1137 
  1138 New in Isabelle2011-1 (October 2011)
  1139 ------------------------------------
  1140 
  1141 *** General ***
  1142 
  1143 * Improved Isabelle/jEdit Prover IDE (PIDE), which can be invoked as
  1144 "isabelle jedit" or "ISABELLE_HOME/Isabelle" on the command line.
  1145 
  1146   - Management of multiple theory files directly from the editor
  1147     buffer store -- bypassing the file-system (no requirement to save
  1148     files for checking).
  1149 
  1150   - Markup of formal entities within the text buffer, with semantic
  1151     highlighting, tooltips and hyperlinks to jump to defining source
  1152     positions.
  1153 
  1154   - Improved text rendering, with sub/superscripts in the source
  1155     buffer (including support for copy/paste wrt. output panel, HTML
  1156     theory output and other non-Isabelle text boxes).
  1157 
  1158   - Refined scheduling of proof checking and printing of results,
  1159     based on interactive editor view.  (Note: jEdit folding and
  1160     narrowing allows to restrict buffer perspectives explicitly.)
  1161 
  1162   - Reduced CPU performance requirements, usable on machines with few
  1163     cores.
  1164 
  1165   - Reduced memory requirements due to pruning of unused document
  1166     versions (garbage collection).
  1167 
  1168 See also ~~/src/Tools/jEdit/README.html for further information,
  1169 including some remaining limitations.
  1170 
  1171 * Theory loader: source files are exclusively located via the master
  1172 directory of each theory node (where the .thy file itself resides).
  1173 The global load path (such as src/HOL/Library) has been discontinued.
  1174 Note that the path element ~~ may be used to reference theories in the
  1175 Isabelle home folder -- for instance, "~~/src/HOL/Library/FuncSet".
  1176 INCOMPATIBILITY.
  1177 
  1178 * Theory loader: source files are identified by content via SHA1
  1179 digests.  Discontinued former path/modtime identification and optional
  1180 ISABELLE_FILE_IDENT plugin scripts.
  1181 
  1182 * Parallelization of nested Isar proofs is subject to
  1183 Goal.parallel_proofs_threshold (default 100).  See also isabelle
  1184 usedir option -Q.
  1185 
  1186 * Name space: former unsynchronized references are now proper
  1187 configuration options, with more conventional names:
  1188 
  1189   long_names   ~> names_long
  1190   short_names  ~> names_short
  1191   unique_names ~> names_unique
  1192 
  1193 Minor INCOMPATIBILITY, need to declare options in context like this:
  1194 
  1195   declare [[names_unique = false]]
  1196 
  1197 * Literal facts `prop` may contain dummy patterns, e.g. `_ = _`.  Note
  1198 that the result needs to be unique, which means fact specifications
  1199 may have to be refined after enriching a proof context.
  1200 
  1201 * Attribute "case_names" has been refined: the assumptions in each case
  1202 can be named now by following the case name with [name1 name2 ...].
  1203 
  1204 * Isabelle/Isar reference manual has been updated and extended:
  1205   - "Synopsis" provides a catalog of main Isar language concepts.
  1206   - Formal references in syntax diagrams, via @{rail} antiquotation.
  1207   - Updated material from classic "ref" manual, notably about
  1208     "Classical Reasoner".
  1209 
  1210 
  1211 *** HOL ***
  1212 
  1213 * Class bot and top require underlying partial order rather than
  1214 preorder: uniqueness of bot and top is guaranteed.  INCOMPATIBILITY.
  1215 
  1216 * Class complete_lattice: generalized a couple of lemmas from sets;
  1217 generalized theorems INF_cong and SUP_cong.  New type classes for
  1218 complete boolean algebras and complete linear orders.  Lemmas
  1219 Inf_less_iff, less_Sup_iff, INF_less_iff, less_SUP_iff now reside in
  1220 class complete_linorder.
  1221 
  1222 Changed proposition of lemmas Inf_bool_def, Sup_bool_def, Inf_fun_def,
  1223 Sup_fun_def, Inf_apply, Sup_apply.
  1224 
  1225 Removed redundant lemmas (the right hand side gives hints how to
  1226 replace them for (metis ...), or (simp only: ...) proofs):
  1227 
  1228   Inf_singleton ~> Inf_insert [where A="{}", unfolded Inf_empty inf_top_right]
  1229   Sup_singleton ~> Sup_insert [where A="{}", unfolded Sup_empty sup_bot_right]
  1230   Inf_binary ~> Inf_insert, Inf_empty, and inf_top_right
  1231   Sup_binary ~> Sup_insert, Sup_empty, and sup_bot_right
  1232   Int_eq_Inter ~> Inf_insert, Inf_empty, and inf_top_right
  1233   Un_eq_Union ~> Sup_insert, Sup_empty, and sup_bot_right
  1234   Inter_def ~> INF_def, image_def
  1235   Union_def ~> SUP_def, image_def
  1236   INT_eq ~> INF_def, and image_def
  1237   UN_eq ~> SUP_def, and image_def
  1238   INF_subset ~> INF_superset_mono [OF _ order_refl]
  1239 
  1240 More consistent and comprehensive names:
  1241 
  1242   INTER_eq_Inter_image ~> INF_def
  1243   UNION_eq_Union_image ~> SUP_def
  1244   INFI_def ~> INF_def
  1245   SUPR_def ~> SUP_def
  1246   INF_leI ~> INF_lower
  1247   INF_leI2 ~> INF_lower2
  1248   le_INFI ~> INF_greatest
  1249   le_SUPI ~> SUP_upper
  1250   le_SUPI2 ~> SUP_upper2
  1251   SUP_leI ~> SUP_least
  1252   INFI_bool_eq ~> INF_bool_eq
  1253   SUPR_bool_eq ~> SUP_bool_eq
  1254   INFI_apply ~> INF_apply
  1255   SUPR_apply ~> SUP_apply
  1256   INTER_def ~> INTER_eq
  1257   UNION_def ~> UNION_eq
  1258 
  1259 INCOMPATIBILITY.
  1260 
  1261 * Renamed theory Complete_Lattice to Complete_Lattices.
  1262 INCOMPATIBILITY.
  1263 
  1264 * Theory Complete_Lattices: lemmas Inf_eq_top_iff, INF_eq_top_iff,
  1265 INF_image, Inf_insert, INF_top, Inf_top_conv, INF_top_conv, SUP_bot,
  1266 Sup_bot_conv, SUP_bot_conv, Sup_eq_top_iff, SUP_eq_top_iff, SUP_image,
  1267 Sup_insert are now declared as [simp].  INCOMPATIBILITY.
  1268 
  1269 * Theory Lattice: lemmas compl_inf_bot, compl_le_comp_iff,
  1270 compl_sup_top, inf_idem, inf_left_idem, inf_sup_absorb, sup_idem,
  1271 sup_inf_absob, sup_left_idem are now declared as [simp].  Minor
  1272 INCOMPATIBILITY.
  1273 
  1274 * Added syntactic classes "inf" and "sup" for the respective
  1275 constants.  INCOMPATIBILITY: Changes in the argument order of the
  1276 (mostly internal) locale predicates for some derived classes.
  1277 
  1278 * Theorem collections ball_simps and bex_simps do not contain theorems
  1279 referring to UNION any longer; these have been moved to collection
  1280 UN_ball_bex_simps.  INCOMPATIBILITY.
  1281 
  1282 * Theory Archimedean_Field: floor now is defined as parameter of a
  1283 separate type class floor_ceiling.
  1284 
  1285 * Theory Finite_Set: more coherent development of fold_set locales:
  1286 
  1287     locale fun_left_comm ~> locale comp_fun_commute
  1288     locale fun_left_comm_idem ~> locale comp_fun_idem
  1289 
  1290 Both use point-free characterization; interpretation proofs may need
  1291 adjustment.  INCOMPATIBILITY.
  1292 
  1293 * Theory Limits: Type "'a net" has been renamed to "'a filter", in
  1294 accordance with standard mathematical terminology. INCOMPATIBILITY.
  1295 
  1296 * Theory Complex_Main: The locale interpretations for the
  1297 bounded_linear and bounded_bilinear locales have been removed, in
  1298 order to reduce the number of duplicate lemmas. Users must use the
  1299 original names for distributivity theorems, potential INCOMPATIBILITY.
  1300 
  1301   divide.add ~> add_divide_distrib
  1302   divide.diff ~> diff_divide_distrib
  1303   divide.setsum ~> setsum_divide_distrib
  1304   mult.add_right ~> right_distrib
  1305   mult.diff_right ~> right_diff_distrib
  1306   mult_right.setsum ~> setsum_right_distrib
  1307   mult_left.diff ~> left_diff_distrib
  1308 
  1309 * Theory Complex_Main: Several redundant theorems have been removed or
  1310 replaced by more general versions. INCOMPATIBILITY.
  1311 
  1312   real_diff_def ~> minus_real_def
  1313   real_divide_def ~> divide_real_def
  1314   real_less_def ~> less_le
  1315   real_abs_def ~> abs_real_def
  1316   real_sgn_def ~> sgn_real_def
  1317   real_mult_commute ~> mult_commute
  1318   real_mult_assoc ~> mult_assoc
  1319   real_mult_1 ~> mult_1_left
  1320   real_add_mult_distrib ~> left_distrib
  1321   real_zero_not_eq_one ~> zero_neq_one
  1322   real_mult_inverse_left ~> left_inverse
  1323   INVERSE_ZERO ~> inverse_zero
  1324   real_le_refl ~> order_refl
  1325   real_le_antisym ~> order_antisym
  1326   real_le_trans ~> order_trans
  1327   real_le_linear ~> linear
  1328   real_le_eq_diff ~> le_iff_diff_le_0
  1329   real_add_left_mono ~> add_left_mono
  1330   real_mult_order ~> mult_pos_pos
  1331   real_mult_less_mono2 ~> mult_strict_left_mono
  1332   real_of_int_real_of_nat ~> real_of_int_of_nat_eq
  1333   real_0_le_divide_iff ~> zero_le_divide_iff
  1334   realpow_two_disj ~> power2_eq_iff
  1335   real_squared_diff_one_factored ~> square_diff_one_factored
  1336   realpow_two_diff ~> square_diff_square_factored
  1337   reals_complete2 ~> complete_real
  1338   real_sum_squared_expand ~> power2_sum
  1339   exp_ln_eq ~> ln_unique
  1340   expi_add ~> exp_add
  1341   expi_zero ~> exp_zero
  1342   lemma_DERIV_subst ~> DERIV_cong
  1343   LIMSEQ_Zfun_iff ~> tendsto_Zfun_iff
  1344   LIMSEQ_const ~> tendsto_const
  1345   LIMSEQ_norm ~> tendsto_norm
  1346   LIMSEQ_add ~> tendsto_add
  1347   LIMSEQ_minus ~> tendsto_minus
  1348   LIMSEQ_minus_cancel ~> tendsto_minus_cancel
  1349   LIMSEQ_diff ~> tendsto_diff
  1350   bounded_linear.LIMSEQ ~> bounded_linear.tendsto
  1351   bounded_bilinear.LIMSEQ ~> bounded_bilinear.tendsto
  1352   LIMSEQ_mult ~> tendsto_mult
  1353   LIMSEQ_inverse ~> tendsto_inverse
  1354   LIMSEQ_divide ~> tendsto_divide
  1355   LIMSEQ_pow ~> tendsto_power
  1356   LIMSEQ_setsum ~> tendsto_setsum
  1357   LIMSEQ_setprod ~> tendsto_setprod
  1358   LIMSEQ_norm_zero ~> tendsto_norm_zero_iff
  1359   LIMSEQ_rabs_zero ~> tendsto_rabs_zero_iff
  1360   LIMSEQ_imp_rabs ~> tendsto_rabs
  1361   LIMSEQ_add_minus ~> tendsto_add [OF _ tendsto_minus]
  1362   LIMSEQ_add_const ~> tendsto_add [OF _ tendsto_const]
  1363   LIMSEQ_diff_const ~> tendsto_diff [OF _ tendsto_const]
  1364   LIMSEQ_Complex ~> tendsto_Complex
  1365   LIM_ident ~> tendsto_ident_at
  1366   LIM_const ~> tendsto_const
  1367   LIM_add ~> tendsto_add
  1368   LIM_add_zero ~> tendsto_add_zero
  1369   LIM_minus ~> tendsto_minus
  1370   LIM_diff ~> tendsto_diff
  1371   LIM_norm ~> tendsto_norm
  1372   LIM_norm_zero ~> tendsto_norm_zero
  1373   LIM_norm_zero_cancel ~> tendsto_norm_zero_cancel
  1374   LIM_norm_zero_iff ~> tendsto_norm_zero_iff
  1375   LIM_rabs ~> tendsto_rabs
  1376   LIM_rabs_zero ~> tendsto_rabs_zero
  1377   LIM_rabs_zero_cancel ~> tendsto_rabs_zero_cancel
  1378   LIM_rabs_zero_iff ~> tendsto_rabs_zero_iff
  1379   LIM_compose ~> tendsto_compose
  1380   LIM_mult ~> tendsto_mult
  1381   LIM_scaleR ~> tendsto_scaleR
  1382   LIM_of_real ~> tendsto_of_real
  1383   LIM_power ~> tendsto_power
  1384   LIM_inverse ~> tendsto_inverse
  1385   LIM_sgn ~> tendsto_sgn
  1386   isCont_LIM_compose ~> isCont_tendsto_compose
  1387   bounded_linear.LIM ~> bounded_linear.tendsto
  1388   bounded_linear.LIM_zero ~> bounded_linear.tendsto_zero
  1389   bounded_bilinear.LIM ~> bounded_bilinear.tendsto
  1390   bounded_bilinear.LIM_prod_zero ~> bounded_bilinear.tendsto_zero
  1391   bounded_bilinear.LIM_left_zero ~> bounded_bilinear.tendsto_left_zero
  1392   bounded_bilinear.LIM_right_zero ~> bounded_bilinear.tendsto_right_zero
  1393   LIM_inverse_fun ~> tendsto_inverse [OF tendsto_ident_at]
  1394 
  1395 * Theory Complex_Main: The definition of infinite series was
  1396 generalized.  Now it is defined on the type class {topological_space,
  1397 comm_monoid_add}.  Hence it is useable also for extended real numbers.
  1398 
  1399 * Theory Complex_Main: The complex exponential function "expi" is now
  1400 a type-constrained abbreviation for "exp :: complex => complex"; thus
  1401 several polymorphic lemmas about "exp" are now applicable to "expi".
  1402 
  1403 * Code generation:
  1404 
  1405   - Theory Library/Code_Char_ord provides native ordering of
  1406     characters in the target language.
  1407 
  1408   - Commands code_module and code_library are legacy, use export_code
  1409     instead.
  1410 
  1411   - Method "evaluation" is legacy, use method "eval" instead.
  1412 
  1413   - Legacy evaluator "SML" is deactivated by default.  May be
  1414     reactivated by the following theory command:
  1415 
  1416       setup {* Value.add_evaluator ("SML", Codegen.eval_term) *}
  1417 
  1418 * Declare ext [intro] by default.  Rare INCOMPATIBILITY.
  1419 
  1420 * New proof method "induction" that gives induction hypotheses the
  1421 name "IH", thus distinguishing them from further hypotheses that come
  1422 from rule induction.  The latter are still called "hyps".  Method
  1423 "induction" is a thin wrapper around "induct" and follows the same
  1424 syntax.
  1425 
  1426 * Method "fastsimp" has been renamed to "fastforce", but "fastsimp" is
  1427 still available as a legacy feature for some time.
  1428 
  1429 * Nitpick:
  1430   - Added "need" and "total_consts" options.
  1431   - Reintroduced "show_skolems" option by popular demand.
  1432   - Renamed attribute: nitpick_def ~> nitpick_unfold.
  1433     INCOMPATIBILITY.
  1434 
  1435 * Sledgehammer:
  1436   - Use quasi-sound (and efficient) translations by default.
  1437   - Added support for the following provers: E-ToFoF, LEO-II,
  1438     Satallax, SNARK, Waldmeister, and Z3 with TPTP syntax.
  1439   - Automatically preplay and minimize proofs before showing them if
  1440     this can be done within reasonable time.
  1441   - sledgehammer available_provers ~> sledgehammer supported_provers.
  1442     INCOMPATIBILITY.
  1443   - Added "preplay_timeout", "slicing", "type_enc", "sound",
  1444     "max_mono_iters", and "max_new_mono_instances" options.
  1445   - Removed "explicit_apply" and "full_types" options as well as "Full
  1446     Types" Proof General menu item. INCOMPATIBILITY.
  1447 
  1448 * Metis:
  1449   - Removed "metisF" -- use "metis" instead. INCOMPATIBILITY.
  1450   - Obsoleted "metisFT" -- use "metis (full_types)" instead.
  1451     INCOMPATIBILITY.
  1452 
  1453 * Command 'try':
  1454   - Renamed 'try_methods' and added "simp:", "intro:", "dest:", and
  1455     "elim:" options. INCOMPATIBILITY.
  1456   - Introduced 'try' that not only runs 'try_methods' but also
  1457     'solve_direct', 'sledgehammer', 'quickcheck', and 'nitpick'.
  1458 
  1459 * Quickcheck:
  1460   - Added "eval" option to evaluate terms for the found counterexample
  1461     (currently only supported by the default (exhaustive) tester).
  1462   - Added post-processing of terms to obtain readable counterexamples
  1463     (currently only supported by the default (exhaustive) tester).
  1464   - New counterexample generator quickcheck[narrowing] enables
  1465     narrowing-based testing.  Requires the Glasgow Haskell compiler
  1466     with its installation location defined in the Isabelle settings
  1467     environment as ISABELLE_GHC.
  1468   - Removed quickcheck tester "SML" based on the SML code generator
  1469     (formly in HOL/Library).
  1470 
  1471 * Function package: discontinued option "tailrec".  INCOMPATIBILITY,
  1472 use 'partial_function' instead.
  1473 
  1474 * Theory Library/Extended_Reals replaces now the positive extended
  1475 reals found in probability theory. This file is extended by
  1476 Multivariate_Analysis/Extended_Real_Limits.
  1477 
  1478 * Theory Library/Old_Recdef: old 'recdef' package has been moved here,
  1479 from where it must be imported explicitly if it is really required.
  1480 INCOMPATIBILITY.
  1481 
  1482 * Theory Library/Wfrec: well-founded recursion combinator "wfrec" has
  1483 been moved here.  INCOMPATIBILITY.
  1484 
  1485 * Theory Library/Saturated provides type of numbers with saturated
  1486 arithmetic.
  1487 
  1488 * Theory Library/Product_Lattice defines a pointwise ordering for the
  1489 product type 'a * 'b, and provides instance proofs for various order
  1490 and lattice type classes.
  1491 
  1492 * Theory Library/Countable now provides the "countable_datatype" proof
  1493 method for proving "countable" class instances for datatypes.
  1494 
  1495 * Theory Library/Cset_Monad allows do notation for computable sets
  1496 (cset) via the generic monad ad-hoc overloading facility.
  1497 
  1498 * Library: Theories of common data structures are split into theories
  1499 for implementation, an invariant-ensuring type, and connection to an
  1500 abstract type. INCOMPATIBILITY.
  1501 
  1502   - RBT is split into RBT and RBT_Mapping.
  1503   - AssocList is split and renamed into AList and AList_Mapping.
  1504   - DList is split into DList_Impl, DList, and DList_Cset.
  1505   - Cset is split into Cset and List_Cset.
  1506 
  1507 * Theory Library/Nat_Infinity has been renamed to
  1508 Library/Extended_Nat, with name changes of the following types and
  1509 constants:
  1510 
  1511   type inat   ~> type enat
  1512   Fin         ~> enat
  1513   Infty       ~> infinity (overloaded)
  1514   iSuc        ~> eSuc
  1515   the_Fin     ~> the_enat
  1516 
  1517 Every theorem name containing "inat", "Fin", "Infty", or "iSuc" has
  1518 been renamed accordingly. INCOMPATIBILITY.
  1519 
  1520 * Session Multivariate_Analysis: The euclidean_space type class now
  1521 fixes a constant "Basis :: 'a set" consisting of the standard
  1522 orthonormal basis for the type. Users now have the option of
  1523 quantifying over this set instead of using the "basis" function, e.g.
  1524 "ALL x:Basis. P x" vs "ALL i<DIM('a). P (basis i)".
  1525 
  1526 * Session Multivariate_Analysis: Type "('a, 'b) cart" has been renamed
  1527 to "('a, 'b) vec" (the syntax "'a ^ 'b" remains unaffected). Constants
  1528 "Cart_nth" and "Cart_lambda" have been respectively renamed to
  1529 "vec_nth" and "vec_lambda"; theorems mentioning those names have
  1530 changed to match. Definition theorems for overloaded constants now use
  1531 the standard "foo_vec_def" naming scheme. A few other theorems have
  1532 been renamed as follows (INCOMPATIBILITY):
  1533 
  1534   Cart_eq          ~> vec_eq_iff
  1535   dist_nth_le_cart ~> dist_vec_nth_le
  1536   tendsto_vector   ~> vec_tendstoI
  1537   Cauchy_vector    ~> vec_CauchyI
  1538 
  1539 * Session Multivariate_Analysis: Several duplicate theorems have been
  1540 removed, and other theorems have been renamed or replaced with more
  1541 general versions. INCOMPATIBILITY.
  1542 
  1543   finite_choice ~> finite_set_choice
  1544   eventually_conjI ~> eventually_conj
  1545   eventually_and ~> eventually_conj_iff
  1546   eventually_false ~> eventually_False
  1547   setsum_norm ~> norm_setsum
  1548   Lim_sequentially ~> LIMSEQ_def
  1549   Lim_ident_at ~> LIM_ident
  1550   Lim_const ~> tendsto_const
  1551   Lim_cmul ~> tendsto_scaleR [OF tendsto_const]
  1552   Lim_neg ~> tendsto_minus
  1553   Lim_add ~> tendsto_add
  1554   Lim_sub ~> tendsto_diff
  1555   Lim_mul ~> tendsto_scaleR
  1556   Lim_vmul ~> tendsto_scaleR [OF _ tendsto_const]
  1557   Lim_null_norm ~> tendsto_norm_zero_iff [symmetric]
  1558   Lim_linear ~> bounded_linear.tendsto
  1559   Lim_component ~> tendsto_euclidean_component
  1560   Lim_component_cart ~> tendsto_vec_nth
  1561   Lim_inner ~> tendsto_inner [OF tendsto_const]
  1562   dot_lsum ~> inner_setsum_left
  1563   dot_rsum ~> inner_setsum_right
  1564   continuous_cmul ~> continuous_scaleR [OF continuous_const]
  1565   continuous_neg ~> continuous_minus
  1566   continuous_sub ~> continuous_diff
  1567   continuous_vmul ~> continuous_scaleR [OF _ continuous_const]
  1568   continuous_mul ~> continuous_scaleR
  1569   continuous_inv ~> continuous_inverse
  1570   continuous_at_within_inv ~> continuous_at_within_inverse
  1571   continuous_at_inv ~> continuous_at_inverse
  1572   continuous_at_norm ~> continuous_norm [OF continuous_at_id]
  1573   continuous_at_infnorm ~> continuous_infnorm [OF continuous_at_id]
  1574   continuous_at_component ~> continuous_component [OF continuous_at_id]
  1575   continuous_on_neg ~> continuous_on_minus
  1576   continuous_on_sub ~> continuous_on_diff
  1577   continuous_on_cmul ~> continuous_on_scaleR [OF continuous_on_const]
  1578   continuous_on_vmul ~> continuous_on_scaleR [OF _ continuous_on_const]
  1579   continuous_on_mul ~> continuous_on_scaleR
  1580   continuous_on_mul_real ~> continuous_on_mult
  1581   continuous_on_inner ~> continuous_on_inner [OF continuous_on_const]
  1582   continuous_on_norm ~> continuous_on_norm [OF continuous_on_id]
  1583   continuous_on_inverse ~> continuous_on_inv
  1584   uniformly_continuous_on_neg ~> uniformly_continuous_on_minus
  1585   uniformly_continuous_on_sub ~> uniformly_continuous_on_diff
  1586   subset_interior ~> interior_mono
  1587   subset_closure ~> closure_mono
  1588   closure_univ ~> closure_UNIV
  1589   real_arch_lt ~> reals_Archimedean2
  1590   real_arch ~> reals_Archimedean3
  1591   real_abs_norm ~> abs_norm_cancel
  1592   real_abs_sub_norm ~> norm_triangle_ineq3
  1593   norm_cauchy_schwarz_abs ~> Cauchy_Schwarz_ineq2
  1594 
  1595 * Session HOL-Probability:
  1596   - Caratheodory's extension lemma is now proved for ring_of_sets.
  1597   - Infinite products of probability measures are now available.
  1598   - Sigma closure is independent, if the generator is independent
  1599   - Use extended reals instead of positive extended
  1600     reals. INCOMPATIBILITY.
  1601 
  1602 * Session HOLCF: Discontinued legacy theorem names, INCOMPATIBILITY.
  1603 
  1604   expand_fun_below ~> fun_below_iff
  1605   below_fun_ext ~> fun_belowI
  1606   expand_cfun_eq ~> cfun_eq_iff
  1607   ext_cfun ~> cfun_eqI
  1608   expand_cfun_below ~> cfun_below_iff
  1609   below_cfun_ext ~> cfun_belowI
  1610   monofun_fun_fun ~> fun_belowD
  1611   monofun_fun_arg ~> monofunE
  1612   monofun_lub_fun ~> adm_monofun [THEN admD]
  1613   cont_lub_fun ~> adm_cont [THEN admD]
  1614   cont2cont_Rep_CFun ~> cont2cont_APP
  1615   cont_Rep_CFun_app ~> cont_APP_app
  1616   cont_Rep_CFun_app_app ~> cont_APP_app_app
  1617   cont_cfun_fun ~> cont_Rep_cfun1 [THEN contE]
  1618   cont_cfun_arg ~> cont_Rep_cfun2 [THEN contE]
  1619   contlub_cfun ~> lub_APP [symmetric]
  1620   contlub_LAM ~> lub_LAM [symmetric]
  1621   thelubI ~> lub_eqI
  1622   UU_I ~> bottomI
  1623   lift_distinct1 ~> lift.distinct(1)
  1624   lift_distinct2 ~> lift.distinct(2)
  1625   Def_not_UU ~> lift.distinct(2)
  1626   Def_inject ~> lift.inject
  1627   below_UU_iff ~> below_bottom_iff
  1628   eq_UU_iff ~> eq_bottom_iff
  1629 
  1630 
  1631 *** Document preparation ***
  1632 
  1633 * Antiquotation @{rail} layouts railroad syntax diagrams, see also
  1634 isar-ref manual, both for description and actual application of the
  1635 same.
  1636 
  1637 * Antiquotation @{value} evaluates the given term and presents its
  1638 result.
  1639 
  1640 * Antiquotations: term style "isub" provides ad-hoc conversion of
  1641 variables x1, y23 into subscripted form x\<^isub>1,
  1642 y\<^isub>2\<^isub>3.
  1643 
  1644 * Predefined LaTeX macros for Isabelle symbols \<bind> and \<then>
  1645 (e.g. see ~~/src/HOL/Library/Monad_Syntax.thy).
  1646 
  1647 * Localized \isabellestyle switch can be used within blocks or groups
  1648 like this:
  1649 
  1650   \isabellestyle{it}  %preferred default
  1651   {\isabellestylett @{text "typewriter stuff"}}
  1652 
  1653 * Discontinued special treatment of hard tabulators.  Implicit
  1654 tab-width is now defined as 1.  Potential INCOMPATIBILITY for visual
  1655 layouts.
  1656 
  1657 
  1658 *** ML ***
  1659 
  1660 * The inner syntax of sort/type/term/prop supports inlined YXML
  1661 representations within quoted string tokens.  By encoding logical
  1662 entities via Term_XML (in ML or Scala) concrete syntax can be
  1663 bypassed, which is particularly useful for producing bits of text
  1664 under external program control.
  1665 
  1666 * Antiquotations for ML and document preparation are managed as theory
  1667 data, which requires explicit setup.
  1668 
  1669 * Isabelle_Process.is_active allows tools to check if the official
  1670 process wrapper is running (Isabelle/Scala/jEdit) or the old TTY loop
  1671 (better known as Proof General).
  1672 
  1673 * Structure Proof_Context follows standard naming scheme.  Old
  1674 ProofContext is still available for some time as legacy alias.
  1675 
  1676 * Structure Timing provides various operations for timing; supersedes
  1677 former start_timing/end_timing etc.
  1678 
  1679 * Path.print is the official way to show file-system paths to users
  1680 (including quotes etc.).
  1681 
  1682 * Inner syntax: identifiers in parse trees of generic categories
  1683 "logic", "aprop", "idt" etc. carry position information (disguised as
  1684 type constraints).  Occasional INCOMPATIBILITY with non-compliant
  1685 translations that choke on unexpected type constraints.  Positions can
  1686 be stripped in ML translations via Syntax.strip_positions /
  1687 Syntax.strip_positions_ast, or via the syntax constant
  1688 "_strip_positions" within parse trees.  As last resort, positions can
  1689 be disabled via the configuration option Syntax.positions, which is
  1690 called "syntax_positions" in Isar attribute syntax.
  1691 
  1692 * Discontinued special status of various ML structures that contribute
  1693 to structure Syntax (Ast, Lexicon, Mixfix, Parser, Printer etc.): less
  1694 pervasive content, no inclusion in structure Syntax.  INCOMPATIBILITY,
  1695 refer directly to Ast.Constant, Lexicon.is_identifier,
  1696 Syntax_Trans.mk_binder_tr etc.
  1697 
  1698 * Typed print translation: discontinued show_sorts argument, which is
  1699 already available via context of "advanced" translation.
  1700 
  1701 * Refined PARALLEL_GOALS tactical: degrades gracefully for schematic
  1702 goal states; body tactic needs to address all subgoals uniformly.
  1703 
  1704 * Slightly more special eq_list/eq_set, with shortcut involving
  1705 pointer equality (assumes that eq relation is reflexive).
  1706 
  1707 * Classical tactics use proper Proof.context instead of historic types
  1708 claset/clasimpset.  Old-style declarations like addIs, addEs, addDs
  1709 operate directly on Proof.context.  Raw type claset retains its use as
  1710 snapshot of the classical context, which can be recovered via
  1711 (put_claset HOL_cs) etc.  Type clasimpset has been discontinued.
  1712 INCOMPATIBILITY, classical tactics and derived proof methods require
  1713 proper Proof.context.
  1714 
  1715 
  1716 *** System ***
  1717 
  1718 * Discontinued support for Poly/ML 5.2, which was the last version
  1719 without proper multithreading and TimeLimit implementation.
  1720 
  1721 * Discontinued old lib/scripts/polyml-platform, which has been
  1722 obsolete since Isabelle2009-2.
  1723 
  1724 * Various optional external tools are referenced more robustly and
  1725 uniformly by explicit Isabelle settings as follows:
  1726 
  1727   ISABELLE_CSDP   (formerly CSDP_EXE)
  1728   ISABELLE_GHC    (formerly EXEC_GHC or GHC_PATH)
  1729   ISABELLE_OCAML  (formerly EXEC_OCAML)
  1730   ISABELLE_SWIPL  (formerly EXEC_SWIPL)
  1731   ISABELLE_YAP    (formerly EXEC_YAP)
  1732 
  1733 Note that automated detection from the file-system or search path has
  1734 been discontinued.  INCOMPATIBILITY.
  1735 
  1736 * Scala layer provides JVM method invocation service for static
  1737 methods of type (String)String, see Invoke_Scala.method in ML.  For
  1738 example:
  1739 
  1740   Invoke_Scala.method "java.lang.System.getProperty" "java.home"
  1741 
  1742 Together with YXML.string_of_body/parse_body and XML.Encode/Decode
  1743 this allows to pass structured values between ML and Scala.
  1744 
  1745 * The IsabelleText fonts includes some further glyphs to support the
  1746 Prover IDE.  Potential INCOMPATIBILITY: users who happen to have
  1747 installed a local copy (which is normally *not* required) need to
  1748 delete or update it from ~~/lib/fonts/.
  1749 
  1750 
  1751 
  1752 New in Isabelle2011 (January 2011)
  1753 ----------------------------------
  1754 
  1755 *** General ***
  1756 
  1757 * Experimental Prover IDE based on Isabelle/Scala and jEdit (see
  1758 src/Tools/jEdit).  This also serves as IDE for Isabelle/ML, with
  1759 useful tooltips and hyperlinks produced from its static analysis.  The
  1760 bundled component provides an executable Isabelle tool that can be run
  1761 like this:
  1762 
  1763   Isabelle2011/bin/isabelle jedit
  1764 
  1765 * Significantly improved Isabelle/Isar implementation manual.
  1766 
  1767 * System settings: ISABELLE_HOME_USER now includes ISABELLE_IDENTIFIER
  1768 (and thus refers to something like $HOME/.isabelle/Isabelle2011),
  1769 while the default heap location within that directory lacks that extra
  1770 suffix.  This isolates multiple Isabelle installations from each
  1771 other, avoiding problems with old settings in new versions.
  1772 INCOMPATIBILITY, need to copy/upgrade old user settings manually.
  1773 
  1774 * Source files are always encoded as UTF-8, instead of old-fashioned
  1775 ISO-Latin-1.  INCOMPATIBILITY.  Isabelle LaTeX documents might require
  1776 the following package declarations:
  1777 
  1778   \usepackage[utf8]{inputenc}
  1779   \usepackage{textcomp}
  1780 
  1781 * Explicit treatment of UTF-8 sequences as Isabelle symbols, such that
  1782 a Unicode character is treated as a single symbol, not a sequence of
  1783 non-ASCII bytes as before.  Since Isabelle/ML string literals may
  1784 contain symbols without further backslash escapes, Unicode can now be
  1785 used here as well.  Recall that Symbol.explode in ML provides a
  1786 consistent view on symbols, while raw explode (or String.explode)
  1787 merely give a byte-oriented representation.
  1788 
  1789 * Theory loader: source files are primarily located via the master
  1790 directory of each theory node (where the .thy file itself resides).
  1791 The global load path is still partially available as legacy feature.
  1792 Minor INCOMPATIBILITY due to subtle change in file lookup: use
  1793 explicit paths, relatively to the theory.
  1794 
  1795 * Special treatment of ML file names has been discontinued.
  1796 Historically, optional extensions .ML or .sml were added on demand --
  1797 at the cost of clarity of file dependencies.  Recall that Isabelle/ML
  1798 files exclusively use the .ML extension.  Minor INCOMPATIBILTY.
  1799 
  1800 * Various options that affect pretty printing etc. are now properly
  1801 handled within the context via configuration options, instead of
  1802 unsynchronized references or print modes.  There are both ML Config.T
  1803 entities and Isar declaration attributes to access these.
  1804 
  1805   ML (Config.T)                 Isar (attribute)
  1806 
  1807   eta_contract                  eta_contract
  1808   show_brackets                 show_brackets
  1809   show_sorts                    show_sorts
  1810   show_types                    show_types
  1811   show_question_marks           show_question_marks
  1812   show_consts                   show_consts
  1813   show_abbrevs                  show_abbrevs
  1814 
  1815   Syntax.ast_trace              syntax_ast_trace
  1816   Syntax.ast_stat               syntax_ast_stat
  1817   Syntax.ambiguity_level        syntax_ambiguity_level
  1818 
  1819   Goal_Display.goals_limit      goals_limit
  1820   Goal_Display.show_main_goal   show_main_goal
  1821 
  1822   Method.rule_trace             rule_trace
  1823 
  1824   Thy_Output.display            thy_output_display
  1825   Thy_Output.quotes             thy_output_quotes
  1826   Thy_Output.indent             thy_output_indent
  1827   Thy_Output.source             thy_output_source
  1828   Thy_Output.break              thy_output_break
  1829 
  1830 Note that corresponding "..._default" references in ML may only be
  1831 changed globally at the ROOT session setup, but *not* within a theory.
  1832 The option "show_abbrevs" supersedes the former print mode
  1833 "no_abbrevs" with inverted meaning.
  1834 
  1835 * More systematic naming of some configuration options.
  1836 INCOMPATIBILITY.
  1837 
  1838   trace_simp  ~>  simp_trace
  1839   debug_simp  ~>  simp_debug
  1840 
  1841 * Support for real valued configuration options, using simplistic
  1842 floating-point notation that coincides with the inner syntax for
  1843 float_token.
  1844 
  1845 * Support for real valued preferences (with approximative PGIP type):
  1846 front-ends need to accept "pgint" values in float notation.
  1847 INCOMPATIBILITY.
  1848 
  1849 * The IsabelleText font now includes Cyrillic, Hebrew, Arabic from
  1850 DejaVu Sans.
  1851 
  1852 * Discontinued support for Poly/ML 5.0 and 5.1 versions.
  1853 
  1854 
  1855 *** Pure ***
  1856 
  1857 * Command 'type_synonym' (with single argument) replaces somewhat
  1858 outdated 'types', which is still available as legacy feature for some
  1859 time.
  1860 
  1861 * Command 'nonterminal' (with 'and' separated list of arguments)
  1862 replaces somewhat outdated 'nonterminals'.  INCOMPATIBILITY.
  1863 
  1864 * Command 'notepad' replaces former 'example_proof' for
  1865 experimentation in Isar without any result.  INCOMPATIBILITY.
  1866 
  1867 * Locale interpretation commands 'interpret' and 'sublocale' accept
  1868 lists of equations to map definitions in a locale to appropriate
  1869 entities in the context of the interpretation.  The 'interpretation'
  1870 command already provided this functionality.
  1871 
  1872 * Diagnostic command 'print_dependencies' prints the locale instances
  1873 that would be activated if the specified expression was interpreted in
  1874 the current context.  Variant "print_dependencies!" assumes a context
  1875 without interpretations.
  1876 
  1877 * Diagnostic command 'print_interps' prints interpretations in proofs
  1878 in addition to interpretations in theories.
  1879 
  1880 * Discontinued obsolete 'global' and 'local' commands to manipulate
  1881 the theory name space.  Rare INCOMPATIBILITY.  The ML functions
  1882 Sign.root_path and Sign.local_path may be applied directly where this
  1883 feature is still required for historical reasons.
  1884 
  1885 * Discontinued obsolete 'constdefs' command.  INCOMPATIBILITY, use
  1886 'definition' instead.
  1887 
  1888 * The "prems" fact, which refers to the accidental collection of
  1889 foundational premises in the context, is now explicitly marked as
  1890 legacy feature and will be discontinued soon.  Consider using "assms"
  1891 of the head statement or reference facts by explicit names.
  1892 
  1893 * Document antiquotations @{class} and @{type} print classes and type
  1894 constructors.
  1895 
  1896 * Document antiquotation @{file} checks file/directory entries within
  1897 the local file system.
  1898 
  1899 
  1900 *** HOL ***
  1901 
  1902 * Coercive subtyping: functions can be declared as coercions and type
  1903 inference will add them as necessary upon input of a term.  Theory
  1904 Complex_Main declares real :: nat => real and real :: int => real as
  1905 coercions. A coercion function f is declared like this:
  1906 
  1907   declare [[coercion f]]
  1908 
  1909 To lift coercions through type constructors (e.g. from nat => real to
  1910 nat list => real list), map functions can be declared, e.g.
  1911 
  1912   declare [[coercion_map map]]
  1913 
  1914 Currently coercion inference is activated only in theories including
  1915 real numbers, i.e. descendants of Complex_Main.  This is controlled by
  1916 the configuration option "coercion_enabled", e.g. it can be enabled in
  1917 other theories like this:
  1918 
  1919   declare [[coercion_enabled]]
  1920 
  1921 * Command 'partial_function' provides basic support for recursive
  1922 function definitions over complete partial orders.  Concrete instances
  1923 are provided for i) the option type, ii) tail recursion on arbitrary
  1924 types, and iii) the heap monad of Imperative_HOL.  See
  1925 src/HOL/ex/Fundefs.thy and src/HOL/Imperative_HOL/ex/Linked_Lists.thy
  1926 for examples.
  1927 
  1928 * Function package: f.psimps rules are no longer implicitly declared
  1929 as [simp].  INCOMPATIBILITY.
  1930 
  1931 * Datatype package: theorems generated for executable equality (class
  1932 "eq") carry proper names and are treated as default code equations.
  1933 
  1934 * Inductive package: now offers command 'inductive_simps' to
  1935 automatically derive instantiated and simplified equations for
  1936 inductive predicates, similar to 'inductive_cases'.
  1937 
  1938 * Command 'enriched_type' allows to register properties of the
  1939 functorial structure of types.
  1940 
  1941 * Improved infrastructure for term evaluation using code generator
  1942 techniques, in particular static evaluation conversions.
  1943 
  1944 * Code generator: Scala (2.8 or higher) has been added to the target
  1945 languages.
  1946 
  1947 * Code generator: globbing constant expressions "*" and "Theory.*"
  1948 have been replaced by the more idiomatic "_" and "Theory._".
  1949 INCOMPATIBILITY.
  1950 
  1951 * Code generator: export_code without explicit file declaration prints
  1952 to standard output.  INCOMPATIBILITY.
  1953 
  1954 * Code generator: do not print function definitions for case
  1955 combinators any longer.
  1956 
  1957 * Code generator: simplification with rules determined with
  1958 src/Tools/Code/code_simp.ML and method "code_simp".
  1959 
  1960 * Code generator for records: more idiomatic representation of record
  1961 types.  Warning: records are not covered by ancient SML code
  1962 generation any longer.  INCOMPATIBILITY.  In cases of need, a suitable
  1963 rep_datatype declaration helps to succeed then:
  1964 
  1965   record 'a foo = ...
  1966   ...
  1967   rep_datatype foo_ext ...
  1968 
  1969 * Records: logical foundation type for records does not carry a
  1970 '_type' suffix any longer (obsolete due to authentic syntax).
  1971 INCOMPATIBILITY.
  1972 
  1973 * Quickcheck now by default uses exhaustive testing instead of random
  1974 testing.  Random testing can be invoked by "quickcheck [random]",
  1975 exhaustive testing by "quickcheck [exhaustive]".
  1976 
  1977 * Quickcheck instantiates polymorphic types with small finite
  1978 datatypes by default. This enables a simple execution mechanism to
  1979 handle quantifiers and function equality over the finite datatypes.
  1980 
  1981 * Quickcheck random generator has been renamed from "code" to
  1982 "random".  INCOMPATIBILITY.
  1983 
  1984 * Quickcheck now has a configurable time limit which is set to 30
  1985 seconds by default. This can be changed by adding [timeout = n] to the
  1986 quickcheck command. The time limit for Auto Quickcheck is still set
  1987 independently.
  1988 
  1989 * Quickcheck in locales considers interpretations of that locale for
  1990 counter example search.
  1991 
  1992 * Sledgehammer:
  1993   - Added "smt" and "remote_smt" provers based on the "smt" proof
  1994     method. See the Sledgehammer manual for details ("isabelle doc
  1995     sledgehammer").
  1996   - Renamed commands:
  1997     sledgehammer atp_info ~> sledgehammer running_provers
  1998     sledgehammer atp_kill ~> sledgehammer kill_provers
  1999     sledgehammer available_atps ~> sledgehammer available_provers
  2000     INCOMPATIBILITY.
  2001   - Renamed options:
  2002     sledgehammer [atps = ...] ~> sledgehammer [provers = ...]
  2003     sledgehammer [atp = ...] ~> sledgehammer [prover = ...]
  2004     sledgehammer [timeout = 77 s] ~> sledgehammer [timeout = 77]
  2005     (and "ms" and "min" are no longer supported)
  2006     INCOMPATIBILITY.
  2007 
  2008 * Nitpick:
  2009   - Renamed options:
  2010     nitpick [timeout = 77 s] ~> nitpick [timeout = 77]
  2011     nitpick [tac_timeout = 777 ms] ~> nitpick [tac_timeout = 0.777]
  2012     INCOMPATIBILITY.
  2013   - Added support for partial quotient types.
  2014   - Added local versions of the "Nitpick.register_xxx" functions.
  2015   - Added "whack" option.
  2016   - Allow registration of quotient types as codatatypes.
  2017   - Improved "merge_type_vars" option to merge more types.
  2018   - Removed unsound "fast_descrs" option.
  2019   - Added custom symmetry breaking for datatypes, making it possible to reach
  2020     higher cardinalities.
  2021   - Prevent the expansion of too large definitions.
  2022 
  2023 * Proof methods "metis" and "meson" now have configuration options
  2024 "meson_trace", "metis_trace", and "metis_verbose" that can be enabled
  2025 to diagnose these tools. E.g.
  2026 
  2027     using [[metis_trace = true]]
  2028 
  2029 * Auto Solve: Renamed "Auto Solve Direct".  The tool is now available
  2030 manually as command 'solve_direct'.
  2031 
  2032 * The default SMT solver Z3 must be enabled explicitly (due to
  2033 licensing issues) by setting the environment variable
  2034 Z3_NON_COMMERCIAL in etc/settings of the component, for example.  For
  2035 commercial applications, the SMT solver CVC3 is provided as fall-back;
  2036 changing the SMT solver is done via the configuration option
  2037 "smt_solver".
  2038 
  2039 * Remote SMT solvers need to be referred to by the "remote_" prefix,
  2040 i.e. "remote_cvc3" and "remote_z3".
  2041 
  2042 * Added basic SMT support for datatypes, records, and typedefs using
  2043 the oracle mode (no proofs).  Direct support of pairs has been dropped
  2044 in exchange (pass theorems fst_conv snd_conv pair_collapse to the SMT
  2045 support for a similar behavior).  Minor INCOMPATIBILITY.
  2046 
  2047 * Changed SMT configuration options:
  2048   - Renamed:
  2049     z3_proofs ~> smt_oracle (with inverted meaning)
  2050     z3_trace_assms ~> smt_trace_used_facts
  2051     INCOMPATIBILITY.
  2052   - Added:
  2053     smt_verbose
  2054     smt_random_seed
  2055     smt_datatypes
  2056     smt_infer_triggers
  2057     smt_monomorph_limit
  2058     cvc3_options
  2059     remote_cvc3_options
  2060     remote_z3_options
  2061     yices_options
  2062 
  2063 * Boogie output files (.b2i files) need to be declared in the theory
  2064 header.
  2065 
  2066 * Simplification procedure "list_to_set_comprehension" rewrites list
  2067 comprehensions applied to List.set to set comprehensions.  Occasional
  2068 INCOMPATIBILITY, may be deactivated like this:
  2069 
  2070   declare [[simproc del: list_to_set_comprehension]]
  2071 
  2072 * Removed old version of primrec package.  INCOMPATIBILITY.
  2073 
  2074 * Removed simplifier congruence rule of "prod_case", as has for long
  2075 been the case with "split".  INCOMPATIBILITY.
  2076 
  2077 * String.literal is a type, but not a datatype.  INCOMPATIBILITY.
  2078 
  2079 * Removed [split_format ... and ... and ...] version of
  2080 [split_format].  Potential INCOMPATIBILITY.
  2081 
  2082 * Predicate "sorted" now defined inductively, with nice induction
  2083 rules.  INCOMPATIBILITY: former sorted.simps now named sorted_simps.
  2084 
  2085 * Constant "contents" renamed to "the_elem", to free the generic name
  2086 contents for other uses.  INCOMPATIBILITY.
  2087 
  2088 * Renamed class eq and constant eq (for code generation) to class
  2089 equal and constant equal, plus renaming of related facts and various
  2090 tuning.  INCOMPATIBILITY.
  2091 
  2092 * Dropped type classes mult_mono and mult_mono1.  INCOMPATIBILITY.
  2093 
  2094 * Removed output syntax "'a ~=> 'b" for "'a => 'b option".
  2095 INCOMPATIBILITY.
  2096 
  2097 * Renamed theory Fset to Cset, type Fset.fset to Cset.set, in order to
  2098 avoid confusion with finite sets.  INCOMPATIBILITY.
  2099 
  2100 * Abandoned locales equiv, congruent and congruent2 for equivalence
  2101 relations.  INCOMPATIBILITY: use equivI rather than equiv_intro (same
  2102 for congruent(2)).
  2103 
  2104 * Some previously unqualified names have been qualified:
  2105 
  2106   types
  2107     bool ~> HOL.bool
  2108     nat ~> Nat.nat
  2109 
  2110   constants
  2111     Trueprop ~> HOL.Trueprop
  2112     True ~> HOL.True
  2113     False ~> HOL.False
  2114     op & ~> HOL.conj
  2115     op | ~> HOL.disj
  2116     op --> ~> HOL.implies
  2117     op = ~> HOL.eq
  2118     Not ~> HOL.Not
  2119     The ~> HOL.The
  2120     All ~> HOL.All
  2121     Ex ~> HOL.Ex
  2122     Ex1 ~> HOL.Ex1
  2123     Let ~> HOL.Let
  2124     If ~> HOL.If
  2125     Ball ~> Set.Ball
  2126     Bex ~> Set.Bex
  2127     Suc ~> Nat.Suc
  2128     Pair ~> Product_Type.Pair
  2129     fst ~> Product_Type.fst
  2130     snd ~> Product_Type.snd
  2131     curry ~> Product_Type.curry
  2132     op : ~> Set.member
  2133     Collect ~> Set.Collect
  2134 
  2135 INCOMPATIBILITY.
  2136 
  2137 * More canonical naming convention for some fundamental definitions:
  2138 
  2139     bot_bool_eq ~> bot_bool_def
  2140     top_bool_eq ~> top_bool_def
  2141     inf_bool_eq ~> inf_bool_def
  2142     sup_bool_eq ~> sup_bool_def
  2143     bot_fun_eq  ~> bot_fun_def
  2144     top_fun_eq  ~> top_fun_def
  2145     inf_fun_eq  ~> inf_fun_def
  2146     sup_fun_eq  ~> sup_fun_def
  2147 
  2148 INCOMPATIBILITY.
  2149 
  2150 * More stylized fact names:
  2151 
  2152   expand_fun_eq ~> fun_eq_iff
  2153   expand_set_eq ~> set_eq_iff
  2154   set_ext       ~> set_eqI
  2155   nat_number    ~> eval_nat_numeral
  2156 
  2157 INCOMPATIBILITY.
  2158 
  2159 * Refactoring of code-generation specific operations in theory List:
  2160 
  2161   constants
  2162     null ~> List.null
  2163 
  2164   facts
  2165     mem_iff ~> member_def
  2166     null_empty ~> null_def
  2167 
  2168 INCOMPATIBILITY.  Note that these were not supposed to be used
  2169 regularly unless for striking reasons; their main purpose was code
  2170 generation.
  2171 
  2172 Various operations from the Haskell prelude are used for generating
  2173 Haskell code.
  2174 
  2175 * Term "bij f" is now an abbreviation of "bij_betw f UNIV UNIV".  Term
  2176 "surj f" is now an abbreviation of "range f = UNIV".  The theorems
  2177 bij_def and surj_def are unchanged.  INCOMPATIBILITY.
  2178 
  2179 * Abolished some non-alphabetic type names: "prod" and "sum" replace
  2180 "*" and "+" respectively.  INCOMPATIBILITY.
  2181 
  2182 * Name "Plus" of disjoint sum operator "<+>" is now hidden.  Write
  2183 "Sum_Type.Plus" instead.
  2184 
  2185 * Constant "split" has been merged with constant "prod_case"; names of
  2186 ML functions, facts etc. involving split have been retained so far,
  2187 though.  INCOMPATIBILITY.
  2188 
  2189 * Dropped old infix syntax "_ mem _" for List.member; use "_ : set _"
  2190 instead.  INCOMPATIBILITY.
  2191 
  2192 * Removed lemma "Option.is_none_none" which duplicates "is_none_def".
  2193 INCOMPATIBILITY.
  2194 
  2195 * Former theory Library/Enum is now part of the HOL-Main image.
  2196 INCOMPATIBILITY: all constants of the Enum theory now have to be
  2197 referred to by its qualified name.
  2198 
  2199   enum    ~>  Enum.enum
  2200   nlists  ~>  Enum.nlists
  2201   product ~>  Enum.product
  2202 
  2203 * Theory Library/Monad_Syntax provides do-syntax for monad types.
  2204 Syntax in Library/State_Monad has been changed to avoid ambiguities.
  2205 INCOMPATIBILITY.
  2206 
  2207 * Theory Library/SetsAndFunctions has been split into
  2208 Library/Function_Algebras and Library/Set_Algebras; canonical names
  2209 for instance definitions for functions; various improvements.
  2210 INCOMPATIBILITY.
  2211 
  2212 * Theory Library/Multiset provides stable quicksort implementation of
  2213 sort_key.
  2214 
  2215 * Theory Library/Multiset: renamed empty_idemp ~> empty_neutral.
  2216 INCOMPATIBILITY.
  2217 
  2218 * Session Multivariate_Analysis: introduced a type class for euclidean
  2219 space.  Most theorems are now stated in terms of euclidean spaces
  2220 instead of finite cartesian products.
  2221 
  2222   types
  2223     real ^ 'n ~>  'a::real_vector
  2224               ~>  'a::euclidean_space
  2225               ~>  'a::ordered_euclidean_space
  2226         (depends on your needs)
  2227 
  2228   constants
  2229      _ $ _        ~> _ $$ _
  2230      \<chi> x. _  ~> \<chi>\<chi> x. _
  2231      CARD('n)     ~> DIM('a)
  2232 
  2233 Also note that the indices are now natural numbers and not from some
  2234 finite type. Finite cartesian products of euclidean spaces, products
  2235 of euclidean spaces the real and complex numbers are instantiated to
  2236 be euclidean_spaces.  INCOMPATIBILITY.
  2237 
  2238 * Session Probability: introduced pextreal as positive extended real
  2239 numbers.  Use pextreal as value for measures.  Introduce the
  2240 Radon-Nikodym derivative, product spaces and Fubini's theorem for
  2241 arbitrary sigma finite measures.  Introduces Lebesgue measure based on
  2242 the integral in Multivariate Analysis.  INCOMPATIBILITY.
  2243 
  2244 * Session Imperative_HOL: revamped, corrected dozens of inadequacies.
  2245 INCOMPATIBILITY.
  2246 
  2247 * Session SPARK (with image HOL-SPARK) provides commands to load and
  2248 prove verification conditions generated by the SPARK Ada program
  2249 verifier.  See also src/HOL/SPARK and src/HOL/SPARK/Examples.
  2250 
  2251 
  2252 *** HOL-Algebra ***
  2253 
  2254 * Theorems for additive ring operations (locale abelian_monoid and
  2255 descendants) are generated by interpretation from their multiplicative
  2256 counterparts.  Names (in particular theorem names) have the mandatory
  2257 qualifier 'add'.  Previous theorem names are redeclared for
  2258 compatibility.
  2259 
  2260 * Structure "int_ring" is now an abbreviation (previously a
  2261 definition).  This fits more natural with advanced interpretations.
  2262 
  2263 
  2264 *** HOLCF ***
  2265 
  2266 * The domain package now runs in definitional mode by default: The
  2267 former command 'new_domain' is now called 'domain'.  To use the domain
  2268 package in its original axiomatic mode, use 'domain (unsafe)'.
  2269 INCOMPATIBILITY.
  2270 
  2271 * The new class "domain" is now the default sort.  Class "predomain"
  2272 is an unpointed version of "domain". Theories can be updated by
  2273 replacing sort annotations as shown below.  INCOMPATIBILITY.
  2274 
  2275   'a::type ~> 'a::countable
  2276   'a::cpo  ~> 'a::predomain
  2277   'a::pcpo ~> 'a::domain
  2278 
  2279 * The old type class "rep" has been superseded by class "domain".
  2280 Accordingly, users of the definitional package must remove any
  2281 "default_sort rep" declarations.  INCOMPATIBILITY.
  2282 
  2283 * The domain package (definitional mode) now supports unpointed
  2284 predomain argument types, as long as they are marked 'lazy'. (Strict
  2285 arguments must be in class "domain".) For example, the following
  2286 domain definition now works:
  2287 
  2288   domain natlist = nil | cons (lazy "nat discr") (lazy "natlist")
  2289 
  2290 * Theory HOLCF/Library/HOL_Cpo provides cpo and predomain class
  2291 instances for types from main HOL: bool, nat, int, char, 'a + 'b,
  2292 'a option, and 'a list.  Additionally, it configures fixrec and the
  2293 domain package to work with these types.  For example:
  2294 
  2295   fixrec isInl :: "('a + 'b) u -> tr"
  2296     where "isInl$(up$(Inl x)) = TT" | "isInl$(up$(Inr y)) = FF"
  2297 
  2298   domain V = VFun (lazy "V -> V") | VCon (lazy "nat") (lazy "V list")
  2299 
  2300 * The "(permissive)" option of fixrec has been replaced with a
  2301 per-equation "(unchecked)" option. See
  2302 src/HOL/HOLCF/Tutorial/Fixrec_ex.thy for examples. INCOMPATIBILITY.
  2303 
  2304 * The "bifinite" class no longer fixes a constant "approx"; the class
  2305 now just asserts that such a function exists.  INCOMPATIBILITY.
  2306 
  2307 * Former type "alg_defl" has been renamed to "defl".  HOLCF no longer
  2308 defines an embedding of type 'a defl into udom by default; instances
  2309 of "bifinite" and "domain" classes are available in
  2310 src/HOL/HOLCF/Library/Defl_Bifinite.thy.
  2311 
  2312 * The syntax "REP('a)" has been replaced with "DEFL('a)".
  2313 
  2314 * The predicate "directed" has been removed.  INCOMPATIBILITY.
  2315 
  2316 * The type class "finite_po" has been removed.  INCOMPATIBILITY.
  2317 
  2318 * The function "cprod_map" has been renamed to "prod_map".
  2319 INCOMPATIBILITY.
  2320 
  2321 * The monadic bind operator on each powerdomain has new binder syntax
  2322 similar to sets, e.g. "\<Union>\<sharp>x\<in>xs. t" represents
  2323 "upper_bind\<cdot>xs\<cdot>(\<Lambda> x. t)".
  2324 
  2325 * The infix syntax for binary union on each powerdomain has changed
  2326 from e.g. "+\<sharp>" to "\<union>\<sharp>", for consistency with set
  2327 syntax.  INCOMPATIBILITY.
  2328 
  2329 * The constant "UU" has been renamed to "bottom".  The syntax "UU" is
  2330 still supported as an input translation.
  2331 
  2332 * Renamed some theorems (the original names are also still available).
  2333 
  2334   expand_fun_below   ~> fun_below_iff
  2335   below_fun_ext      ~> fun_belowI
  2336   expand_cfun_eq     ~> cfun_eq_iff
  2337   ext_cfun           ~> cfun_eqI
  2338   expand_cfun_below  ~> cfun_below_iff
  2339   below_cfun_ext     ~> cfun_belowI
  2340   cont2cont_Rep_CFun ~> cont2cont_APP
  2341 
  2342 * The Abs and Rep functions for various types have changed names.
  2343 Related theorem names have also changed to match. INCOMPATIBILITY.
  2344 
  2345   Rep_CFun  ~> Rep_cfun
  2346   Abs_CFun  ~> Abs_cfun
  2347   Rep_Sprod ~> Rep_sprod
  2348   Abs_Sprod ~> Abs_sprod
  2349   Rep_Ssum  ~> Rep_ssum
  2350   Abs_Ssum  ~> Abs_ssum
  2351 
  2352 * Lemmas with names of the form *_defined_iff or *_strict_iff have
  2353 been renamed to *_bottom_iff.  INCOMPATIBILITY.
  2354 
  2355 * Various changes to bisimulation/coinduction with domain package:
  2356 
  2357   - Definitions of "bisim" constants no longer mention definedness.
  2358   - With mutual recursion, "bisim" predicate is now curried.
  2359   - With mutual recursion, each type gets a separate coind theorem.
  2360   - Variable names in bisim_def and coinduct rules have changed.
  2361 
  2362 INCOMPATIBILITY.
  2363 
  2364 * Case combinators generated by the domain package for type "foo" are
  2365 now named "foo_case" instead of "foo_when".  INCOMPATIBILITY.
  2366 
  2367 * Several theorems have been renamed to more accurately reflect the
  2368 names of constants and types involved.  INCOMPATIBILITY.
  2369 
  2370   thelub_const    ~> lub_const
  2371   lub_const       ~> is_lub_const
  2372   thelubI         ~> lub_eqI
  2373   is_lub_lub      ~> is_lubD2
  2374   lubI            ~> is_lub_lub
  2375   unique_lub      ~> is_lub_unique
  2376   is_ub_lub       ~> is_lub_rangeD1
  2377   lub_bin_chain   ~> is_lub_bin_chain
  2378   lub_fun         ~> is_lub_fun
  2379   thelub_fun      ~> lub_fun
  2380   thelub_cfun     ~> lub_cfun
  2381   thelub_Pair     ~> lub_Pair
  2382   lub_cprod       ~> is_lub_prod
  2383   thelub_cprod    ~> lub_prod
  2384   minimal_cprod   ~> minimal_prod
  2385   inst_cprod_pcpo ~> inst_prod_pcpo
  2386   UU_I            ~> bottomI
  2387   compact_UU      ~> compact_bottom
  2388   deflation_UU    ~> deflation_bottom
  2389   finite_deflation_UU ~> finite_deflation_bottom
  2390 
  2391 * Many legacy theorem names have been discontinued.  INCOMPATIBILITY.
  2392 
  2393   sq_ord_less_eq_trans ~> below_eq_trans
  2394   sq_ord_eq_less_trans ~> eq_below_trans
  2395   refl_less            ~> below_refl
  2396   trans_less           ~> below_trans
  2397   antisym_less         ~> below_antisym
  2398   antisym_less_inverse ~> po_eq_conv [THEN iffD1]
  2399   box_less             ~> box_below
  2400   rev_trans_less       ~> rev_below_trans
  2401   not_less2not_eq      ~> not_below2not_eq
  2402   less_UU_iff          ~> below_UU_iff
  2403   flat_less_iff        ~> flat_below_iff
  2404   adm_less             ~> adm_below
  2405   adm_not_less         ~> adm_not_below
  2406   adm_compact_not_less ~> adm_compact_not_below
  2407   less_fun_def         ~> below_fun_def
  2408   expand_fun_less      ~> fun_below_iff
  2409   less_fun_ext         ~> fun_belowI
  2410   less_discr_def       ~> below_discr_def
  2411   discr_less_eq        ~> discr_below_eq
  2412   less_unit_def        ~> below_unit_def
  2413   less_cprod_def       ~> below_prod_def
  2414   prod_lessI           ~> prod_belowI
  2415   Pair_less_iff        ~> Pair_below_iff
  2416   fst_less_iff         ~> fst_below_iff
  2417   snd_less_iff         ~> snd_below_iff
  2418   expand_cfun_less     ~> cfun_below_iff
  2419   less_cfun_ext        ~> cfun_belowI
  2420   injection_less       ~> injection_below
  2421   less_up_def          ~> below_up_def
  2422   not_Iup_less         ~> not_Iup_below
  2423   Iup_less             ~> Iup_below
  2424   up_less              ~> up_below
  2425   Def_inject_less_eq   ~> Def_below_Def
  2426   Def_less_is_eq       ~> Def_below_iff
  2427   spair_less_iff       ~> spair_below_iff
  2428   less_sprod           ~> below_sprod
  2429   spair_less           ~> spair_below
  2430   sfst_less_iff        ~> sfst_below_iff
  2431   ssnd_less_iff        ~> ssnd_below_iff
  2432   fix_least_less       ~> fix_least_below
  2433   dist_less_one        ~> dist_below_one
  2434   less_ONE             ~> below_ONE
  2435   ONE_less_iff         ~> ONE_below_iff
  2436   less_sinlD           ~> below_sinlD
  2437   less_sinrD           ~> below_sinrD
  2438 
  2439 
  2440 *** FOL and ZF ***
  2441 
  2442 * All constant names are now qualified internally and use proper
  2443 identifiers, e.g. "IFOL.eq" instead of "op =".  INCOMPATIBILITY.
  2444 
  2445 
  2446 *** ML ***
  2447 
  2448 * Antiquotation @{assert} inlines a function bool -> unit that raises
  2449 Fail if the argument is false.  Due to inlining the source position of
  2450 failed assertions is included in the error output.
  2451 
  2452 * Discontinued antiquotation @{theory_ref}, which is obsolete since ML
  2453 text is in practice always evaluated with a stable theory checkpoint.
  2454 Minor INCOMPATIBILITY, use (Theory.check_thy @{theory}) instead.
  2455 
  2456 * Antiquotation @{theory A} refers to theory A from the ancestry of
  2457 the current context, not any accidental theory loader state as before.
  2458 Potential INCOMPATIBILITY, subtle change in semantics.
  2459 
  2460 * Syntax.pretty_priority (default 0) configures the required priority
  2461 of pretty-printed output and thus affects insertion of parentheses.
  2462 
  2463 * Syntax.default_root (default "any") configures the inner syntax
  2464 category (nonterminal symbol) for parsing of terms.
  2465 
  2466 * Former exception Library.UnequalLengths now coincides with
  2467 ListPair.UnequalLengths.
  2468 
  2469 * Renamed structure MetaSimplifier to Raw_Simplifier.  Note that the
  2470 main functionality is provided by structure Simplifier.
  2471 
  2472 * Renamed raw "explode" function to "raw_explode" to emphasize its
  2473 meaning.  Note that internally to Isabelle, Symbol.explode is used in
  2474 almost all situations.
  2475 
  2476 * Discontinued obsolete function sys_error and exception SYS_ERROR.
  2477 See implementation manual for further details on exceptions in
  2478 Isabelle/ML.
  2479 
  2480 * Renamed setmp_noncritical to Unsynchronized.setmp to emphasize its
  2481 meaning.
  2482 
  2483 * Renamed structure PureThy to Pure_Thy and moved most of its
  2484 operations to structure Global_Theory, to emphasize that this is
  2485 rarely-used global-only stuff.
  2486 
  2487 * Discontinued Output.debug.  Minor INCOMPATIBILITY, use plain writeln
  2488 instead (or tracing for high-volume output).
  2489 
  2490 * Configuration option show_question_marks only affects regular pretty
  2491 printing of types and terms, not raw Term.string_of_vname.
  2492 
  2493 * ML_Context.thm and ML_Context.thms are no longer pervasive.  Rare
  2494 INCOMPATIBILITY, superseded by static antiquotations @{thm} and
  2495 @{thms} for most purposes.
  2496 
  2497 * ML structure Unsynchronized is never opened, not even in Isar
  2498 interaction mode as before.  Old Unsynchronized.set etc. have been
  2499 discontinued -- use plain := instead.  This should be *rare* anyway,
  2500 since modern tools always work via official context data, notably
  2501 configuration options.
  2502 
  2503 * Parallel and asynchronous execution requires special care concerning
  2504 interrupts.  Structure Exn provides some convenience functions that
  2505 avoid working directly with raw Interrupt.  User code must not absorb
  2506 interrupts -- intermediate handling (for cleanup etc.) needs to be
  2507 followed by re-raising of the original exception.  Another common
  2508 source of mistakes are "handle _" patterns, which make the meaning of
  2509 the program subject to physical effects of the environment.
  2510 
  2511 
  2512 
  2513 New in Isabelle2009-2 (June 2010)
  2514 ---------------------------------
  2515 
  2516 *** General ***
  2517 
  2518 * Authentic syntax for *all* logical entities (type classes, type
  2519 constructors, term constants): provides simple and robust
  2520 correspondence between formal entities and concrete syntax.  Within
  2521 the parse tree / AST representations, "constants" are decorated by
  2522 their category (class, type, const) and spelled out explicitly with
  2523 their full internal name.
  2524 
  2525 Substantial INCOMPATIBILITY concerning low-level syntax declarations
  2526 and translations (translation rules and translation functions in ML).
  2527 Some hints on upgrading:
  2528 
  2529   - Many existing uses of 'syntax' and 'translations' can be replaced
  2530     by more modern 'type_notation', 'notation' and 'abbreviation',
  2531     which are independent of this issue.
  2532 
  2533   - 'translations' require markup within the AST; the term syntax
  2534     provides the following special forms:
  2535 
  2536       CONST c   -- produces syntax version of constant c from context
  2537       XCONST c  -- literally c, checked as constant from context
  2538       c         -- literally c, if declared by 'syntax'
  2539 
  2540     Plain identifiers are treated as AST variables -- occasionally the
  2541     system indicates accidental variables via the error "rhs contains
  2542     extra variables".
  2543 
  2544     Type classes and type constructors are marked according to their
  2545     concrete syntax.  Some old translations rules need to be written
  2546     for the "type" category, using type constructor application
  2547     instead of pseudo-term application of the default category
  2548     "logic".
  2549 
  2550   - 'parse_translation' etc. in ML may use the following
  2551     antiquotations:
  2552 
  2553       @{class_syntax c}   -- type class c within parse tree / AST
  2554       @{term_syntax c}    -- type constructor c within parse tree / AST
  2555       @{const_syntax c}   -- ML version of "CONST c" above
  2556       @{syntax_const c}   -- literally c (checked wrt. 'syntax' declarations)
  2557 
  2558   - Literal types within 'typed_print_translations', i.e. those *not*
  2559     represented as pseudo-terms are represented verbatim.  Use @{class
  2560     c} or @{type_name c} here instead of the above syntax
  2561     antiquotations.
  2562 
  2563 Note that old non-authentic syntax was based on unqualified base
  2564 names, so all of the above "constant" names would coincide.  Recall
  2565 that 'print_syntax' and ML_command "set Syntax.trace_ast" help to
  2566 diagnose syntax problems.
  2567 
  2568 * Type constructors admit general mixfix syntax, not just infix.
  2569 
  2570 * Concrete syntax may be attached to local entities without a proof
  2571 body, too.  This works via regular mixfix annotations for 'fix',
  2572 'def', 'obtain' etc. or via the explicit 'write' command, which is
  2573 similar to the 'notation' command in theory specifications.
  2574 
  2575 * Discontinued unnamed infix syntax (legacy feature for many years) --
  2576 need to specify constant name and syntax separately.  Internal ML
  2577 datatype constructors have been renamed from InfixName to Infix etc.
  2578 Minor INCOMPATIBILITY.
  2579 
  2580 * Schematic theorem statements need to be explicitly markup as such,
  2581 via commands 'schematic_lemma', 'schematic_theorem',
  2582 'schematic_corollary'.  Thus the relevance of the proof is made
  2583 syntactically clear, which impacts performance in a parallel or
  2584 asynchronous interactive environment.  Minor INCOMPATIBILITY.
  2585 
  2586 * Use of cumulative prems via "!" in some proof methods has been
  2587 discontinued (old legacy feature).
  2588 
  2589 * References 'trace_simp' and 'debug_simp' have been replaced by
  2590 configuration options stored in the context. Enabling tracing (the
  2591 case of debugging is similar) in proofs works via
  2592 
  2593   using [[trace_simp = true]]
  2594 
  2595 Tracing is then active for all invocations of the simplifier in
  2596 subsequent goal refinement steps. Tracing may also still be enabled or
  2597 disabled via the ProofGeneral settings menu.
  2598 
  2599 * Separate commands 'hide_class', 'hide_type', 'hide_const',
  2600 'hide_fact' replace the former 'hide' KIND command.  Minor
  2601 INCOMPATIBILITY.
  2602 
  2603 * Improved parallelism of proof term normalization: usedir -p2 -q0 is
  2604 more efficient than combinations with -q1 or -q2.
  2605 
  2606 
  2607 *** Pure ***
  2608 
  2609 * Proofterms record type-class reasoning explicitly, using the
  2610 "unconstrain" operation internally.  This eliminates all sort
  2611 constraints from a theorem and proof, introducing explicit
  2612 OFCLASS-premises.  On the proof term level, this operation is
  2613 automatically applied at theorem boundaries, such that closed proofs
  2614 are always free of sort constraints.  INCOMPATIBILITY for tools that
  2615 inspect proof terms.
  2616 
  2617 * Local theory specifications may depend on extra type variables that
  2618 are not present in the result type -- arguments TYPE('a) :: 'a itself
  2619 are added internally.  For example:
  2620 
  2621   definition unitary :: bool where "unitary = (ALL (x::'a) y. x = y)"
  2622 
  2623 * Predicates of locales introduced by classes carry a mandatory
  2624 "class" prefix.  INCOMPATIBILITY.
  2625 
  2626 * Vacuous class specifications observe default sort.  INCOMPATIBILITY.
  2627 
  2628 * Old 'axclass' command has been discontinued.  INCOMPATIBILITY, use
  2629 'class' instead.
  2630 
  2631 * Command 'code_reflect' allows to incorporate generated ML code into
  2632 runtime environment; replaces immature code_datatype antiquotation.
  2633 INCOMPATIBILITY.
  2634 
  2635 * Code generator: simple concept for abstract datatypes obeying
  2636 invariants.
  2637 
  2638 * Code generator: details of internal data cache have no impact on the
  2639 user space functionality any longer.
  2640 
  2641 * Methods "unfold_locales" and "intro_locales" ignore non-locale
  2642 subgoals.  This is more appropriate for interpretations with 'where'.
  2643 INCOMPATIBILITY.
  2644 
  2645 * Command 'example_proof' opens an empty proof body.  This allows to
  2646 experiment with Isar, without producing any persistent result.
  2647 
  2648 * Commands 'type_notation' and 'no_type_notation' declare type syntax
  2649 within a local theory context, with explicit checking of the
  2650 constructors involved (in contrast to the raw 'syntax' versions).
  2651 
  2652 * Commands 'types' and 'typedecl' now work within a local theory
  2653 context -- without introducing dependencies on parameters or
  2654 assumptions, which is not possible in Isabelle/Pure.
  2655 
  2656 * Command 'defaultsort' has been renamed to 'default_sort', it works
  2657 within a local theory context.  Minor INCOMPATIBILITY.
  2658 
  2659 
  2660 *** HOL ***
  2661 
  2662 * Command 'typedef' now works within a local theory context -- without
  2663 introducing dependencies on parameters or assumptions, which is not
  2664 possible in Isabelle/Pure/HOL.  Note that the logical environment may
  2665 contain multiple interpretations of local typedefs (with different
  2666 non-emptiness proofs), even in a global theory context.
  2667 
  2668 * New package for quotient types.  Commands 'quotient_type' and
  2669 'quotient_definition' may be used for defining types and constants by
  2670 quotient constructions.  An example is the type of integers created by
  2671 quotienting pairs of natural numbers:
  2672 
  2673   fun
  2674     intrel :: "(nat * nat) => (nat * nat) => bool"
  2675   where
  2676     "intrel (x, y) (u, v) = (x + v = u + y)"
  2677 
  2678   quotient_type int = "nat * nat" / intrel
  2679     by (auto simp add: equivp_def expand_fun_eq)
  2680 
  2681   quotient_definition
  2682     "0::int" is "(0::nat, 0::nat)"
  2683 
  2684 The method "lifting" can be used to lift of theorems from the
  2685 underlying "raw" type to the quotient type.  The example
  2686 src/HOL/Quotient_Examples/FSet.thy includes such a quotient
  2687 construction and provides a reasoning infrastructure for finite sets.
  2688 
  2689 * Renamed Library/Quotient.thy to Library/Quotient_Type.thy to avoid
  2690 clash with new theory Quotient in Main HOL.
  2691 
  2692 * Moved the SMT binding into the main HOL session, eliminating
  2693 separate HOL-SMT session.
  2694 
  2695 * List membership infix mem operation is only an input abbreviation.
  2696 INCOMPATIBILITY.
  2697 
  2698 * Theory Library/Word.thy has been removed.  Use library Word/Word.thy
  2699 for future developements; former Library/Word.thy is still present in
  2700 the AFP entry RSAPPS.
  2701 
  2702 * Theorem Int.int_induct renamed to Int.int_of_nat_induct and is no
  2703 longer shadowed.  INCOMPATIBILITY.
  2704 
  2705 * Dropped theorem duplicate comp_arith; use semiring_norm instead.
  2706 INCOMPATIBILITY.
  2707 
  2708 * Dropped theorem RealPow.real_sq_order; use power2_le_imp_le instead.
  2709 INCOMPATIBILITY.
  2710 
  2711 * Dropped normalizing_semiring etc; use the facts in semiring classes
  2712 instead.  INCOMPATIBILITY.
  2713 
  2714 * Dropped several real-specific versions of lemmas about floor and
  2715 ceiling; use the generic lemmas from theory "Archimedean_Field"
  2716 instead.  INCOMPATIBILITY.
  2717 
  2718   floor_number_of_eq         ~> floor_number_of
  2719   le_floor_eq_number_of      ~> number_of_le_floor
  2720   le_floor_eq_zero           ~> zero_le_floor
  2721   le_floor_eq_one            ~> one_le_floor
  2722   floor_less_eq_number_of    ~> floor_less_number_of
  2723   floor_less_eq_zero         ~> floor_less_zero
  2724   floor_less_eq_one          ~> floor_less_one
  2725   less_floor_eq_number_of    ~> number_of_less_floor
  2726   less_floor_eq_zero         ~> zero_less_floor
  2727   less_floor_eq_one          ~> one_less_floor
  2728   floor_le_eq_number_of      ~> floor_le_number_of
  2729   floor_le_eq_zero           ~> floor_le_zero
  2730   floor_le_eq_one            ~> floor_le_one
  2731   floor_subtract_number_of   ~> floor_diff_number_of
  2732   floor_subtract_one         ~> floor_diff_one
  2733   ceiling_number_of_eq       ~> ceiling_number_of
  2734   ceiling_le_eq_number_of    ~> ceiling_le_number_of
  2735   ceiling_le_zero_eq         ~> ceiling_le_zero
  2736   ceiling_le_eq_one          ~> ceiling_le_one
  2737   less_ceiling_eq_number_of  ~> number_of_less_ceiling
  2738   less_ceiling_eq_zero       ~> zero_less_ceiling
  2739   less_ceiling_eq_one        ~> one_less_ceiling
  2740   ceiling_less_eq_number_of  ~> ceiling_less_number_of
  2741   ceiling_less_eq_zero       ~> ceiling_less_zero
  2742   ceiling_less_eq_one        ~> ceiling_less_one
  2743   le_ceiling_eq_number_of    ~> number_of_le_ceiling
  2744   le_ceiling_eq_zero         ~> zero_le_ceiling
  2745   le_ceiling_eq_one          ~> one_le_ceiling
  2746   ceiling_subtract_number_of ~> ceiling_diff_number_of
  2747   ceiling_subtract_one       ~> ceiling_diff_one
  2748 
  2749 * Theory "Finite_Set": various folding_XXX locales facilitate the
  2750 application of the various fold combinators on finite sets.
  2751 
  2752 * Library theory "RBT" renamed to "RBT_Impl"; new library theory "RBT"
  2753 provides abstract red-black tree type which is backed by "RBT_Impl" as
  2754 implementation.  INCOMPATIBILTY.
  2755 
  2756 * Theory Library/Coinductive_List has been removed -- superseded by
  2757 AFP/thys/Coinductive.
  2758 
  2759 * Theory PReal, including the type "preal" and related operations, has
  2760 been removed.  INCOMPATIBILITY.
  2761 
  2762 * Real: new development using Cauchy Sequences.
  2763 
  2764 * Split off theory "Big_Operators" containing setsum, setprod,
  2765 Inf_fin, Sup_fin, Min, Max from theory Finite_Set.  INCOMPATIBILITY.
  2766 
  2767 * Theory "Rational" renamed to "Rat", for consistency with "Nat",
  2768 "Int" etc.  INCOMPATIBILITY.
  2769 
  2770 * Constant Rat.normalize needs to be qualified.  INCOMPATIBILITY.
  2771 
  2772 * New set of rules "ac_simps" provides combined assoc / commute
  2773 rewrites for all interpretations of the appropriate generic locales.
  2774 
  2775 * Renamed theory "OrderedGroup" to "Groups" and split theory
  2776 "Ring_and_Field" into theories "Rings" and "Fields"; for more
  2777 appropriate and more consistent names suitable for name prefixes
  2778 within the HOL theories.  INCOMPATIBILITY.
  2779 
  2780 * Some generic constants have been put to appropriate theories:
  2781   - less_eq, less: Orderings
  2782   - zero, one, plus, minus, uminus, times, abs, sgn: Groups
  2783   - inverse, divide: Rings
  2784 INCOMPATIBILITY.
  2785 
  2786 * More consistent naming of type classes involving orderings (and
  2787 lattices):
  2788 
  2789     lower_semilattice                   ~> semilattice_inf
  2790     upper_semilattice                   ~> semilattice_sup
  2791 
  2792     dense_linear_order                  ~> dense_linorder
  2793 
  2794     pordered_ab_group_add               ~> ordered_ab_group_add
  2795     pordered_ab_group_add_abs           ~> ordered_ab_group_add_abs
  2796     pordered_ab_semigroup_add           ~> ordered_ab_semigroup_add
  2797     pordered_ab_semigroup_add_imp_le    ~> ordered_ab_semigroup_add_imp_le
  2798     pordered_cancel_ab_semigroup_add    ~> ordered_cancel_ab_semigroup_add
  2799     pordered_cancel_comm_semiring       ~> ordered_cancel_comm_semiring
  2800     pordered_cancel_semiring            ~> ordered_cancel_semiring
  2801     pordered_comm_monoid_add            ~> ordered_comm_monoid_add
  2802     pordered_comm_ring                  ~> ordered_comm_ring
  2803     pordered_comm_semiring              ~> ordered_comm_semiring
  2804     pordered_ring                       ~> ordered_ring
  2805     pordered_ring_abs                   ~> ordered_ring_abs
  2806     pordered_semiring                   ~> ordered_semiring
  2807 
  2808     ordered_ab_group_add                ~> linordered_ab_group_add
  2809     ordered_ab_semigroup_add            ~> linordered_ab_semigroup_add
  2810     ordered_cancel_ab_semigroup_add     ~> linordered_cancel_ab_semigroup_add
  2811     ordered_comm_semiring_strict        ~> linordered_comm_semiring_strict
  2812     ordered_field                       ~> linordered_field
  2813     ordered_field_no_lb                 ~> linordered_field_no_lb
  2814     ordered_field_no_ub                 ~> linordered_field_no_ub
  2815     ordered_field_dense_linear_order    ~> dense_linordered_field
  2816     ordered_idom                        ~> linordered_idom
  2817     ordered_ring                        ~> linordered_ring
  2818     ordered_ring_le_cancel_factor       ~> linordered_ring_le_cancel_factor
  2819     ordered_ring_less_cancel_factor     ~> linordered_ring_less_cancel_factor
  2820     ordered_ring_strict                 ~> linordered_ring_strict
  2821     ordered_semidom                     ~> linordered_semidom
  2822     ordered_semiring                    ~> linordered_semiring
  2823     ordered_semiring_1                  ~> linordered_semiring_1
  2824     ordered_semiring_1_strict           ~> linordered_semiring_1_strict
  2825     ordered_semiring_strict             ~> linordered_semiring_strict
  2826 
  2827   The following slightly odd type classes have been moved to a
  2828   separate theory Library/Lattice_Algebras:
  2829 
  2830     lordered_ab_group_add               ~> lattice_ab_group_add
  2831     lordered_ab_group_add_abs           ~> lattice_ab_group_add_abs
  2832     lordered_ab_group_add_meet          ~> semilattice_inf_ab_group_add
  2833     lordered_ab_group_add_join          ~> semilattice_sup_ab_group_add
  2834     lordered_ring                       ~> lattice_ring
  2835 
  2836 INCOMPATIBILITY.
  2837 
  2838 * Refined field classes:
  2839   - classes division_ring_inverse_zero, field_inverse_zero,
  2840     linordered_field_inverse_zero include rule inverse 0 = 0 --
  2841     subsumes former division_by_zero class;
  2842   - numerous lemmas have been ported from field to division_ring.
  2843 INCOMPATIBILITY.
  2844 
  2845 * Refined algebra theorem collections:
  2846   - dropped theorem group group_simps, use algebra_simps instead;
  2847   - dropped theorem group ring_simps, use field_simps instead;
  2848   - proper theorem collection field_simps subsumes former theorem
  2849     groups field_eq_simps and field_simps;
  2850   - dropped lemma eq_minus_self_iff which is a duplicate for
  2851     equal_neg_zero.
  2852 INCOMPATIBILITY.
  2853 
  2854 * Theory Finite_Set and List: some lemmas have been generalized from
  2855 sets to lattices:
  2856 
  2857   fun_left_comm_idem_inter      ~> fun_left_comm_idem_inf
  2858   fun_left_comm_idem_union      ~> fun_left_comm_idem_sup
  2859   inter_Inter_fold_inter        ~> inf_Inf_fold_inf
  2860   union_Union_fold_union        ~> sup_Sup_fold_sup
  2861   Inter_fold_inter              ~> Inf_fold_inf
  2862   Union_fold_union              ~> Sup_fold_sup
  2863   inter_INTER_fold_inter        ~> inf_INFI_fold_inf
  2864   union_UNION_fold_union        ~> sup_SUPR_fold_sup
  2865   INTER_fold_inter              ~> INFI_fold_inf
  2866   UNION_fold_union              ~> SUPR_fold_sup
  2867 
  2868 * Theory "Complete_Lattice": lemmas top_def and bot_def have been
  2869 replaced by the more convenient lemmas Inf_empty and Sup_empty.
  2870 Dropped lemmas Inf_insert_simp and Sup_insert_simp, which are subsumed
  2871 by Inf_insert and Sup_insert.  Lemmas Inf_UNIV and Sup_UNIV replace
  2872 former Inf_Univ and Sup_Univ.  Lemmas inf_top_right and sup_bot_right
  2873 subsume inf_top and sup_bot respectively.  INCOMPATIBILITY.
  2874 
  2875 * Reorganized theory Multiset: swapped notation of pointwise and
  2876 multiset order:
  2877 
  2878   - pointwise ordering is instance of class order with standard syntax
  2879     <= and <;
  2880   - multiset ordering has syntax <=# and <#; partial order properties
  2881     are provided by means of interpretation with prefix
  2882     multiset_order;
  2883   - less duplication, less historical organization of sections,
  2884     conversion from associations lists to multisets, rudimentary code
  2885     generation;
  2886   - use insert_DiffM2 [symmetric] instead of elem_imp_eq_diff_union,
  2887     if needed.
  2888 
  2889 Renamed:
  2890 
  2891   multiset_eq_conv_count_eq  ~>  multiset_ext_iff
  2892   multi_count_ext  ~>  multiset_ext
  2893   diff_union_inverse2  ~>  diff_union_cancelR
  2894 
  2895 INCOMPATIBILITY.
  2896 
  2897 * Theory Permutation: replaced local "remove" by List.remove1.
  2898 
  2899 * Code generation: ML and OCaml code is decorated with signatures.
  2900 
  2901 * Theory List: added transpose.
  2902 
  2903 * Library/Nat_Bijection.thy is a collection of bijective functions
  2904 between nat and other types, which supersedes the older libraries
  2905 Library/Nat_Int_Bij.thy and HOLCF/NatIso.thy.  INCOMPATIBILITY.
  2906 
  2907   Constants:
  2908   Nat_Int_Bij.nat2_to_nat         ~> prod_encode
  2909   Nat_Int_Bij.nat_to_nat2         ~> prod_decode
  2910   Nat_Int_Bij.int_to_nat_bij      ~> int_encode
  2911   Nat_Int_Bij.nat_to_int_bij      ~> int_decode
  2912   Countable.pair_encode           ~> prod_encode
  2913   NatIso.prod2nat                 ~> prod_encode
  2914   NatIso.nat2prod                 ~> prod_decode
  2915   NatIso.sum2nat                  ~> sum_encode
  2916   NatIso.nat2sum                  ~> sum_decode
  2917   NatIso.list2nat                 ~> list_encode
  2918   NatIso.nat2list                 ~> list_decode
  2919   NatIso.set2nat                  ~> set_encode
  2920   NatIso.nat2set                  ~> set_decode
  2921 
  2922   Lemmas:
  2923   Nat_Int_Bij.bij_nat_to_int_bij  ~> bij_int_decode
  2924   Nat_Int_Bij.nat2_to_nat_inj     ~> inj_prod_encode
  2925   Nat_Int_Bij.nat2_to_nat_surj    ~> surj_prod_encode
  2926   Nat_Int_Bij.nat_to_nat2_inj     ~> inj_prod_decode
  2927   Nat_Int_Bij.nat_to_nat2_surj    ~> surj_prod_decode
  2928   Nat_Int_Bij.i2n_n2i_id          ~> int_encode_inverse
  2929   Nat_Int_Bij.n2i_i2n_id          ~> int_decode_inverse
  2930   Nat_Int_Bij.surj_nat_to_int_bij ~> surj_int_encode
  2931   Nat_Int_Bij.surj_int_to_nat_bij ~> surj_int_decode
  2932   Nat_Int_Bij.inj_nat_to_int_bij  ~> inj_int_encode
  2933   Nat_Int_Bij.inj_int_to_nat_bij  ~> inj_int_decode
  2934   Nat_Int_Bij.bij_nat_to_int_bij  ~> bij_int_encode
  2935   Nat_Int_Bij.bij_int_to_nat_bij  ~> bij_int_decode
  2936 
  2937 * Sledgehammer:
  2938   - Renamed ATP commands:
  2939     atp_info     ~> sledgehammer running_atps
  2940     atp_kill     ~> sledgehammer kill_atps
  2941     atp_messages ~> sledgehammer messages
  2942     atp_minimize ~> sledgehammer minimize
  2943     print_atps   ~> sledgehammer available_atps
  2944     INCOMPATIBILITY.
  2945   - Added user's manual ("isabelle doc sledgehammer").
  2946   - Added option syntax and "sledgehammer_params" to customize
  2947     Sledgehammer's behavior.  See the manual for details.
  2948   - Modified the Isar proof reconstruction code so that it produces
  2949     direct proofs rather than proofs by contradiction.  (This feature
  2950     is still experimental.)
  2951   - Made Isar proof reconstruction work for SPASS, remote ATPs, and in
  2952     full-typed mode.
  2953   - Added support for TPTP syntax for SPASS via the "spass_tptp" ATP.
  2954 
  2955 * Nitpick:
  2956   - Added and implemented "binary_ints" and "bits" options.
  2957   - Added "std" option and implemented support for nonstandard models.
  2958   - Added and implemented "finitize" option to improve the precision
  2959     of infinite datatypes based on a monotonicity analysis.
  2960   - Added support for quotient types.
  2961   - Added support for "specification" and "ax_specification"
  2962     constructs.
  2963   - Added support for local definitions (for "function" and
  2964     "termination" proofs).
  2965   - Added support for term postprocessors.
  2966   - Optimized "Multiset.multiset" and "FinFun.finfun".
  2967   - Improved efficiency of "destroy_constrs" optimization.
  2968   - Fixed soundness bugs related to "destroy_constrs" optimization and
  2969     record getters.
  2970   - Fixed soundness bug related to higher-order constructors.
  2971   - Fixed soundness bug when "full_descrs" is enabled.
  2972   - Improved precision of set constructs.
  2973   - Added "atoms" option.
  2974   - Added cache to speed up repeated Kodkod invocations on the same
  2975     problems.
  2976   - Renamed "MiniSatJNI", "zChaffJNI", "BerkMinAlloy", and
  2977     "SAT4JLight" to "MiniSat_JNI", "zChaff_JNI", "BerkMin_Alloy", and
  2978     "SAT4J_Light".  INCOMPATIBILITY.
  2979   - Removed "skolemize", "uncurry", "sym_break", "flatten_prop",
  2980     "sharing_depth", and "show_skolems" options.  INCOMPATIBILITY.
  2981   - Removed "nitpick_intro" attribute.  INCOMPATIBILITY.
  2982 
  2983 * Method "induct" now takes instantiations of the form t, where t is not
  2984   a variable, as a shorthand for "x == t", where x is a fresh variable.
  2985   If this is not intended, t has to be enclosed in parentheses.
  2986   By default, the equalities generated by definitional instantiations
  2987   are pre-simplified, which may cause parameters of inductive cases
  2988   to disappear, or may even delete some of the inductive cases.
  2989   Use "induct (no_simp)" instead of "induct" to restore the old
  2990   behaviour. The (no_simp) option is also understood by the "cases"
  2991   and "nominal_induct" methods, which now perform pre-simplification, too.
  2992   INCOMPATIBILITY.
  2993 
  2994 
  2995 *** HOLCF ***
  2996 
  2997 * Variable names in lemmas generated by the domain package have
  2998 changed; the naming scheme is now consistent with the HOL datatype
  2999 package.  Some proof scripts may be affected, INCOMPATIBILITY.
  3000 
  3001 * The domain package no longer defines the function "foo_copy" for
  3002 recursive domain "foo".  The reach lemma is now stated directly in
  3003 terms of "foo_take".  Lemmas and proofs that mention "foo_copy" must
  3004 be reformulated in terms of "foo_take", INCOMPATIBILITY.
  3005 
  3006 * Most definedness lemmas generated by the domain package (previously
  3007 of the form "x ~= UU ==> foo$x ~= UU") now have an if-and-only-if form
  3008 like "foo$x = UU <-> x = UU", which works better as a simp rule.
  3009 Proofs that used definedness lemmas as intro rules may break,
  3010 potential INCOMPATIBILITY.
  3011 
  3012 * Induction and casedist rules generated by the domain package now
  3013 declare proper case_names (one called "bottom", and one named for each
  3014 constructor).  INCOMPATIBILITY.
  3015 
  3016 * For mutually-recursive domains, separate "reach" and "take_lemma"
  3017 rules are generated for each domain, INCOMPATIBILITY.
  3018 
  3019   foo_bar.reach       ~> foo.reach  bar.reach
  3020   foo_bar.take_lemmas ~> foo.take_lemma  bar.take_lemma
  3021 
  3022 * Some lemmas generated by the domain package have been renamed for
  3023 consistency with the datatype package, INCOMPATIBILITY.
  3024 
  3025   foo.ind        ~> foo.induct
  3026   foo.finite_ind ~> foo.finite_induct
  3027   foo.coind      ~> foo.coinduct
  3028   foo.casedist   ~> foo.exhaust
  3029   foo.exhaust    ~> foo.nchotomy
  3030 
  3031 * For consistency with other definition packages, the fixrec package
  3032 now generates qualified theorem names, INCOMPATIBILITY.
  3033 
  3034   foo_simps  ~> foo.simps
  3035   foo_unfold ~> foo.unfold
  3036   foo_induct ~> foo.induct
  3037 
  3038 * The "fixrec_simp" attribute has been removed.  The "fixrec_simp"
  3039 method and internal fixrec proofs now use the default simpset instead.
  3040 INCOMPATIBILITY.
  3041 
  3042 * The "contlub" predicate has been removed.  Proof scripts should use
  3043 lemma contI2 in place of monocontlub2cont, INCOMPATIBILITY.
  3044 
  3045 * The "admw" predicate has been removed, INCOMPATIBILITY.
  3046 
  3047 * The constants cpair, cfst, and csnd have been removed in favor of
  3048 Pair, fst, and snd from Isabelle/HOL, INCOMPATIBILITY.
  3049 
  3050 
  3051 *** ML ***
  3052 
  3053 * Antiquotations for basic formal entities:
  3054 
  3055     @{class NAME}         -- type class
  3056     @{class_syntax NAME}  -- syntax representation of the above
  3057 
  3058     @{type_name NAME}     -- logical type
  3059     @{type_abbrev NAME}   -- type abbreviation
  3060     @{nonterminal NAME}   -- type of concrete syntactic category
  3061     @{type_syntax NAME}   -- syntax representation of any of the above
  3062 
  3063     @{const_name NAME}    -- logical constant (INCOMPATIBILITY)
  3064     @{const_abbrev NAME}  -- abbreviated constant
  3065     @{const_syntax NAME}  -- syntax representation of any of the above
  3066 
  3067 * Antiquotation @{syntax_const NAME} ensures that NAME refers to a raw
  3068 syntax constant (cf. 'syntax' command).
  3069 
  3070 * Antiquotation @{make_string} inlines a function to print arbitrary
  3071 values similar to the ML toplevel.  The result is compiler dependent
  3072 and may fall back on "?" in certain situations.
  3073 
  3074 * Diagnostic commands 'ML_val' and 'ML_command' may refer to
  3075 antiquotations @{Isar.state} and @{Isar.goal}.  This replaces impure
  3076 Isar.state() and Isar.goal(), which belong to the old TTY loop and do
  3077 not work with the asynchronous Isar document model.
  3078 
  3079 * Configuration options now admit dynamic default values, depending on
  3080 the context or even global references.
  3081 
  3082 * SHA1.digest digests strings according to SHA-1 (see RFC 3174).  It
  3083 uses an efficient external library if available (for Poly/ML).
  3084 
  3085 * Renamed some important ML structures, while keeping the old names
  3086 for some time as aliases within the structure Legacy:
  3087 
  3088   OuterKeyword  ~>  Keyword
  3089   OuterLex      ~>  Token
  3090   OuterParse    ~>  Parse
  3091   OuterSyntax   ~>  Outer_Syntax
  3092   PrintMode     ~>  Print_Mode
  3093   SpecParse     ~>  Parse_Spec
  3094   ThyInfo       ~>  Thy_Info
  3095   ThyLoad       ~>  Thy_Load
  3096   ThyOutput     ~>  Thy_Output
  3097   TypeInfer     ~>  Type_Infer
  3098 
  3099 Note that "open Legacy" simplifies porting of sources, but forgetting
  3100 to remove it again will complicate porting again in the future.
  3101 
  3102 * Most operations that refer to a global context are named
  3103 accordingly, e.g. Simplifier.global_context or
  3104 ProofContext.init_global.  There are some situations where a global
  3105 context actually works, but under normal circumstances one needs to
  3106 pass the proper local context through the code!
  3107 
  3108 * Discontinued old TheoryDataFun with its copy/init operation -- data
  3109 needs to be pure.  Functor Theory_Data_PP retains the traditional
  3110 Pretty.pp argument to merge, which is absent in the standard
  3111 Theory_Data version.
  3112 
  3113 * Sorts.certify_sort and derived "cert" operations for types and terms
  3114 no longer minimize sorts.  Thus certification at the boundary of the
  3115 inference kernel becomes invariant under addition of class relations,
  3116 which is an important monotonicity principle.  Sorts are now minimized
  3117 in the syntax layer only, at the boundary between the end-user and the
  3118 system.  Subtle INCOMPATIBILITY, may have to use Sign.minimize_sort
  3119 explicitly in rare situations.
  3120 
  3121 * Renamed old-style Drule.standard to Drule.export_without_context, to
  3122 emphasize that this is in no way a standard operation.
  3123 INCOMPATIBILITY.
  3124 
  3125 * Subgoal.FOCUS (and variants): resulting goal state is normalized as
  3126 usual for resolution.  Rare INCOMPATIBILITY.
  3127 
  3128 * Renamed varify/unvarify operations to varify_global/unvarify_global
  3129 to emphasize that these only work in a global situation (which is
  3130 quite rare).
  3131 
  3132 * Curried take and drop in library.ML; negative length is interpreted
  3133 as infinity (as in chop).  Subtle INCOMPATIBILITY.
  3134 
  3135 * Proof terms: type substitutions on proof constants now use canonical
  3136 order of type variables.  INCOMPATIBILITY for tools working with proof
  3137 terms.
  3138 
  3139 * Raw axioms/defs may no longer carry sort constraints, and raw defs
  3140 may no longer carry premises.  User-level specifications are
  3141 transformed accordingly by Thm.add_axiom/add_def.
  3142 
  3143 
  3144 *** System ***
  3145 
  3146 * Discontinued special HOL_USEDIR_OPTIONS for the main HOL image;
  3147 ISABELLE_USEDIR_OPTIONS applies uniformly to all sessions.  Note that
  3148 proof terms are enabled unconditionally in the new HOL-Proofs image.
  3149 
  3150 * Discontinued old ISABELLE and ISATOOL environment settings (legacy
  3151 feature since Isabelle2009).  Use ISABELLE_PROCESS and ISABELLE_TOOL,
  3152 respectively.
  3153 
  3154 * Old lib/scripts/polyml-platform is superseded by the
  3155 ISABELLE_PLATFORM setting variable, which defaults to the 32 bit
  3156 variant, even on a 64 bit machine.  The following example setting
  3157 prefers 64 bit if available:
  3158 
  3159   ML_PLATFORM="${ISABELLE_PLATFORM64:-$ISABELLE_PLATFORM}"
  3160 
  3161 * The preliminary Isabelle/jEdit application demonstrates the emerging
  3162 Isabelle/Scala layer for advanced prover interaction and integration.
  3163 See src/Tools/jEdit or "isabelle jedit" provided by the properly built
  3164 component.
  3165 
  3166 * "IsabelleText" is a Unicode font derived from Bitstream Vera Mono
  3167 and Bluesky TeX fonts.  It provides the usual Isabelle symbols,
  3168 similar to the default assignment of the document preparation system
  3169 (cf. isabellesym.sty).  The Isabelle/Scala class Isabelle_System
  3170 provides some operations for direct access to the font without asking
  3171 the user for manual installation.
  3172 
  3173 
  3174 
  3175 New in Isabelle2009-1 (December 2009)
  3176 -------------------------------------
  3177 
  3178 *** General ***
  3179 
  3180 * Discontinued old form of "escaped symbols" such as \\<forall>.  Only
  3181 one backslash should be used, even in ML sources.
  3182 
  3183 
  3184 *** Pure ***
  3185 
  3186 * Locale interpretation propagates mixins along the locale hierarchy.
  3187 The currently only available mixins are the equations used to map
  3188 local definitions to terms of the target domain of an interpretation.
  3189 
  3190 * Reactivated diagnostic command 'print_interps'.  Use "print_interps
  3191 loc" to print all interpretations of locale "loc" in the theory.
  3192 Interpretations in proofs are not shown.
  3193 
  3194 * Thoroughly revised locales tutorial.  New section on conditional
  3195 interpretation.
  3196 
  3197 * On instantiation of classes, remaining undefined class parameters
  3198 are formally declared.  INCOMPATIBILITY.
  3199 
  3200 
  3201 *** Document preparation ***
  3202 
  3203 * New generalized style concept for printing terms: @{foo (style) ...}
  3204 instead of @{foo_style style ...}  (old form is still retained for
  3205 backward compatibility).  Styles can be also applied for
  3206 antiquotations prop, term_type and typeof.
  3207 
  3208 
  3209 *** HOL ***
  3210 
  3211 * New proof method "smt" for a combination of first-order logic with
  3212 equality, linear and nonlinear (natural/integer/real) arithmetic, and
  3213 fixed-size bitvectors; there is also basic support for higher-order
  3214 features (esp. lambda abstractions).  It is an incomplete decision
  3215 procedure based on external SMT solvers using the oracle mechanism;
  3216 for the SMT solver Z3, this method is proof-producing.  Certificates
  3217 are provided to avoid calling the external solvers solely for
  3218 re-checking proofs.  Due to a remote SMT service there is no need for
  3219 installing SMT solvers locally.  See src/HOL/SMT.
  3220 
  3221 * New commands to load and prove verification conditions generated by
  3222 the Boogie program verifier or derived systems (e.g. the Verifying C
  3223 Compiler (VCC) or Spec#).  See src/HOL/Boogie.
  3224 
  3225 * New counterexample generator tool 'nitpick' based on the Kodkod
  3226 relational model finder.  See src/HOL/Tools/Nitpick and
  3227 src/HOL/Nitpick_Examples.
  3228 
  3229 * New commands 'code_pred' and 'values' to invoke the predicate
  3230 compiler and to enumerate values of inductive predicates.
  3231 
  3232 * A tabled implementation of the reflexive transitive closure.
  3233 
  3234 * New implementation of quickcheck uses generic code generator;
  3235 default generators are provided for all suitable HOL types, records
  3236 and datatypes.  Old quickcheck can be re-activated importing theory
  3237 Library/SML_Quickcheck.
  3238 
  3239 * New testing tool Mirabelle for automated proof tools.  Applies
  3240 several tools and tactics like sledgehammer, metis, or quickcheck, to
  3241 every proof step in a theory.  To be used in batch mode via the
  3242 "mirabelle" utility.
  3243 
  3244 * New proof method "sos" (sum of squares) for nonlinear real
  3245 arithmetic (originally due to John Harison). It requires theory
  3246 Library/Sum_Of_Squares.  It is not a complete decision procedure but
  3247 works well in practice on quantifier-free real arithmetic with +, -,
  3248 *, ^, =, <= and <, i.e. boolean combinations of equalities and
  3249 inequalities between polynomials.  It makes use of external
  3250 semidefinite programming solvers.  Method "sos" generates a
  3251 certificate that can be pasted into the proof thus avoiding the need
  3252 to call an external tool every time the proof is checked.  See
  3253 src/HOL/Library/Sum_Of_Squares.
  3254 
  3255 * New method "linarith" invokes existing linear arithmetic decision
  3256 procedure only.
  3257 
  3258 * New command 'atp_minimal' reduces result produced by Sledgehammer.
  3259 
  3260 * New Sledgehammer option "Full Types" in Proof General settings menu.
  3261 Causes full type information to be output to the ATPs.  This slows
  3262 ATPs down considerably but eliminates a source of unsound "proofs"
  3263 that fail later.
  3264 
  3265 * New method "metisFT": A version of metis that uses full type
  3266 information in order to avoid failures of proof reconstruction.
  3267 
  3268 * New evaluator "approximate" approximates an real valued term using
  3269 the same method as the approximation method.
  3270 
  3271 * Method "approximate" now supports arithmetic expressions as
  3272 boundaries of intervals and implements interval splitting and Taylor
  3273 series expansion.
  3274 
  3275 * ML antiquotation @{code_datatype} inserts definition of a datatype
  3276 generated by the code generator; e.g. see src/HOL/Predicate.thy.
  3277 
  3278 * New theory SupInf of the supremum and infimum operators for sets of
  3279 reals.
  3280 
  3281 * New theory Probability, which contains a development of measure
  3282 theory, eventually leading to Lebesgue integration and probability.
  3283 
  3284 * Extended Multivariate Analysis to include derivation and Brouwer's
  3285 fixpoint theorem.
  3286 
  3287 * Reorganization of number theory, INCOMPATIBILITY:
  3288   - new number theory development for nat and int, in theories Divides
  3289     and GCD as well as in new session Number_Theory
  3290   - some constants and facts now suffixed with _nat and _int
  3291     accordingly
  3292   - former session NumberTheory now named Old_Number_Theory, including
  3293     theories Legacy_GCD and Primes (prefer Number_Theory if possible)
  3294   - moved theory Pocklington from src/HOL/Library to
  3295     src/HOL/Old_Number_Theory
  3296 
  3297 * Theory GCD includes functions Gcd/GCD and Lcm/LCM for the gcd and
  3298 lcm of finite and infinite sets. It is shown that they form a complete
  3299 lattice.
  3300 
  3301 * Class semiring_div requires superclass no_zero_divisors and proof of
  3302 div_mult_mult1; theorems div_mult_mult1, div_mult_mult2,
  3303 div_mult_mult1_if, div_mult_mult1 and div_mult_mult2 have been
  3304 generalized to class semiring_div, subsuming former theorems
  3305 zdiv_zmult_zmult1, zdiv_zmult_zmult1_if, zdiv_zmult_zmult1 and
  3306 zdiv_zmult_zmult2.  div_mult_mult1 is now [simp] by default.
  3307 INCOMPATIBILITY.
  3308 
  3309 * Refinements to lattice classes and sets:
  3310   - less default intro/elim rules in locale variant, more default
  3311     intro/elim rules in class variant: more uniformity
  3312   - lemma ge_sup_conv renamed to le_sup_iff, in accordance with
  3313     le_inf_iff
  3314   - dropped lemma alias inf_ACI for inf_aci (same for sup_ACI and
  3315     sup_aci)
  3316   - renamed ACI to inf_sup_aci
  3317   - new class "boolean_algebra"
  3318   - class "complete_lattice" moved to separate theory
  3319     "Complete_Lattice"; corresponding constants (and abbreviations)
  3320     renamed and with authentic syntax:
  3321     Set.Inf ~>    Complete_Lattice.Inf
  3322     Set.Sup ~>    Complete_Lattice.Sup
  3323     Set.INFI ~>   Complete_Lattice.INFI
  3324     Set.SUPR ~>   Complete_Lattice.SUPR
  3325     Set.Inter ~>  Complete_Lattice.Inter
  3326     Set.Union ~>  Complete_Lattice.Union
  3327     Set.INTER ~>  Complete_Lattice.INTER
  3328     Set.UNION ~>  Complete_Lattice.UNION
  3329   - authentic syntax for
  3330     Set.Pow
  3331     Set.image
  3332   - mere abbreviations:
  3333     Set.empty               (for bot)
  3334     Set.UNIV                (for top)
  3335     Set.inter               (for inf, formerly Set.Int)
  3336     Set.union               (for sup, formerly Set.Un)
  3337     Complete_Lattice.Inter  (for Inf)
  3338     Complete_Lattice.Union  (for Sup)
  3339     Complete_Lattice.INTER  (for INFI)
  3340     Complete_Lattice.UNION  (for SUPR)
  3341   - object-logic definitions as far as appropriate
  3342 
  3343 INCOMPATIBILITY.  Care is required when theorems Int_subset_iff or
  3344 Un_subset_iff are explicitly deleted as default simp rules; then also
  3345 their lattice counterparts le_inf_iff and le_sup_iff have to be
  3346 deleted to achieve the desired effect.
  3347 
  3348 * Rules inf_absorb1, inf_absorb2, sup_absorb1, sup_absorb2 are no simp
  3349 rules by default any longer; the same applies to min_max.inf_absorb1
  3350 etc.  INCOMPATIBILITY.
  3351 
  3352 * Rules sup_Int_eq and sup_Un_eq are no longer declared as
  3353 pred_set_conv by default.  INCOMPATIBILITY.
  3354 
  3355 * Power operations on relations and functions are now one dedicated
  3356 constant "compow" with infix syntax "^^".  Power operation on
  3357 multiplicative monoids retains syntax "^" and is now defined generic
  3358 in class power.  INCOMPATIBILITY.
  3359 
  3360 * Relation composition "R O S" now has a more standard argument order:
  3361 "R O S = {(x, z). EX y. (x, y) : R & (y, z) : S}".  INCOMPATIBILITY,
  3362 rewrite propositions with "S O R" --> "R O S". Proofs may occasionally
  3363 break, since the O_assoc rule was not rewritten like this.  Fix using
  3364 O_assoc[symmetric].  The same applies to the curried version "R OO S".
  3365 
  3366 * Function "Inv" is renamed to "inv_into" and function "inv" is now an
  3367 abbreviation for "inv_into UNIV".  Lemmas are renamed accordingly.
  3368 INCOMPATIBILITY.
  3369 
  3370 * Most rules produced by inductive and datatype package have mandatory
  3371 prefixes.  INCOMPATIBILITY.
  3372 
  3373 * Changed "DERIV_intros" to a dynamic fact, which can be augmented by
  3374 the attribute of the same name.  Each of the theorems in the list
  3375 DERIV_intros assumes composition with an additional function and
  3376 matches a variable to the derivative, which has to be solved by the
  3377 Simplifier.  Hence (auto intro!: DERIV_intros) computes the derivative
  3378 of most elementary terms.  Former Maclauren.DERIV_tac and
  3379 Maclauren.deriv_tac should be replaced by (auto intro!: DERIV_intros).
  3380 INCOMPATIBILITY.
  3381 
  3382 * Code generator attributes follow the usual underscore convention:
  3383     code_unfold     replaces    code unfold
  3384     code_post       replaces    code post
  3385     etc.
  3386   INCOMPATIBILITY.
  3387 
  3388 * Renamed methods:
  3389     sizechange -> size_change
  3390     induct_scheme -> induction_schema
  3391   INCOMPATIBILITY.
  3392 
  3393 * Discontinued abbreviation "arbitrary" of constant "undefined".
  3394 INCOMPATIBILITY, use "undefined" directly.
  3395 
  3396 * Renamed theorems:
  3397     Suc_eq_add_numeral_1 -> Suc_eq_plus1
  3398     Suc_eq_add_numeral_1_left -> Suc_eq_plus1_left
  3399     Suc_plus1 -> Suc_eq_plus1
  3400     *anti_sym -> *antisym*
  3401     vector_less_eq_def -> vector_le_def
  3402   INCOMPATIBILITY.
  3403 
  3404 * Added theorem List.map_map as [simp].  Removed List.map_compose.
  3405 INCOMPATIBILITY.
  3406 
  3407 * Removed predicate "M hassize n" (<--> card M = n & finite M).
  3408 INCOMPATIBILITY.
  3409 
  3410 
  3411 *** HOLCF ***
  3412 
  3413 * Theory Representable defines a class "rep" of domains that are
  3414 representable (via an ep-pair) in the universal domain type "udom".
  3415 Instances are provided for all type constructors defined in HOLCF.
  3416 
  3417 * The 'new_domain' command is a purely definitional version of the
  3418 domain package, for representable domains.  Syntax is identical to the
  3419 old domain package.  The 'new_domain' package also supports indirect
  3420 recursion using previously-defined type constructors.  See
  3421 src/HOLCF/ex/New_Domain.thy for examples.
  3422 
  3423 * Method "fixrec_simp" unfolds one step of a fixrec-defined constant
  3424 on the left-hand side of an equation, and then performs
  3425 simplification.  Rewriting is done using rules declared with the
  3426 "fixrec_simp" attribute.  The "fixrec_simp" method is intended as a
  3427 replacement for "fixpat"; see src/HOLCF/ex/Fixrec_ex.thy for examples.
  3428 
  3429 * The pattern-match compiler in 'fixrec' can now handle constructors
  3430 with HOL function types.  Pattern-match combinators for the Pair
  3431 constructor are pre-configured.
  3432 
  3433 * The 'fixrec' package now produces better fixed-point induction rules
  3434 for mutually-recursive definitions:  Induction rules have conclusions
  3435 of the form "P foo bar" instead of "P <foo, bar>".
  3436 
  3437 * The constant "sq_le" (with infix syntax "<<" or "\<sqsubseteq>") has
  3438 been renamed to "below".  The name "below" now replaces "less" in many
  3439 theorem names.  (Legacy theorem names using "less" are still supported
  3440 as well.)
  3441 
  3442 * The 'fixrec' package now supports "bottom patterns".  Bottom
  3443 patterns can be used to generate strictness rules, or to make
  3444 functions more strict (much like the bang-patterns supported by the
  3445 Glasgow Haskell Compiler).  See src/HOLCF/ex/Fixrec_ex.thy for
  3446 examples.
  3447 
  3448 
  3449 *** ML ***
  3450 
  3451 * Support for Poly/ML 5.3.0, with improved reporting of compiler
  3452 errors and run-time exceptions, including detailed source positions.
  3453 
  3454 * Structure Name_Space (formerly NameSpace) now manages uniquely
  3455 identified entries, with some additional information such as source
  3456 position, logical grouping etc.
  3457 
  3458 * Theory and context data is now introduced by the simplified and
  3459 modernized functors Theory_Data, Proof_Data, Generic_Data.  Data needs
  3460 to be pure, but the old TheoryDataFun for mutable data (with explicit
  3461 copy operation) is still available for some time.
  3462 
  3463 * Structure Synchronized (cf. src/Pure/Concurrent/synchronized.ML)
  3464 provides a high-level programming interface to synchronized state
  3465 variables with atomic update.  This works via pure function
  3466 application within a critical section -- its runtime should be as
  3467 short as possible; beware of deadlocks if critical code is nested,
  3468 either directly or indirectly via other synchronized variables!
  3469 
  3470 * Structure Unsynchronized (cf. src/Pure/ML-Systems/unsynchronized.ML)
  3471 wraps raw ML references, explicitly indicating their non-thread-safe
  3472 behaviour.  The Isar toplevel keeps this structure open, to
  3473 accommodate Proof General as well as quick and dirty interactive
  3474 experiments with references.
  3475 
  3476 * PARALLEL_CHOICE and PARALLEL_GOALS provide basic support for
  3477 parallel tactical reasoning.
  3478 
  3479 * Tacticals Subgoal.FOCUS, Subgoal.FOCUS_PREMS, Subgoal.FOCUS_PARAMS
  3480 are similar to SUBPROOF, but are slightly more flexible: only the
  3481 specified parts of the subgoal are imported into the context, and the
  3482 body tactic may introduce new subgoals and schematic variables.
  3483 
  3484 * Old tactical METAHYPS, which does not observe the proof context, has
  3485 been renamed to Old_Goals.METAHYPS and awaits deletion.  Use SUBPROOF
  3486 or Subgoal.FOCUS etc.
  3487 
  3488 * Renamed functor TableFun to Table, and GraphFun to Graph.  (Since
  3489 functors have their own ML name space there is no point to mark them
  3490 separately.)  Minor INCOMPATIBILITY.
  3491 
  3492 * Renamed NamedThmsFun to Named_Thms.  INCOMPATIBILITY.
  3493 
  3494 * Renamed several structures FooBar to Foo_Bar.  Occasional,
  3495 INCOMPATIBILITY.
  3496 
  3497 * Operations of structure Skip_Proof no longer require quick_and_dirty
  3498 mode, which avoids critical setmp.
  3499 
  3500 * Eliminated old Attrib.add_attributes, Method.add_methods and related
  3501 combinators for "args".  INCOMPATIBILITY, need to use simplified
  3502 Attrib/Method.setup introduced in Isabelle2009.
  3503 
  3504 * Proper context for simpset_of, claset_of, clasimpset_of.  May fall
  3505 back on global_simpset_of, global_claset_of, global_clasimpset_of as
  3506 last resort.  INCOMPATIBILITY.
  3507 
  3508 * Display.pretty_thm now requires a proper context (cf. former
  3509 ProofContext.pretty_thm).  May fall back on Display.pretty_thm_global
  3510 or even Display.pretty_thm_without_context as last resort.
  3511 INCOMPATIBILITY.
  3512 
  3513 * Discontinued Display.pretty_ctyp/cterm etc.  INCOMPATIBILITY, use
  3514 Syntax.pretty_typ/term directly, preferably with proper context
  3515 instead of global theory.
  3516 
  3517 
  3518 *** System ***
  3519 
  3520 * Further fine tuning of parallel proof checking, scales up to 8 cores
  3521 (max. speedup factor 5.0).  See also Goal.parallel_proofs in ML and
  3522 usedir option -q.
  3523 
  3524 * Support for additional "Isabelle components" via etc/components, see
  3525 also the system manual.
  3526 
  3527 * The isabelle makeall tool now operates on all components with
  3528 IsaMakefile, not just hardwired "logics".
  3529 
  3530 * Removed "compress" option from isabelle-process and isabelle usedir;
  3531 this is always enabled.
  3532 
  3533 * Discontinued support for Poly/ML 4.x versions.
  3534 
  3535 * Isabelle tool "wwwfind" provides web interface for 'find_theorems'
  3536 on a given logic image.  This requires the lighttpd webserver and is
  3537 currently supported on Linux only.
  3538 
  3539 
  3540 
  3541 New in Isabelle2009 (April 2009)
  3542 --------------------------------
  3543 
  3544 *** General ***
  3545 
  3546 * Simplified main Isabelle executables, with less surprises on
  3547 case-insensitive file-systems (such as Mac OS).
  3548 
  3549   - The main Isabelle tool wrapper is now called "isabelle" instead of
  3550     "isatool."
  3551 
  3552   - The former "isabelle" alias for "isabelle-process" has been
  3553     removed (should rarely occur to regular users).
  3554 
  3555   - The former "isabelle-interface" and its alias "Isabelle" have been
  3556     removed (interfaces are now regular Isabelle tools).
  3557 
  3558 Within scripts and make files, the Isabelle environment variables
  3559 ISABELLE_TOOL and ISABELLE_PROCESS replace old ISATOOL and ISABELLE,
  3560 respectively.  (The latter are still available as legacy feature.)
  3561 
  3562 The old isabelle-interface wrapper could react in confusing ways if
  3563 the interface was uninstalled or changed otherwise.  Individual
  3564 interface tool configuration is now more explicit, see also the
  3565 Isabelle system manual.  In particular, Proof General is now available
  3566 via "isabelle emacs".
  3567 
  3568 INCOMPATIBILITY, need to adapt derivative scripts.  Users may need to
  3569 purge installed copies of Isabelle executables and re-run "isabelle
  3570 install -p ...", or use symlinks.
  3571 
  3572 * The default for ISABELLE_HOME_USER is now ~/.isabelle instead of the
  3573 old ~/isabelle, which was slightly non-standard and apt to cause
  3574 surprises on case-insensitive file-systems (such as Mac OS).
  3575 
  3576 INCOMPATIBILITY, need to move existing ~/isabelle/etc,
  3577 ~/isabelle/heaps, ~/isabelle/browser_info to the new place.  Special
  3578 care is required when using older releases of Isabelle.  Note that
  3579 ISABELLE_HOME_USER can be changed in Isabelle/etc/settings of any
  3580 Isabelle distribution, in order to use the new ~/.isabelle uniformly.
  3581 
  3582 * Proofs of fully specified statements are run in parallel on
  3583 multi-core systems.  A speedup factor of 2.5 to 3.2 can be expected on
  3584 a regular 4-core machine, if the initial heap space is made reasonably
  3585 large (cf. Poly/ML option -H).  (Requires Poly/ML 5.2.1 or later.)
  3586 
  3587 * The main reference manuals ("isar-ref", "implementation", and
  3588 "system") have been updated and extended.  Formally checked references
  3589 as hyperlinks are now available uniformly.
  3590 
  3591 
  3592 *** Pure ***
  3593 
  3594 * Complete re-implementation of locales.  INCOMPATIBILITY in several
  3595 respects.  The most important changes are listed below.  See the
  3596 Tutorial on Locales ("locales" manual) for details.
  3597 
  3598 - In locale expressions, instantiation replaces renaming.  Parameters
  3599 must be declared in a for clause.  To aid compatibility with previous
  3600 parameter inheritance, in locale declarations, parameters that are not
  3601 'touched' (instantiation position "_" or omitted) are implicitly added
  3602 with their syntax at the beginning of the for clause.
  3603 
  3604 - Syntax from abbreviations and definitions in locales is available in
  3605 locale expressions and context elements.  The latter is particularly
  3606 useful in locale declarations.
  3607 
  3608 - More flexible mechanisms to qualify names generated by locale
  3609 expressions.  Qualifiers (prefixes) may be specified in locale
  3610 expressions, and can be marked as mandatory (syntax: "name!:") or
  3611 optional (syntax "name?:").  The default depends for plain "name:"
  3612 depends on the situation where a locale expression is used: in
  3613 commands 'locale' and 'sublocale' prefixes are optional, in
  3614 'interpretation' and 'interpret' prefixes are mandatory.  The old
  3615 implicit qualifiers derived from the parameter names of a locale are
  3616 no longer generated.
  3617 
  3618 - Command "sublocale l < e" replaces "interpretation l < e".  The
  3619 instantiation clause in "interpretation" and "interpret" (square
  3620 brackets) is no longer available.  Use locale expressions.
  3621 
  3622 - When converting proof scripts, mandatory qualifiers in
  3623 'interpretation' and 'interpret' should be retained by default, even
  3624 if this is an INCOMPATIBILITY compared to former behavior.  In the
  3625 worst case, use the "name?:" form for non-mandatory ones.  Qualifiers
  3626 in locale expressions range over a single locale instance only.
  3627 
  3628 - Dropped locale element "includes".  This is a major INCOMPATIBILITY.
  3629 In existing theorem specifications replace the includes element by the
  3630 respective context elements of the included locale, omitting those
  3631 that are already present in the theorem specification.  Multiple
  3632 assume elements of a locale should be replaced by a single one
  3633 involving the locale predicate.  In the proof body, declarations (most
  3634 notably theorems) may be regained by interpreting the respective
  3635 locales in the proof context as required (command "interpret").
  3636 
  3637 If using "includes" in replacement of a target solely because the
  3638 parameter types in the theorem are not as general as in the target,
  3639 consider declaring a new locale with additional type constraints on
  3640 the parameters (context element "constrains").
  3641 
  3642 - Discontinued "locale (open)".  INCOMPATIBILITY.
  3643 
  3644 - Locale interpretation commands no longer attempt to simplify goal.
  3645 INCOMPATIBILITY: in rare situations the generated goal differs.  Use
  3646 methods intro_locales and unfold_locales to clarify.
  3647 
  3648 - Locale interpretation commands no longer accept interpretation
  3649 attributes.  INCOMPATIBILITY.
  3650 
  3651 * Class declaration: so-called "base sort" must not be given in import
  3652 list any longer, but is inferred from the specification.  Particularly
  3653 in HOL, write
  3654 
  3655     class foo = ...
  3656 
  3657 instead of
  3658 
  3659     class foo = type + ...
  3660 
  3661 * Class target: global versions of theorems stemming do not carry a
  3662 parameter prefix any longer.  INCOMPATIBILITY.
  3663 
  3664 * Class 'instance' command no longer accepts attached definitions.
  3665 INCOMPATIBILITY, use proper 'instantiation' target instead.
  3666 
  3667 * Recovered hiding of consts, which was accidentally broken in
  3668 Isabelle2007.  Potential INCOMPATIBILITY, ``hide const c'' really
  3669 makes c inaccessible; consider using ``hide (open) const c'' instead.
  3670 
  3671 * Slightly more coherent Pure syntax, with updated documentation in
  3672 isar-ref manual.  Removed locales meta_term_syntax and
  3673 meta_conjunction_syntax: TERM and &&& (formerly &&) are now permanent,
  3674 INCOMPATIBILITY in rare situations.  Note that &&& should not be used
  3675 directly in regular applications.
  3676 
  3677 * There is a new syntactic category "float_const" for signed decimal
  3678 fractions (e.g. 123.45 or -123.45).
  3679 
  3680 * Removed exotic 'token_translation' command.  INCOMPATIBILITY, use ML
  3681 interface with 'setup' command instead.
  3682 
  3683 * Command 'local_setup' is similar to 'setup', but operates on a local
  3684 theory context.
  3685 
  3686 * The 'axiomatization' command now only works within a global theory
  3687 context.  INCOMPATIBILITY.
  3688 
  3689 * Goal-directed proof now enforces strict proof irrelevance wrt. sort
  3690 hypotheses.  Sorts required in the course of reasoning need to be
  3691 covered by the constraints in the initial statement, completed by the
  3692 type instance information of the background theory.  Non-trivial sort
  3693 hypotheses, which rarely occur in practice, may be specified via
  3694 vacuous propositions of the form SORT_CONSTRAINT('a::c).  For example:
  3695 
  3696   lemma assumes "SORT_CONSTRAINT('a::empty)" shows False ...
  3697 
  3698 The result contains an implicit sort hypotheses as before --
  3699 SORT_CONSTRAINT premises are eliminated as part of the canonical rule
  3700 normalization.
  3701 
  3702 * Generalized Isar history, with support for linear undo, direct state
  3703 addressing etc.
  3704 
  3705 * Changed defaults for unify configuration options:
  3706 
  3707   unify_trace_bound = 50 (formerly 25)
  3708   unify_search_bound = 60 (formerly 30)
  3709 
  3710 * Different bookkeeping for code equations (INCOMPATIBILITY):
  3711 
  3712   a) On theory merge, the last set of code equations for a particular
  3713      constant is taken (in accordance with the policy applied by other
  3714      parts of the code generator framework).
  3715 
  3716   b) Code equations stemming from explicit declarations (e.g. code
  3717      attribute) gain priority over default code equations stemming
  3718      from definition, primrec, fun etc.
  3719 
  3720 * Keyword 'code_exception' now named 'code_abort'.  INCOMPATIBILITY.
  3721 
  3722 * Unified theorem tables for both code generators.  Thus [code
  3723 func] has disappeared and only [code] remains.  INCOMPATIBILITY.
  3724 
  3725 * Command 'find_consts' searches for constants based on type and name
  3726 patterns, e.g.
  3727 
  3728     find_consts "_ => bool"
  3729 
  3730 By default, matching is against subtypes, but it may be restricted to
  3731 the whole type.  Searching by name is possible.  Multiple queries are
  3732 conjunctive and queries may be negated by prefixing them with a
  3733 hyphen:
  3734 
  3735     find_consts strict: "_ => bool" name: "Int" -"int => int"
  3736 
  3737 * New 'find_theorems' criterion "solves" matches theorems that
  3738 directly solve the current goal (modulo higher-order unification).
  3739 
  3740 * Auto solve feature for main theorem statements: whenever a new goal
  3741 is stated, "find_theorems solves" is called; any theorems that could
  3742 solve the lemma directly are listed as part of the goal state.
  3743 Cf. associated options in Proof General Isabelle settings menu,
  3744 enabled by default, with reasonable timeout for pathological cases of
  3745 higher-order unification.
  3746 
  3747 
  3748 *** Document preparation ***
  3749 
  3750 * Antiquotation @{lemma} now imitates a regular terminal proof,
  3751 demanding keyword 'by' and supporting the full method expression
  3752 syntax just like the Isar command 'by'.
  3753 
  3754 
  3755 *** HOL ***
  3756 
  3757 * Integrated main parts of former image HOL-Complex with HOL.  Entry
  3758 points Main and Complex_Main remain as before.
  3759 
  3760 * Logic image HOL-Plain provides a minimal HOL with the most important
  3761 tools available (inductive, datatype, primrec, ...).  This facilitates
  3762 experimentation and tool development.  Note that user applications
  3763 (and library theories) should never refer to anything below theory
  3764 Main, as before.
  3765 
  3766 * Logic image HOL-Main stops at theory Main, and thus facilitates
  3767 experimentation due to shorter build times.
  3768 
  3769 * Logic image HOL-NSA contains theories of nonstandard analysis which
  3770 were previously part of former HOL-Complex.  Entry point Hyperreal
  3771 remains valid, but theories formerly using Complex_Main should now use
  3772 new entry point Hypercomplex.
  3773 
  3774 * Generic ATP manager for Sledgehammer, based on ML threads instead of
  3775 Posix processes.  Avoids potentially expensive forking of the ML
  3776 process.  New thread-based implementation also works on non-Unix
  3777 platforms (Cygwin).  Provers are no longer hardwired, but defined
  3778 within the theory via plain ML wrapper functions.  Basic Sledgehammer
  3779 commands are covered in the isar-ref manual.
  3780 
  3781 * Wrapper scripts for remote SystemOnTPTP service allows to use
  3782 sledgehammer without local ATP installation (Vampire etc.). Other
  3783 provers may be included via suitable ML wrappers, see also
  3784 src/HOL/ATP_Linkup.thy.
  3785 
  3786 * ATP selection (E/Vampire/Spass) is now via Proof General's settings
  3787 menu.
  3788 
  3789 * The metis method no longer fails because the theorem is too trivial
  3790 (contains the empty clause).
  3791 
  3792 * The metis method now fails in the usual manner, rather than raising
  3793 an exception, if it determines that it cannot prove the theorem.
  3794 
  3795 * Method "coherent" implements a prover for coherent logic (see also
  3796 src/Tools/coherent.ML).
  3797 
  3798 * Constants "undefined" and "default" replace "arbitrary".  Usually
  3799 "undefined" is the right choice to replace "arbitrary", though
  3800 logically there is no difference.  INCOMPATIBILITY.
  3801 
  3802 * Command "value" now integrates different evaluation mechanisms.  The
  3803 result of the first successful evaluation mechanism is printed.  In
  3804 square brackets a particular named evaluation mechanisms may be
  3805 specified (currently, [SML], [code] or [nbe]).  See further
  3806 src/HOL/ex/Eval_Examples.thy.
  3807 
  3808 * Normalization by evaluation now allows non-leftlinear equations.
  3809 Declare with attribute [code nbe].
  3810 
  3811 * Methods "case_tac" and "induct_tac" now refer to the very same rules
  3812 as the structured Isar versions "cases" and "induct", cf. the
  3813 corresponding "cases" and "induct" attributes.  Mutual induction rules
  3814 are now presented as a list of individual projections
  3815 (e.g. foo_bar.inducts for types foo and bar); the old format with
  3816 explicit HOL conjunction is no longer supported.  INCOMPATIBILITY, in
  3817 rare situations a different rule is selected --- notably nested tuple
  3818 elimination instead of former prod.exhaust: use explicit (case_tac t
  3819 rule: prod.exhaust) here.
  3820 
  3821 * Attributes "cases", "induct", "coinduct" support "del" option.
  3822 
  3823 * Removed fact "case_split_thm", which duplicates "case_split".
  3824 
  3825 * The option datatype has been moved to a new theory Option.  Renamed
  3826 option_map to Option.map, and o2s to Option.set, INCOMPATIBILITY.
  3827 
  3828 * New predicate "strict_mono" classifies strict functions on partial
  3829 orders.  With strict functions on linear orders, reasoning about
  3830 (in)equalities is facilitated by theorems "strict_mono_eq",
  3831 "strict_mono_less_eq" and "strict_mono_less".
  3832 
  3833 * Some set operations are now proper qualified constants with
  3834 authentic syntax.  INCOMPATIBILITY:
  3835 
  3836     op Int ~>   Set.Int
  3837     op Un ~>    Set.Un
  3838     INTER ~>    Set.INTER
  3839     UNION ~>    Set.UNION
  3840     Inter ~>    Set.Inter
  3841     Union ~>    Set.Union
  3842     {} ~>       Set.empty
  3843     UNIV ~>     Set.UNIV
  3844 
  3845 * Class complete_lattice with operations Inf, Sup, INFI, SUPR now in
  3846 theory Set.
  3847 
  3848 * Auxiliary class "itself" has disappeared -- classes without any
  3849 parameter are treated as expected by the 'class' command.
  3850 
  3851 * Leibnitz's Series for Pi and the arcus tangens and logarithm series.
  3852 
  3853 * Common decision procedures (Cooper, MIR, Ferrack, Approximation,
  3854 Dense_Linear_Order) are now in directory HOL/Decision_Procs.
  3855 
  3856 * Theory src/HOL/Decision_Procs/Approximation provides the new proof
  3857 method "approximation".  It proves formulas on real values by using
  3858 interval arithmetic.  In the formulas are also the transcendental
  3859 functions sin, cos, tan, atan, ln, exp and the constant pi are
  3860 allowed. For examples see
  3861 src/HOL/Descision_Procs/ex/Approximation_Ex.thy.
  3862 
  3863 * Theory "Reflection" now resides in HOL/Library.
  3864 
  3865 * Entry point to Word library now simply named "Word".
  3866 INCOMPATIBILITY.
  3867 
  3868 * Made source layout more coherent with logical distribution
  3869 structure:
  3870 
  3871     src/HOL/Library/RType.thy ~> src/HOL/Typerep.thy
  3872     src/HOL/Library/Code_Message.thy ~> src/HOL/
  3873     src/HOL/Library/GCD.thy ~> src/HOL/
  3874     src/HOL/Library/Order_Relation.thy ~> src/HOL/
  3875     src/HOL/Library/Parity.thy ~> src/HOL/
  3876     src/HOL/Library/Univ_Poly.thy ~> src/HOL/
  3877     src/HOL/Real/ContNotDenum.thy ~> src/HOL/Library/
  3878     src/HOL/Real/Lubs.thy ~> src/HOL/
  3879     src/HOL/Real/PReal.thy ~> src/HOL/
  3880     src/HOL/Real/Rational.thy ~> src/HOL/
  3881     src/HOL/Real/RComplete.thy ~> src/HOL/
  3882     src/HOL/Real/RealDef.thy ~> src/HOL/
  3883     src/HOL/Real/RealPow.thy ~> src/HOL/
  3884     src/HOL/Real/Real.thy ~> src/HOL/
  3885     src/HOL/Complex/Complex_Main.thy ~> src/HOL/
  3886     src/HOL/Complex/Complex.thy ~> src/HOL/
  3887     src/HOL/Complex/FrechetDeriv.thy ~> src/HOL/Library/
  3888     src/HOL/Complex/Fundamental_Theorem_Algebra.thy ~> src/HOL/Library/
  3889     src/HOL/Hyperreal/Deriv.thy ~> src/HOL/
  3890     src/HOL/Hyperreal/Fact.thy ~> src/HOL/
  3891     src/HOL/Hyperreal/Integration.thy ~> src/HOL/
  3892     src/HOL/Hyperreal/Lim.thy ~> src/HOL/
  3893     src/HOL/Hyperreal/Ln.thy ~> src/HOL/
  3894     src/HOL/Hyperreal/Log.thy ~> src/HOL/
  3895     src/HOL/Hyperreal/MacLaurin.thy ~> src/HOL/
  3896     src/HOL/Hyperreal/NthRoot.thy ~> src/HOL/
  3897     src/HOL/Hyperreal/Series.thy ~> src/HOL/
  3898     src/HOL/Hyperreal/SEQ.thy ~> src/HOL/
  3899     src/HOL/Hyperreal/Taylor.thy ~> src/HOL/
  3900     src/HOL/Hyperreal/Transcendental.thy ~> src/HOL/
  3901     src/HOL/Real/Float ~> src/HOL/Library/
  3902     src/HOL/Real/HahnBanach ~> src/HOL/HahnBanach
  3903     src/HOL/Real/RealVector.thy ~> src/HOL/
  3904 
  3905     src/HOL/arith_data.ML ~> src/HOL/Tools
  3906     src/HOL/hologic.ML ~> src/HOL/Tools
  3907     src/HOL/simpdata.ML ~> src/HOL/Tools
  3908     src/HOL/int_arith1.ML ~> src/HOL/Tools/int_arith.ML
  3909     src/HOL/int_factor_simprocs.ML ~> src/HOL/Tools
  3910     src/HOL/nat_simprocs.ML ~> src/HOL/Tools
  3911     src/HOL/Real/float_arith.ML ~> src/HOL/Tools
  3912     src/HOL/Real/float_syntax.ML ~> src/HOL/Tools
  3913     src/HOL/Real/rat_arith.ML ~> src/HOL/Tools
  3914     src/HOL/Real/real_arith.ML ~> src/HOL/Tools
  3915 
  3916     src/HOL/Library/Array.thy ~> src/HOL/Imperative_HOL
  3917     src/HOL/Library/Heap_Monad.thy ~> src/HOL/Imperative_HOL
  3918     src/HOL/Library/Heap.thy ~> src/HOL/Imperative_HOL
  3919     src/HOL/Library/Imperative_HOL.thy ~> src/HOL/Imperative_HOL
  3920     src/HOL/Library/Ref.thy ~> src/HOL/Imperative_HOL
  3921     src/HOL/Library/Relational.thy ~> src/HOL/Imperative_HOL
  3922 
  3923 * If methods "eval" and "evaluation" encounter a structured proof
  3924 state with !!/==>, only the conclusion is evaluated to True (if
  3925 possible), avoiding strange error messages.
  3926 
  3927 * Method "sizechange" automates termination proofs using (a
  3928 modification of) the size-change principle.  Requires SAT solver.  See
  3929 src/HOL/ex/Termination.thy for examples.
  3930 
  3931 * Simplifier: simproc for let expressions now unfolds if bound
  3932 variable occurs at most once in let expression body.  INCOMPATIBILITY.
  3933 
  3934 * Method "arith": Linear arithmetic now ignores all inequalities when
  3935 fast_arith_neq_limit is exceeded, instead of giving up entirely.
  3936 
  3937 * New attribute "arith" for facts that should always be used
  3938 automatically by arithmetic. It is intended to be used locally in
  3939 proofs, e.g.
  3940 
  3941   assumes [arith]: "x > 0"
  3942 
  3943 Global usage is discouraged because of possible performance impact.
  3944 
  3945 * New classes "top" and "bot" with corresponding operations "top" and
  3946 "bot" in theory Orderings; instantiation of class "complete_lattice"
  3947 requires instantiation of classes "top" and "bot".  INCOMPATIBILITY.
  3948 
  3949 * Changed definition lemma "less_fun_def" in order to provide an
  3950 instance for preorders on functions; use lemma "less_le" instead.
  3951 INCOMPATIBILITY.
  3952 
  3953 * Theory Orderings: class "wellorder" moved here, with explicit
  3954 induction rule "less_induct" as assumption.  For instantiation of
  3955 "wellorder" by means of predicate "wf", use rule wf_wellorderI.
  3956 INCOMPATIBILITY.
  3957 
  3958 * Theory Orderings: added class "preorder" as superclass of "order".
  3959 INCOMPATIBILITY: Instantiation proofs for order, linorder
  3960 etc. slightly changed.  Some theorems named order_class.* now named
  3961 preorder_class.*.
  3962 
  3963 * Theory Relation: renamed "refl" to "refl_on", "reflexive" to "refl,
  3964 "diag" to "Id_on".
  3965 
  3966 * Theory Finite_Set: added a new fold combinator of type
  3967 
  3968   ('a => 'b => 'b) => 'b => 'a set => 'b
  3969 
  3970 Occasionally this is more convenient than the old fold combinator
  3971 which is now defined in terms of the new one and renamed to
  3972 fold_image.
  3973 
  3974 * Theories Ring_and_Field and OrderedGroup: The lemmas "group_simps"
  3975 and "ring_simps" have been replaced by "algebra_simps" (which can be
  3976 extended with further lemmas!).  At the moment both still exist but
  3977 the former will disappear at some point.
  3978 
  3979 * Theory Power: Lemma power_Suc is now declared as a simp rule in
  3980 class recpower.  Type-specific simp rules for various recpower types
  3981 have been removed.  INCOMPATIBILITY, rename old lemmas as follows:
  3982 
  3983 rat_power_0    -> power_0
  3984 rat_power_Suc  -> power_Suc
  3985 realpow_0      -> power_0
  3986 realpow_Suc    -> power_Suc
  3987 complexpow_0   -> power_0
  3988 complexpow_Suc -> power_Suc
  3989 power_poly_0   -> power_0
  3990 power_poly_Suc -> power_Suc
  3991 
  3992 * Theories Ring_and_Field and Divides: Definition of "op dvd" has been
  3993 moved to separate class dvd in Ring_and_Field; a couple of lemmas on
  3994 dvd has been generalized to class comm_semiring_1.  Likewise a bunch
  3995 of lemmas from Divides has been generalized from nat to class
  3996 semiring_div.  INCOMPATIBILITY.  This involves the following theorem
  3997 renames resulting from duplicate elimination:
  3998 
  3999     dvd_def_mod ~>          dvd_eq_mod_eq_0
  4000     zero_dvd_iff ~>         dvd_0_left_iff
  4001     dvd_0 ~>                dvd_0_right
  4002     DIVISION_BY_ZERO_DIV ~> div_by_0
  4003     DIVISION_BY_ZERO_MOD ~> mod_by_0
  4004     mult_div ~>             div_mult_self2_is_id
  4005     mult_mod ~>             mod_mult_self2_is_0
  4006 
  4007 * Theory IntDiv: removed many lemmas that are instances of class-based
  4008 generalizations (from Divides and Ring_and_Field).  INCOMPATIBILITY,
  4009 rename old lemmas as follows:
  4010 
  4011 dvd_diff               -> nat_dvd_diff
  4012 dvd_zminus_iff         -> dvd_minus_iff
  4013 mod_add1_eq            -> mod_add_eq
  4014 mod_mult1_eq           -> mod_mult_right_eq
  4015 mod_mult1_eq'          -> mod_mult_left_eq
  4016 mod_mult_distrib_mod   -> mod_mult_eq
  4017 nat_mod_add_left_eq    -> mod_add_left_eq
  4018 nat_mod_add_right_eq   -> mod_add_right_eq
  4019 nat_mod_div_trivial    -> mod_div_trivial
  4020 nat_mod_mod_trivial    -> mod_mod_trivial
  4021 zdiv_zadd_self1        -> div_add_self1
  4022 zdiv_zadd_self2        -> div_add_self2
  4023 zdiv_zmult_self1       -> div_mult_self2_is_id
  4024 zdiv_zmult_self2       -> div_mult_self1_is_id
  4025 zdvd_triv_left         -> dvd_triv_left
  4026 zdvd_triv_right        -> dvd_triv_right
  4027 zdvd_zmult_cancel_disj -> dvd_mult_cancel_left
  4028 zmod_eq0_zdvd_iff      -> dvd_eq_mod_eq_0[symmetric]
  4029 zmod_zadd_left_eq      -> mod_add_left_eq
  4030 zmod_zadd_right_eq     -> mod_add_right_eq
  4031 zmod_zadd_self1        -> mod_add_self1
  4032 zmod_zadd_self2        -> mod_add_self2
  4033 zmod_zadd1_eq          -> mod_add_eq
  4034 zmod_zdiff1_eq         -> mod_diff_eq
  4035 zmod_zdvd_zmod         -> mod_mod_cancel
  4036 zmod_zmod_cancel       -> mod_mod_cancel
  4037 zmod_zmult_self1       -> mod_mult_self2_is_0
  4038 zmod_zmult_self2       -> mod_mult_self1_is_0
  4039 zmod_1                 -> mod_by_1
  4040 zdiv_1                 -> div_by_1
  4041 zdvd_abs1              -> abs_dvd_iff
  4042 zdvd_abs2              -> dvd_abs_iff
  4043 zdvd_refl              -> dvd_refl
  4044 zdvd_trans             -> dvd_trans
  4045 zdvd_zadd              -> dvd_add
  4046 zdvd_zdiff             -> dvd_diff
  4047 zdvd_zminus_iff        -> dvd_minus_iff
  4048 zdvd_zminus2_iff       -> minus_dvd_iff
  4049 zdvd_zmultD            -> dvd_mult_right
  4050 zdvd_zmultD2           -> dvd_mult_left
  4051 zdvd_zmult_mono        -> mult_dvd_mono
  4052 zdvd_0_right           -> dvd_0_right
  4053 zdvd_0_left            -> dvd_0_left_iff
  4054 zdvd_1_left            -> one_dvd
  4055 zminus_dvd_iff         -> minus_dvd_iff
  4056 
  4057 * Theory Rational: 'Fract k 0' now equals '0'.  INCOMPATIBILITY.
  4058 
  4059 * The real numbers offer decimal input syntax: 12.34 is translated
  4060 into 1234/10^2. This translation is not reversed upon output.
  4061 
  4062 * Theory Library/Polynomial defines an abstract type 'a poly of
  4063 univariate polynomials with coefficients of type 'a.  In addition to
  4064 the standard ring operations, it also supports div and mod.  Code
  4065 generation is also supported, using list-style constructors.
  4066 
  4067 * Theory Library/Inner_Product defines a class of real_inner for real
  4068 inner product spaces, with an overloaded operation inner :: 'a => 'a
  4069 => real.  Class real_inner is a subclass of real_normed_vector from
  4070 theory RealVector.
  4071 
  4072 * Theory Library/Product_Vector provides instances for the product
  4073 type 'a * 'b of several classes from RealVector and Inner_Product.
  4074 Definitions of addition, subtraction, scalar multiplication, norms,
  4075 and inner products are included.
  4076 
  4077 * Theory Library/Bit defines the field "bit" of integers modulo 2.  In
  4078 addition to the field operations, numerals and case syntax are also
  4079 supported.
  4080 
  4081 * Theory Library/Diagonalize provides constructive version of Cantor's
  4082 first diagonalization argument.
  4083 
  4084 * Theory Library/GCD: Curried operations gcd, lcm (for nat) and zgcd,
  4085 zlcm (for int); carried together from various gcd/lcm developements in
  4086 the HOL Distribution.  Constants zgcd and zlcm replace former igcd and
  4087 ilcm; corresponding theorems renamed accordingly.  INCOMPATIBILITY,
  4088 may recover tupled syntax as follows:
  4089 
  4090     hide (open) const gcd
  4091     abbreviation gcd where
  4092       "gcd == (%(a, b). GCD.gcd a b)"
  4093     notation (output)
  4094       GCD.gcd ("gcd '(_, _')")
  4095 
  4096 The same works for lcm, zgcd, zlcm.
  4097 
  4098 * Theory Library/Nat_Infinity: added addition, numeral syntax and more
  4099 instantiations for algebraic structures.  Removed some duplicate
  4100 theorems.  Changes in simp rules.  INCOMPATIBILITY.
  4101 
  4102 * ML antiquotation @{code} takes a constant as argument and generates
  4103 corresponding code in background and inserts name of the corresponding
  4104 resulting ML value/function/datatype constructor binding in place.
  4105 All occurrences of @{code} with a single ML block are generated
  4106 simultaneously.  Provides a generic and safe interface for
  4107 instrumentalizing code generation.  See
  4108 src/HOL/Decision_Procs/Ferrack.thy for a more ambitious application.
  4109 In future you ought to refrain from ad-hoc compiling generated SML
  4110 code on the ML toplevel.  Note that (for technical reasons) @{code}
  4111 cannot refer to constants for which user-defined serializations are
  4112 set.  Refer to the corresponding ML counterpart directly in that
  4113 cases.
  4114 
  4115 * Command 'rep_datatype': instead of theorem names the command now
  4116 takes a list of terms denoting the constructors of the type to be
  4117 represented as datatype.  The characteristic theorems have to be
  4118 proven.  INCOMPATIBILITY.  Also observe that the following theorems
  4119 have disappeared in favour of existing ones:
  4120 
  4121     unit_induct                 ~> unit.induct
  4122     prod_induct                 ~> prod.induct
  4123     sum_induct                  ~> sum.induct
  4124     Suc_Suc_eq                  ~> nat.inject
  4125     Suc_not_Zero Zero_not_Suc   ~> nat.distinct
  4126 
  4127 
  4128 *** HOL-Algebra ***
  4129 
  4130 * New locales for orders and lattices where the equivalence relation
  4131 is not restricted to equality.  INCOMPATIBILITY: all order and lattice
  4132 locales use a record structure with field eq for the equivalence.
  4133 
  4134 * New theory of factorial domains.
  4135 
  4136 * Units_l_inv and Units_r_inv are now simp rules by default.
  4137 INCOMPATIBILITY.  Simplifier proof that require deletion of l_inv
  4138 and/or r_inv will now also require deletion of these lemmas.
  4139 
  4140 * Renamed the following theorems, INCOMPATIBILITY:
  4141 
  4142 UpperD ~> Upper_memD
  4143 LowerD ~> Lower_memD
  4144 least_carrier ~> least_closed
  4145 greatest_carrier ~> greatest_closed
  4146 greatest_Lower_above ~> greatest_Lower_below
  4147 one_zero ~> carrier_one_zero
  4148 one_not_zero ~> carrier_one_not_zero  (collision with assumption)
  4149 
  4150 
  4151 *** HOL-Nominal ***
  4152 
  4153 * Nominal datatypes can now contain type-variables.
  4154 
  4155 * Commands 'nominal_inductive' and 'equivariance' work with local
  4156 theory targets.
  4157 
  4158 * Nominal primrec can now works with local theory targets and its
  4159 specification syntax now conforms to the general format as seen in
  4160 'inductive' etc.
  4161 
  4162 * Method "perm_simp" honours the standard simplifier attributes
  4163 (no_asm), (no_asm_use) etc.
  4164 
  4165 * The new predicate #* is defined like freshness, except that on the
  4166 left hand side can be a set or list of atoms.
  4167 
  4168 * Experimental command 'nominal_inductive2' derives strong induction
  4169 principles for inductive definitions.  In contrast to
  4170 'nominal_inductive', which can only deal with a fixed number of
  4171 binders, it can deal with arbitrary expressions standing for sets of
  4172 atoms to be avoided.  The only inductive definition we have at the
  4173 moment that needs this generalisation is the typing rule for Lets in
  4174 the algorithm W:
  4175 
  4176  Gamma |- t1 : T1   (x,close Gamma T1)::Gamma |- t2 : T2   x#Gamma
  4177  -----------------------------------------------------------------
  4178          Gamma |- Let x be t1 in t2 : T2
  4179 
  4180 In this rule one wants to avoid all the binders that are introduced by
  4181 "close Gamma T1".  We are looking for other examples where this
  4182 feature might be useful.  Please let us know.
  4183 
  4184 
  4185 *** HOLCF ***
  4186 
  4187 * Reimplemented the simplification procedure for proving continuity
  4188 subgoals.  The new simproc is extensible; users can declare additional
  4189 continuity introduction rules with the attribute [cont2cont].
  4190 
  4191 * The continuity simproc now uses a different introduction rule for
  4192 solving continuity subgoals on terms with lambda abstractions.  In
  4193 some rare cases the new simproc may fail to solve subgoals that the
  4194 old one could solve, and "simp add: cont2cont_LAM" may be necessary.
  4195 Potential INCOMPATIBILITY.
  4196 
  4197 * Command 'fixrec': specification syntax now conforms to the general
  4198 format as seen in 'inductive' etc.  See src/HOLCF/ex/Fixrec_ex.thy for
  4199 examples.  INCOMPATIBILITY.
  4200 
  4201 
  4202 *** ZF ***
  4203 
  4204 * Proof of Zorn's Lemma for partial orders.
  4205 
  4206 
  4207 *** ML ***
  4208 
  4209 * Multithreading for Poly/ML 5.1/5.2 is no longer supported, only for
  4210 Poly/ML 5.2.1 or later.  Important note: the TimeLimit facility
  4211 depends on multithreading, so timouts will not work before Poly/ML
  4212 5.2.1!
  4213 
  4214 * High-level support for concurrent ML programming, see
  4215 src/Pure/Cuncurrent.  The data-oriented model of "future values" is
  4216 particularly convenient to organize independent functional
  4217 computations.  The concept of "synchronized variables" provides a
  4218 higher-order interface for components with shared state, avoiding the
  4219 delicate details of mutexes and condition variables.  (Requires
  4220 Poly/ML 5.2.1 or later.)
  4221 
  4222 * ML bindings produced via Isar commands are stored within the Isar
  4223 context (theory or proof).  Consequently, commands like 'use' and 'ML'
  4224 become thread-safe and work with undo as expected (concerning
  4225 top-level bindings, not side-effects on global references).
  4226 INCOMPATIBILITY, need to provide proper Isar context when invoking the
  4227 compiler at runtime; really global bindings need to be given outside a
  4228 theory.  (Requires Poly/ML 5.2 or later.)
  4229 
  4230 * Command 'ML_prf' is analogous to 'ML' but works within a proof
  4231 context.  Top-level ML bindings are stored within the proof context in
  4232 a purely sequential fashion, disregarding the nested proof structure.
  4233 ML bindings introduced by 'ML_prf' are discarded at the end of the
  4234 proof.  (Requires Poly/ML 5.2 or later.)
  4235 
  4236 * Simplified ML attribute and method setup, cf. functions Attrib.setup
  4237 and Method.setup, as well as Isar commands 'attribute_setup' and
  4238 'method_setup'.  INCOMPATIBILITY for 'method_setup', need to simplify
  4239 existing code accordingly, or use plain 'setup' together with old
  4240 Method.add_method.
  4241 
  4242 * Simplified ML oracle interface Thm.add_oracle promotes 'a -> cterm
  4243 to 'a -> thm, while results are always tagged with an authentic oracle
  4244 name.  The Isar command 'oracle' is now polymorphic, no argument type
  4245 is specified.  INCOMPATIBILITY, need to simplify existing oracle code
  4246 accordingly.  Note that extra performance may be gained by producing
  4247 the cterm carefully, avoiding slow Thm.cterm_of.
  4248 
  4249 * Simplified interface for defining document antiquotations via
  4250 ThyOutput.antiquotation, ThyOutput.output, and optionally
  4251 ThyOutput.maybe_pretty_source.  INCOMPATIBILITY, need to simplify user
  4252 antiquotations accordingly, see src/Pure/Thy/thy_output.ML for common
  4253 examples.
  4254 
  4255 * More systematic treatment of long names, abstract name bindings, and
  4256 name space operations.  Basic operations on qualified names have been
  4257 move from structure NameSpace to Long_Name, e.g. Long_Name.base_name,
  4258 Long_Name.append.  Old type bstring has been mostly replaced by
  4259 abstract type binding (see structure Binding), which supports precise
  4260 qualification by packages and local theory targets, as well as proper
  4261 tracking of source positions.  INCOMPATIBILITY, need to wrap old
  4262 bstring values into Binding.name, or better pass through abstract
  4263 bindings everywhere.  See further src/Pure/General/long_name.ML,
  4264 src/Pure/General/binding.ML and src/Pure/General/name_space.ML
  4265 
  4266 * Result facts (from PureThy.note_thms, ProofContext.note_thms,
  4267 LocalTheory.note etc.) now refer to the *full* internal name, not the
  4268 bstring as before.  INCOMPATIBILITY, not detected by ML type-checking!
  4269 
  4270 * Disposed old type and term read functions (Sign.read_def_typ,
  4271 Sign.read_typ, Sign.read_def_terms, Sign.read_term,
  4272 Thm.read_def_cterms, Thm.read_cterm etc.).  INCOMPATIBILITY, should
  4273 use regular Syntax.read_typ, Syntax.read_term, Syntax.read_typ_global,
  4274 Syntax.read_term_global etc.; see also OldGoals.read_term as last
  4275 resort for legacy applications.
  4276 
  4277 * Disposed old declarations, tactics, tactic combinators that refer to
  4278 the simpset or claset of an implicit theory (such as Addsimps,
  4279 Simp_tac, SIMPSET).  INCOMPATIBILITY, should use @{simpset} etc. in
  4280 embedded ML text, or local_simpset_of with a proper context passed as
  4281 explicit runtime argument.
  4282 
  4283 * Rules and tactics that read instantiations (read_instantiate,
  4284 res_inst_tac, thin_tac, subgoal_tac etc.) now demand a proper proof
  4285 context, which is required for parsing and type-checking.  Moreover,
  4286 the variables are specified as plain indexnames, not string encodings
  4287 thereof.  INCOMPATIBILITY.
  4288 
  4289 * Generic Toplevel.add_hook interface allows to analyze the result of
  4290 transactions.  E.g. see src/Pure/ProofGeneral/proof_general_pgip.ML
  4291 for theorem dependency output of transactions resulting in a new
  4292 theory state.
  4293 
  4294 * ML antiquotations: block-structured compilation context indicated by
  4295 \<lbrace> ... \<rbrace>; additional antiquotation forms:
  4296 
  4297   @{binding name}                         - basic name binding
  4298   @{let ?pat = term}                      - term abbreviation (HO matching)
  4299   @{note name = fact}                     - fact abbreviation
  4300   @{thm fact}                             - singleton fact (with attributes)
  4301   @{thms fact}                            - general fact (with attributes)
  4302   @{lemma prop by method}                 - singleton goal
  4303   @{lemma prop by meth1 meth2}            - singleton goal
  4304   @{lemma prop1 ... propN by method}      - general goal
  4305   @{lemma prop1 ... propN by meth1 meth2} - general goal
  4306   @{lemma (open) ...}                     - open derivation
  4307 
  4308 
  4309 *** System ***
  4310 
  4311 * The Isabelle "emacs" tool provides a specific interface to invoke
  4312 Proof General / Emacs, with more explicit failure if that is not
  4313 installed (the old isabelle-interface script silently falls back on
  4314 isabelle-process).  The PROOFGENERAL_HOME setting determines the
  4315 installation location of the Proof General distribution.
  4316 
  4317 * Isabelle/lib/classes/Pure.jar provides basic support to integrate
  4318 the Isabelle process into a JVM/Scala application.  See
  4319 Isabelle/lib/jedit/plugin for a minimal example.  (The obsolete Java
  4320 process wrapper has been discontinued.)
  4321 
  4322 * Added homegrown Isabelle font with unicode layout, see lib/fonts.
  4323 
  4324 * Various status messages (with exact source position information) are
  4325 emitted, if proper markup print mode is enabled.  This allows
  4326 user-interface components to provide detailed feedback on internal
  4327 prover operations.
  4328 
  4329 
  4330 
  4331 New in Isabelle2008 (June 2008)
  4332 -------------------------------
  4333 
  4334 *** General ***
  4335 
  4336 * The Isabelle/Isar Reference Manual (isar-ref) has been reorganized
  4337 and updated, with formally checked references as hyperlinks.
  4338 
  4339 * Theory loader: use_thy (and similar operations) no longer set the
  4340 implicit ML context, which was occasionally hard to predict and in
  4341 conflict with concurrency.  INCOMPATIBILITY, use ML within Isar which
  4342 provides a proper context already.
  4343 
  4344 * Theory loader: old-style ML proof scripts being *attached* to a thy
  4345 file are no longer supported.  INCOMPATIBILITY, regular 'uses' and
  4346 'use' within a theory file will do the job.
  4347 
  4348 * Name space merge now observes canonical order, i.e. the second space
  4349 is inserted into the first one, while existing entries in the first
  4350 space take precedence.  INCOMPATIBILITY in rare situations, may try to
  4351 swap theory imports.
  4352 
  4353 * Syntax: symbol \<chi> is now considered a letter.  Potential
  4354 INCOMPATIBILITY in identifier syntax etc.
  4355 
  4356 * Outer syntax: string tokens no longer admit escaped white space,
  4357 which was an accidental (undocumented) feature.  INCOMPATIBILITY, use
  4358 white space without escapes.
  4359 
  4360 * Outer syntax: string tokens may contain arbitrary character codes
  4361 specified via 3 decimal digits (as in SML).  E.g. "foo\095bar" for
  4362 "foo_bar".
  4363 
  4364 
  4365 *** Pure ***
  4366 
  4367 * Context-dependent token translations.  Default setup reverts locally
  4368 fixed variables, and adds hilite markup for undeclared frees.
  4369 
  4370 * Unused theorems can be found using the new command 'unused_thms'.
  4371 There are three ways of invoking it:
  4372 
  4373 (1) unused_thms
  4374      Only finds unused theorems in the current theory.
  4375 
  4376 (2) unused_thms thy_1 ... thy_n -
  4377      Finds unused theorems in the current theory and all of its ancestors,
  4378      excluding the theories thy_1 ... thy_n and all of their ancestors.
  4379 
  4380 (3) unused_thms thy_1 ... thy_n - thy'_1 ... thy'_m
  4381      Finds unused theorems in the theories thy'_1 ... thy'_m and all of
  4382      their ancestors, excluding the theories thy_1 ... thy_n and all of
  4383      their ancestors.
  4384 
  4385 In order to increase the readability of the list produced by
  4386 unused_thms, theorems that have been created by a particular instance
  4387 of a theory command such as 'inductive' or 'function' are considered
  4388 to belong to the same "group", meaning that if at least one theorem in
  4389 this group is used, the other theorems in the same group are no longer
  4390 reported as unused.  Moreover, if all theorems in the group are
  4391 unused, only one theorem in the group is displayed.
  4392 
  4393 Note that proof objects have to be switched on in order for
  4394 unused_thms to work properly (i.e. !proofs must be >= 1, which is
  4395 usually the case when using Proof General with the default settings).
  4396 
  4397 * Authentic naming of facts disallows ad-hoc overwriting of previous
  4398 theorems within the same name space.  INCOMPATIBILITY, need to remove
  4399 duplicate fact bindings, or even accidental fact duplications.  Note
  4400 that tools may maintain dynamically scoped facts systematically, using
  4401 PureThy.add_thms_dynamic.
  4402 
  4403 * Command 'hide' now allows to hide from "fact" name space as well.
  4404 
  4405 * Eliminated destructive theorem database, simpset, claset, and
  4406 clasimpset.  Potential INCOMPATIBILITY, really need to observe linear
  4407 update of theories within ML code.
  4408 
  4409 * Eliminated theory ProtoPure and CPure, leaving just one Pure theory.
  4410 INCOMPATIBILITY, object-logics depending on former Pure require
  4411 additional setup PureThy.old_appl_syntax_setup; object-logics
  4412 depending on former CPure need to refer to Pure.
  4413 
  4414 * Commands 'use' and 'ML' are now purely functional, operating on
  4415 theory/local_theory.  Removed former 'ML_setup' (on theory), use 'ML'
  4416 instead.  Added 'ML_val' as mere diagnostic replacement for 'ML'.
  4417 INCOMPATIBILITY.
  4418 
  4419 * Command 'setup': discontinued implicit version with ML reference.
  4420 
  4421 * Instantiation target allows for simultaneous specification of class
  4422 instance operations together with an instantiation proof.
  4423 Type-checking phase allows to refer to class operations uniformly.
  4424 See src/HOL/Complex/Complex.thy for an Isar example and
  4425 src/HOL/Library/Eval.thy for an ML example.
  4426 
  4427 * Indexing of literal facts: be more serious about including only
  4428 facts from the visible specification/proof context, but not the
  4429 background context (locale etc.).  Affects `prop` notation and method
  4430 "fact".  INCOMPATIBILITY: need to name facts explicitly in rare
  4431 situations.
  4432 
  4433 * Method "cases", "induct", "coinduct": removed obsolete/undocumented
  4434 "(open)" option, which used to expose internal bound variables to the
  4435 proof text.
  4436 
  4437 * Isar statements: removed obsolete case "rule_context".
  4438 INCOMPATIBILITY, better use explicit fixes/assumes.
  4439 
  4440 * Locale proofs: default proof step now includes 'unfold_locales';
  4441 hence 'proof' without argument may be used to unfold locale
  4442 predicates.
  4443 
  4444 
  4445 *** Document preparation ***
  4446 
  4447 * Simplified pdfsetup.sty: color/hyperref is used unconditionally for
  4448 both pdf and dvi (hyperlinks usually work in xdvi as well); removed
  4449 obsolete thumbpdf setup (contemporary PDF viewers do this on the
  4450 spot); renamed link color from "darkblue" to "linkcolor" (default
  4451 value unchanged, can be redefined via \definecolor); no longer sets
  4452 "a4paper" option (unnecessary or even intrusive).
  4453 
  4454 * Antiquotation @{lemma A method} proves proposition A by the given
  4455 method (either a method name or a method name plus (optional) method
  4456 arguments in parentheses) and prints A just like @{prop A}.
  4457 
  4458 
  4459 *** HOL ***
  4460 
  4461 * New primrec package.  Specification syntax conforms in style to
  4462 definition/function/....  No separate induction rule is provided.  The
  4463 "primrec" command distinguishes old-style and new-style specifications
  4464 by syntax.  The former primrec package is now named OldPrimrecPackage.
  4465 When adjusting theories, beware: constants stemming from new-style
  4466 primrec specifications have authentic syntax.
  4467 
  4468 * Metis prover is now an order of magnitude faster, and also works
  4469 with multithreading.
  4470 
  4471 * Metis: the maximum number of clauses that can be produced from a
  4472 theorem is now given by the attribute max_clauses.  Theorems that
  4473 exceed this number are ignored, with a warning printed.
  4474 
  4475 * Sledgehammer no longer produces structured proofs by default. To
  4476 enable, declare [[sledgehammer_full = true]].  Attributes
  4477 reconstruction_modulus, reconstruction_sorts renamed
  4478 sledgehammer_modulus, sledgehammer_sorts.  INCOMPATIBILITY.
  4479 
  4480 * Method "induct_scheme" derives user-specified induction rules
  4481 from well-founded induction and completeness of patterns. This factors
  4482 out some operations that are done internally by the function package
  4483 and makes them available separately.  See
  4484 src/HOL/ex/Induction_Scheme.thy for examples.
  4485 
  4486 * More flexible generation of measure functions for termination
  4487 proofs: Measure functions can be declared by proving a rule of the
  4488 form "is_measure f" and giving it the [measure_function] attribute.
  4489 The "is_measure" predicate is logically meaningless (always true), and
  4490 just guides the heuristic.  To find suitable measure functions, the
  4491 termination prover sets up the goal "is_measure ?f" of the appropriate
  4492 type and generates all solutions by prolog-style backwards proof using
  4493 the declared rules.
  4494 
  4495 This setup also deals with rules like 
  4496 
  4497   "is_measure f ==> is_measure (list_size f)"
  4498 
  4499 which accommodates nested datatypes that recurse through lists.
  4500 Similar rules are predeclared for products and option types.
  4501 
  4502 * Turned the type of sets "'a set" into an abbreviation for "'a => bool"
  4503 
  4504   INCOMPATIBILITIES:
  4505 
  4506   - Definitions of overloaded constants on sets have to be replaced by
  4507     definitions on => and bool.
  4508 
  4509   - Some definitions of overloaded operators on sets can now be proved
  4510     using the definitions of the operators on => and bool.  Therefore,
  4511     the following theorems have been renamed:
  4512 
  4513       subset_def   -> subset_eq
  4514       psubset_def  -> psubset_eq
  4515       set_diff_def -> set_diff_eq
  4516       Compl_def    -> Compl_eq
  4517       Sup_set_def  -> Sup_set_eq
  4518       Inf_set_def  -> Inf_set_eq
  4519       sup_set_def  -> sup_set_eq
  4520       inf_set_def  -> inf_set_eq
  4521 
  4522   - Due to the incompleteness of the HO unification algorithm, some
  4523     rules such as subst may require manual instantiation, if some of
  4524     the unknowns in the rule is a set.
  4525 
  4526   - Higher order unification and forward proofs:
  4527     The proof pattern
  4528 
  4529       have "P (S::'a set)" <...>
  4530       then have "EX S. P S" ..
  4531 
  4532     no longer works (due to the incompleteness of the HO unification
  4533     algorithm) and must be replaced by the pattern
  4534 
  4535       have "EX S. P S"
  4536       proof
  4537         show "P S" <...>
  4538       qed
  4539 
  4540   - Calculational reasoning with subst (or similar rules):
  4541     The proof pattern
  4542 
  4543       have "P (S::'a set)" <...>
  4544       also have "S = T" <...>
  4545       finally have "P T" .
  4546 
  4547     no longer works (for similar reasons as the previous example) and
  4548     must be replaced by something like
  4549 
  4550       have "P (S::'a set)" <...>
  4551       moreover have "S = T" <...>
  4552       ultimately have "P T" by simp
  4553 
  4554   - Tactics or packages written in ML code:
  4555     Code performing pattern matching on types via
  4556 
  4557       Type ("set", [T]) => ...
  4558 
  4559     must be rewritten. Moreover, functions like strip_type or
  4560     binder_types no longer return the right value when applied to a
  4561     type of the form
  4562 
  4563       T1 => ... => Tn => U => bool
  4564 
  4565     rather than
  4566 
  4567       T1 => ... => Tn => U set
  4568 
  4569 * Merged theories Wellfounded_Recursion, Accessible_Part and
  4570 Wellfounded_Relations to theory Wellfounded.
  4571 
  4572 * Explicit class "eq" for executable equality.  INCOMPATIBILITY.
  4573 
  4574 * Class finite no longer treats UNIV as class parameter.  Use class
  4575 enum from theory Library/Enum instead to achieve a similar effect.
  4576 INCOMPATIBILITY.
  4577 
  4578 * Theory List: rule list_induct2 now has explicitly named cases "Nil"
  4579 and "Cons".  INCOMPATIBILITY.
  4580 
  4581 * HOL (and FOL): renamed variables in rules imp_elim and swap.
  4582 Potential INCOMPATIBILITY.
  4583 
  4584 * Theory Product_Type: duplicated lemmas split_Pair_apply and
  4585 injective_fst_snd removed, use split_eta and prod_eqI instead.
  4586 Renamed upd_fst to apfst and upd_snd to apsnd.  INCOMPATIBILITY.
  4587 
  4588 * Theory Nat: removed redundant lemmas that merely duplicate lemmas of
  4589 the same name in theory Orderings:
  4590 
  4591   less_trans
  4592   less_linear
  4593   le_imp_less_or_eq
  4594   le_less_trans
  4595   less_le_trans
  4596   less_not_sym
  4597   less_asym
  4598 
  4599 Renamed less_imp_le to less_imp_le_nat, and less_irrefl to
  4600 less_irrefl_nat.  Potential INCOMPATIBILITY due to more general types
  4601 and different variable names.
  4602 
  4603 * Library/Option_ord.thy: Canonical order on option type.
  4604 
  4605 * Library/RBT.thy: Red-black trees, an efficient implementation of
  4606 finite maps.
  4607 
  4608 * Library/Countable.thy: Type class for countable types.
  4609 
  4610 * Theory Int: The representation of numerals has changed.  The infix
  4611 operator BIT and the bit datatype with constructors B0 and B1 have
  4612 disappeared.  INCOMPATIBILITY, use "Int.Bit0 x" and "Int.Bit1 y" in
  4613 place of "x BIT bit.B0" and "y BIT bit.B1", respectively.  Theorems
  4614 involving BIT, B0, or B1 have been renamed with "Bit0" or "Bit1"
  4615 accordingly.
  4616 
  4617 * Theory Nat: definition of <= and < on natural numbers no longer
  4618 depend on well-founded relations.  INCOMPATIBILITY.  Definitions
  4619 le_def and less_def have disappeared.  Consider lemmas not_less
  4620 [symmetric, where ?'a = nat] and less_eq [symmetric] instead.
  4621 
  4622 * Theory Finite_Set: locales ACf, ACe, ACIf, ACIfSL and ACIfSLlin
  4623 (whose purpose mainly is for various fold_set functionals) have been
  4624 abandoned in favor of the existing algebraic classes
  4625 ab_semigroup_mult, comm_monoid_mult, ab_semigroup_idem_mult,
  4626 lower_semilattice (resp. upper_semilattice) and linorder.
  4627 INCOMPATIBILITY.
  4628 
  4629 * Theory Transitive_Closure: induct and cases rules now declare proper
  4630 case_names ("base" and "step").  INCOMPATIBILITY.
  4631 
  4632 * Theorem Inductive.lfp_ordinal_induct generalized to complete
  4633 lattices.  The form set-specific version is available as
  4634 Inductive.lfp_ordinal_induct_set.
  4635 
  4636 * Renamed theorems "power.simps" to "power_int.simps".
  4637 INCOMPATIBILITY.
  4638 
  4639 * Class semiring_div provides basic abstract properties of semirings
  4640 with division and modulo operations.  Subsumes former class dvd_mod.
  4641 
  4642 * Merged theories IntDef, Numeral and IntArith into unified theory
  4643 Int.  INCOMPATIBILITY.
  4644 
  4645 * Theory Library/Code_Index: type "index" now represents natural
  4646 numbers rather than integers.  INCOMPATIBILITY.
  4647 
  4648 * New class "uminus" with operation "uminus" (split of from class
  4649 "minus" which now only has operation "minus", binary).
  4650 INCOMPATIBILITY.
  4651 
  4652 * Constants "card", "internal_split", "option_map" now with authentic
  4653 syntax.  INCOMPATIBILITY.
  4654 
  4655 * Definitions subset_def, psubset_def, set_diff_def, Compl_def,
  4656 le_bool_def, less_bool_def, le_fun_def, less_fun_def, inf_bool_def,
  4657 sup_bool_def, Inf_bool_def, Sup_bool_def, inf_fun_def, sup_fun_def,
  4658 Inf_fun_def, Sup_fun_def, inf_set_def, sup_set_def, Inf_set_def,
  4659 Sup_set_def, le_def, less_def, option_map_def now with object
  4660 equality.  INCOMPATIBILITY.
  4661 
  4662 * Records. Removed K_record, and replaced it by pure lambda term
  4663 %x. c. The simplifier setup is now more robust against eta expansion.
  4664 INCOMPATIBILITY: in cases explicitly referring to K_record.
  4665 
  4666 * Library/Multiset: {#a, b, c#} abbreviates {#a#} + {#b#} + {#c#}.
  4667 
  4668 * Library/ListVector: new theory of arithmetic vector operations.
  4669 
  4670 * Library/Order_Relation: new theory of various orderings as sets of
  4671 pairs.  Defines preorders, partial orders, linear orders and
  4672 well-orders on sets and on types.
  4673 
  4674 
  4675 *** ZF ***
  4676 
  4677 * Renamed some theories to allow to loading both ZF and HOL in the
  4678 same session:
  4679 
  4680   Datatype  -> Datatype_ZF
  4681   Inductive -> Inductive_ZF
  4682   Int       -> Int_ZF
  4683   IntDiv    -> IntDiv_ZF
  4684   Nat       -> Nat_ZF
  4685   List      -> List_ZF
  4686   Main      -> Main_ZF
  4687 
  4688 INCOMPATIBILITY: ZF theories that import individual theories below
  4689 Main might need to be adapted.  Regular theory Main is still
  4690 available, as trivial extension of Main_ZF.
  4691 
  4692 
  4693 *** ML ***
  4694 
  4695 * ML within Isar: antiquotation @{const name} or @{const
  4696 name(typargs)} produces statically-checked Const term.
  4697 
  4698 * Functor NamedThmsFun: data is available to the user as dynamic fact
  4699 (of the same name).  Removed obsolete print command.
  4700 
  4701 * Removed obsolete "use_legacy_bindings" function.
  4702 
  4703 * The ``print mode'' is now a thread-local value derived from a global
  4704 template (the former print_mode reference), thus access becomes
  4705 non-critical.  The global print_mode reference is for session
  4706 management only; user-code should use print_mode_value,
  4707 print_mode_active, PrintMode.setmp etc.  INCOMPATIBILITY.
  4708 
  4709 * Functions system/system_out provide a robust way to invoke external
  4710 shell commands, with propagation of interrupts (requires Poly/ML
  4711 5.2.1).  Do not use OS.Process.system etc. from the basis library!
  4712 
  4713 
  4714 *** System ***
  4715 
  4716 * Default settings: PROOFGENERAL_OPTIONS no longer impose xemacs ---
  4717 in accordance with Proof General 3.7, which prefers GNU emacs.
  4718 
  4719 * isatool tty runs Isabelle process with plain tty interaction;
  4720 optional line editor may be specified via ISABELLE_LINE_EDITOR
  4721 setting, the default settings attempt to locate "ledit" and "rlwrap".
  4722 
  4723 * isatool browser now works with Cygwin as well, using general
  4724 "javapath" function defined in Isabelle process environment.
  4725 
  4726 * YXML notation provides a simple and efficient alternative to
  4727 standard XML transfer syntax.  See src/Pure/General/yxml.ML and
  4728 isatool yxml as described in the Isabelle system manual.
  4729 
  4730 * JVM class isabelle.IsabelleProcess (located in Isabelle/lib/classes)
  4731 provides general wrapper for managing an Isabelle process in a robust
  4732 fashion, with ``cooked'' output from stdin/stderr.
  4733 
  4734 * Rudimentary Isabelle plugin for jEdit (see Isabelle/lib/jedit),
  4735 based on Isabelle/JVM process wrapper (see Isabelle/lib/classes).
  4736 
  4737 * Removed obsolete THIS_IS_ISABELLE_BUILD feature.  NB: the documented
  4738 way of changing the user's settings is via
  4739 ISABELLE_HOME_USER/etc/settings, which is a fully featured bash
  4740 script.
  4741 
  4742 * Multithreading.max_threads := 0 refers to the number of actual CPU
  4743 cores of the underlying machine, which is a good starting point for
  4744 optimal performance tuning.  The corresponding usedir option -M allows
  4745 "max" as an alias for "0".  WARNING: does not work on certain versions
  4746 of Mac OS (with Poly/ML 5.1).
  4747 
  4748 * isabelle-process: non-ML sessions are run with "nice", to reduce the
  4749 adverse effect of Isabelle flooding interactive front-ends (notably
  4750 ProofGeneral / XEmacs).
  4751 
  4752 
  4753 
  4754 New in Isabelle2007 (November 2007)
  4755 -----------------------------------
  4756 
  4757 *** General ***
  4758 
  4759 * More uniform information about legacy features, notably a
  4760 warning/error of "Legacy feature: ...", depending on the state of the
  4761 tolerate_legacy_features flag (default true). FUTURE INCOMPATIBILITY:
  4762 legacy features will disappear eventually.
  4763 
  4764 * Theory syntax: the header format ``theory A = B + C:'' has been
  4765 discontinued in favour of ``theory A imports B C begin''.  Use isatool
  4766 fixheaders to convert existing theory files.  INCOMPATIBILITY.
  4767 
  4768 * Theory syntax: the old non-Isar theory file format has been
  4769 discontinued altogether.  Note that ML proof scripts may still be used
  4770 with Isar theories; migration is usually quite simple with the ML
  4771 function use_legacy_bindings.  INCOMPATIBILITY.
  4772 
  4773 * Theory syntax: some popular names (e.g. 'class', 'declaration',
  4774 'fun', 'help', 'if') are now keywords.  INCOMPATIBILITY, use double
  4775 quotes.
  4776 
  4777 * Theory loader: be more serious about observing the static theory
  4778 header specifications (including optional directories), but not the
  4779 accidental file locations of previously successful loads.  The strict
  4780 update policy of former update_thy is now already performed by
  4781 use_thy, so the former has been removed; use_thys updates several
  4782 theories simultaneously, just as 'imports' within a theory header
  4783 specification, but without merging the results.  Potential
  4784 INCOMPATIBILITY: may need to refine theory headers and commands
  4785 ROOT.ML which depend on load order.
  4786 
  4787 * Theory loader: optional support for content-based file
  4788 identification, instead of the traditional scheme of full physical
  4789 path plus date stamp; configured by the ISABELLE_FILE_IDENT setting
  4790 (cf. the system manual).  The new scheme allows to work with
  4791 non-finished theories in persistent session images, such that source
  4792 files may be moved later on without requiring reloads.
  4793 
  4794 * Theory loader: old-style ML proof scripts being *attached* to a thy
  4795 file (with the same base name as the theory) are considered a legacy
  4796 feature, which will disappear eventually. Even now, the theory loader
  4797 no longer maintains dependencies on such files.
  4798 
  4799 * Syntax: the scope for resolving ambiguities via type-inference is
  4800 now limited to individual terms, instead of whole simultaneous
  4801 specifications as before. This greatly reduces the complexity of the
  4802 syntax module and improves flexibility by separating parsing and
  4803 type-checking. INCOMPATIBILITY: additional type-constraints (explicit
  4804 'fixes' etc.) are required in rare situations.
  4805 
  4806 * Syntax: constants introduced by new-style packages ('definition',
  4807 'abbreviation' etc.) are passed through the syntax module in
  4808 ``authentic mode''. This means that associated mixfix annotations
  4809 really stick to such constants, independently of potential name space
  4810 ambiguities introduced later on. INCOMPATIBILITY: constants in parse
  4811 trees are represented slightly differently, may need to adapt syntax
  4812 translations accordingly. Use CONST marker in 'translations' and
  4813 @{const_syntax} antiquotation in 'parse_translation' etc.
  4814 
  4815 * Legacy goal package: reduced interface to the bare minimum required
  4816 to keep existing proof scripts running.  Most other user-level
  4817 functions are now part of the OldGoals structure, which is *not* open
  4818 by default (consider isatool expandshort before open OldGoals).
  4819 Removed top_sg, prin, printyp, pprint_term/typ altogether, because
  4820 these tend to cause confusion about the actual goal (!) context being
  4821 used here, which is not necessarily the same as the_context().
  4822 
  4823 * Command 'find_theorems': supports "*" wild-card in "name:"
  4824 criterion; "with_dups" option.  Certain ProofGeneral versions might
  4825 support a specific search form (see ProofGeneral/CHANGES).
  4826 
  4827 * The ``prems limit'' option (cf. ProofContext.prems_limit) is now -1
  4828 by default, which means that "prems" (and also "fixed variables") are
  4829 suppressed from proof state output.  Note that the ProofGeneral
  4830 settings mechanism allows to change and save options persistently, but
  4831 older versions of Isabelle will fail to start up if a negative prems
  4832 limit is imposed.
  4833 
  4834 * Local theory targets may be specified by non-nested blocks of
  4835 ``context/locale/class ... begin'' followed by ``end''.  The body may
  4836 contain definitions, theorems etc., including any derived mechanism
  4837 that has been implemented on top of these primitives.  This concept
  4838 generalizes the existing ``theorem (in ...)'' towards more versatility
  4839 and scalability.
  4840 
  4841 * Proof General interface: proper undo of final 'end' command;
  4842 discontinued Isabelle/classic mode (ML proof scripts).
  4843 
  4844 
  4845 *** Document preparation ***
  4846 
  4847 * Added antiquotation @{theory name} which prints the given name,
  4848 after checking that it refers to a valid ancestor theory in the
  4849 current context.
  4850 
  4851 * Added antiquotations @{ML_type text} and @{ML_struct text} which
  4852 check the given source text as ML type/structure, printing verbatim.
  4853 
  4854 * Added antiquotation @{abbrev "c args"} which prints the abbreviation
  4855 "c args == rhs" given in the current context.  (Any number of
  4856 arguments may be given on the LHS.)
  4857 
  4858 
  4859 *** Pure ***
  4860 
  4861 * The 'class' package offers a combination of axclass and locale to
  4862 achieve Haskell-like type classes in Isabelle.  Definitions and
  4863 theorems within a class context produce both relative results (with
  4864 implicit parameters according to the locale context), and polymorphic
  4865 constants with qualified polymorphism (according to the class
  4866 context).  Within the body context of a 'class' target, a separate
  4867 syntax layer ("user space type system") takes care of converting
  4868 between global polymorphic consts and internal locale representation.
  4869 See src/HOL/ex/Classpackage.thy for examples (as well as main HOL).
  4870 "isatool doc classes" provides a tutorial.
  4871 
  4872 * Generic code generator framework allows to generate executable
  4873 code for ML and Haskell (including Isabelle classes).  A short usage
  4874 sketch:
  4875 
  4876     internal compilation:
  4877         export_code <list of constants (term syntax)> in SML
  4878     writing SML code to a file:
  4879         export_code <list of constants (term syntax)> in SML <filename>
  4880     writing OCaml code to a file:
  4881         export_code <list of constants (term syntax)> in OCaml <filename>
  4882     writing Haskell code to a bunch of files:
  4883         export_code <list of constants (term syntax)> in Haskell <filename>
  4884 
  4885     evaluating closed propositions to True/False using code generation:
  4886         method ``eval''
  4887 
  4888 Reasonable default setup of framework in HOL.
  4889 
  4890 Theorem attributs for selecting and transforming function equations theorems:
  4891 
  4892     [code fun]:        select a theorem as function equation for a specific constant
  4893     [code fun del]:    deselect a theorem as function equation for a specific constant
  4894     [code inline]:     select an equation theorem for unfolding (inlining) in place
  4895     [code inline del]: deselect an equation theorem for unfolding (inlining) in place
  4896 
  4897 User-defined serializations (target in {SML, OCaml, Haskell}):
  4898 
  4899     code_const <and-list of constants (term syntax)>
  4900       {(target) <and-list of const target syntax>}+
  4901 
  4902     code_type <and-list of type constructors>
  4903       {(target) <and-list of type target syntax>}+
  4904 
  4905     code_instance <and-list of instances>
  4906       {(target)}+
  4907         where instance ::= <type constructor> :: <class>
  4908 
  4909     code_class <and_list of classes>
  4910       {(target) <and-list of class target syntax>}+
  4911         where class target syntax ::= <class name> {where {<classop> == <target syntax>}+}?
  4912 
  4913 code_instance and code_class only are effective to target Haskell.
  4914 
  4915 For example usage see src/HOL/ex/Codegenerator.thy and
  4916 src/HOL/ex/Codegenerator_Pretty.thy.  A separate tutorial on code
  4917 generation from Isabelle/HOL theories is available via "isatool doc
  4918 codegen".
  4919 
  4920 * Code generator: consts in 'consts_code' Isar commands are now
  4921 referred to by usual term syntax (including optional type
  4922 annotations).
  4923 
  4924 * Command 'no_translations' removes translation rules from theory
  4925 syntax.
  4926 
  4927 * Overloaded definitions are now actually checked for acyclic
  4928 dependencies.  The overloading scheme is slightly more general than
  4929 that of Haskell98, although Isabelle does not demand an exact
  4930 correspondence to type class and instance declarations.
  4931 INCOMPATIBILITY, use ``defs (unchecked overloaded)'' to admit more
  4932 exotic versions of overloading -- at the discretion of the user!
  4933 
  4934 Polymorphic constants are represented via type arguments, i.e. the
  4935 instantiation that matches an instance against the most general
  4936 declaration given in the signature.  For example, with the declaration
  4937 c :: 'a => 'a => 'a, an instance c :: nat => nat => nat is represented
  4938 as c(nat).  Overloading is essentially simultaneous structural
  4939 recursion over such type arguments.  Incomplete specification patterns
  4940 impose global constraints on all occurrences, e.g. c('a * 'a) on the
  4941 LHS means that more general c('a * 'b) will be disallowed on any RHS.
  4942 Command 'print_theory' outputs the normalized system of recursive
  4943 equations, see section "definitions".
  4944 
  4945 * Configuration options are maintained within the theory or proof
  4946 context (with name and type bool/int/string), providing a very simple
  4947 interface to a poor-man's version of general context data.  Tools may
  4948 declare options in ML (e.g. using Attrib.config_int) and then refer to
  4949 these values using Config.get etc.  Users may change options via an
  4950 associated attribute of the same name.  This form of context
  4951 declaration works particularly well with commands 'declare' or
  4952 'using', for example ``declare [[foo = 42]]''.  Thus it has become
  4953 very easy to avoid global references, which would not observe Isar
  4954 toplevel undo/redo and fail to work with multithreading.
  4955 
  4956 Various global ML references of Pure and HOL have been turned into
  4957 configuration options:
  4958 
  4959   Unify.search_bound		unify_search_bound
  4960   Unify.trace_bound		unify_trace_bound
  4961   Unify.trace_simp		unify_trace_simp
  4962   Unify.trace_types		unify_trace_types
  4963   Simplifier.simp_depth_limit	simp_depth_limit
  4964   Blast.depth_limit		blast_depth_limit
  4965   DatatypeProp.dtK		datatype_distinctness_limit
  4966   fast_arith_neq_limit  	fast_arith_neq_limit
  4967   fast_arith_split_limit	fast_arith_split_limit
  4968 
  4969 * Named collections of theorems may be easily installed as context
  4970 data using the functor NamedThmsFun (see also
  4971 src/Pure/Tools/named_thms.ML).  The user may add or delete facts via
  4972 attributes; there is also a toplevel print command.  This facility is
  4973 just a common case of general context data, which is the preferred way
  4974 for anything more complex than just a list of facts in canonical
  4975 order.
  4976 
  4977 * Isar: command 'declaration' augments a local theory by generic
  4978 declaration functions written in ML.  This enables arbitrary content
  4979 being added to the context, depending on a morphism that tells the
  4980 difference of the original declaration context wrt. the application
  4981 context encountered later on.
  4982 
  4983 * Isar: proper interfaces for simplification procedures.  Command
  4984 'simproc_setup' declares named simprocs (with match patterns, and body
  4985 text in ML).  Attribute "simproc" adds/deletes simprocs in the current
  4986 context.  ML antiquotation @{simproc name} retrieves named simprocs.
  4987 
  4988 * Isar: an extra pair of brackets around attribute declarations
  4989 abbreviates a theorem reference involving an internal dummy fact,
  4990 which will be ignored later --- only the effect of the attribute on
  4991 the background context will persist.  This form of in-place
  4992 declarations is particularly useful with commands like 'declare' and
  4993 'using', for example ``have A using [[simproc a]] by simp''.
  4994 
  4995 * Isar: method "assumption" (and implicit closing of subproofs) now
  4996 takes simple non-atomic goal assumptions into account: after applying
  4997 an assumption as a rule the resulting subgoals are solved by atomic
  4998 assumption steps.  This is particularly useful to finish 'obtain'
  4999 goals, such as "!!x. (!!x. P x ==> thesis) ==> P x ==> thesis",
  5000 without referring to the original premise "!!x. P x ==> thesis" in the
  5001 Isar proof context.  POTENTIAL INCOMPATIBILITY: method "assumption" is
  5002 more permissive.
  5003 
  5004 * Isar: implicit use of prems from the Isar proof context is
  5005 considered a legacy feature.  Common applications like ``have A .''
  5006 may be replaced by ``have A by fact'' or ``note `A`''.  In general,
  5007 referencing facts explicitly here improves readability and
  5008 maintainability of proof texts.
  5009 
  5010 * Isar: improper proof element 'guess' is like 'obtain', but derives
  5011 the obtained context from the course of reasoning!  For example:
  5012 
  5013   assume "EX x y. A x & B y"   -- "any previous fact"
  5014   then guess x and y by clarify
  5015 
  5016 This technique is potentially adventurous, depending on the facts and
  5017 proof tools being involved here.
  5018 
  5019 * Isar: known facts from the proof context may be specified as literal
  5020 propositions, using ASCII back-quote syntax.  This works wherever
  5021 named facts used to be allowed so far, in proof commands, proof
  5022 methods, attributes etc.  Literal facts are retrieved from the context
  5023 according to unification of type and term parameters.  For example,
  5024 provided that "A" and "A ==> B" and "!!x. P x ==> Q x" are known
  5025 theorems in the current context, then these are valid literal facts:
  5026 `A` and `A ==> B` and `!!x. P x ==> Q x" as well as `P a ==> Q a` etc.
  5027 
  5028 There is also a proof method "fact" which does the same composition
  5029 for explicit goal states, e.g. the following proof texts coincide with
  5030 certain special cases of literal facts:
  5031 
  5032   have "A" by fact                 ==  note `A`
  5033   have "A ==> B" by fact           ==  note `A ==> B`
  5034   have "!!x. P x ==> Q x" by fact  ==  note `!!x. P x ==> Q x`
  5035   have "P a ==> Q a" by fact       ==  note `P a ==> Q a`
  5036 
  5037 * Isar: ":" (colon) is no longer a symbolic identifier character in
  5038 outer syntax.  Thus symbolic identifiers may be used without
  5039 additional white space in declarations like this: ``assume *: A''.
  5040 
  5041 * Isar: 'print_facts' prints all local facts of the current context,
  5042 both named and unnamed ones.
  5043 
  5044 * Isar: 'def' now admits simultaneous definitions, e.g.:
  5045 
  5046   def x == "t" and y == "u"
  5047 
  5048 * Isar: added command 'unfolding', which is structurally similar to
  5049 'using', but affects both the goal state and facts by unfolding given
  5050 rewrite rules.  Thus many occurrences of the 'unfold' method or
  5051 'unfolded' attribute may be replaced by first-class proof text.
  5052 
  5053 * Isar: methods 'unfold' / 'fold', attributes 'unfolded' / 'folded',
  5054 and command 'unfolding' now all support object-level equalities
  5055 (potentially conditional).  The underlying notion of rewrite rule is
  5056 analogous to the 'rule_format' attribute, but *not* that of the
  5057 Simplifier (which is usually more generous).
  5058 
  5059 * Isar: the new attribute [rotated n] (default n = 1) rotates the
  5060 premises of a theorem by n. Useful in conjunction with drule.
  5061 
  5062 * Isar: the goal restriction operator [N] (default N = 1) evaluates a
  5063 method expression within a sandbox consisting of the first N
  5064 sub-goals, which need to exist.  For example, ``simp_all [3]''
  5065 simplifies the first three sub-goals, while (rule foo, simp_all)[]
  5066 simplifies all new goals that emerge from applying rule foo to the
  5067 originally first one.
  5068 
  5069 * Isar: schematic goals are no longer restricted to higher-order
  5070 patterns; e.g. ``lemma "?P(?x)" by (rule TrueI)'' now works as
  5071 expected.
  5072 
  5073 * Isar: the conclusion of a long theorem statement is now either
  5074 'shows' (a simultaneous conjunction, as before), or 'obtains'
  5075 (essentially a disjunction of cases with local parameters and
  5076 assumptions).  The latter allows to express general elimination rules
  5077 adequately; in this notation common elimination rules look like this:
  5078 
  5079   lemma exE:    -- "EX x. P x ==> (!!x. P x ==> thesis) ==> thesis"
  5080     assumes "EX x. P x"
  5081     obtains x where "P x"
  5082 
  5083   lemma conjE:  -- "A & B ==> (A ==> B ==> thesis) ==> thesis"
  5084     assumes "A & B"
  5085     obtains A and B
  5086 
  5087   lemma disjE:  -- "A | B ==> (A ==> thesis) ==> (B ==> thesis) ==> thesis"
  5088     assumes "A | B"
  5089     obtains
  5090       A
  5091     | B
  5092 
  5093 The subsequent classical rules even refer to the formal "thesis"
  5094 explicitly:
  5095 
  5096   lemma classical:     -- "(~ thesis ==> thesis) ==> thesis"
  5097     obtains "~ thesis"
  5098 
  5099   lemma Peirce's_Law:  -- "((thesis ==> something) ==> thesis) ==> thesis"
  5100     obtains "thesis ==> something"
  5101 
  5102 The actual proof of an 'obtains' statement is analogous to that of the
  5103 Isar proof element 'obtain', only that there may be several cases.
  5104 Optional case names may be specified in parentheses; these will be
  5105 available both in the present proof and as annotations in the
  5106 resulting rule, for later use with the 'cases' method (cf. attribute
  5107 case_names).
  5108 
  5109 * Isar: the assumptions of a long theorem statement are available as
  5110 "assms" fact in the proof context.  This is more appropriate than the
  5111 (historical) "prems", which refers to all assumptions of the current
  5112 context, including those from the target locale, proof body etc.
  5113 
  5114 * Isar: 'print_statement' prints theorems from the current theory or
  5115 proof context in long statement form, according to the syntax of a
  5116 top-level lemma.
  5117 
  5118 * Isar: 'obtain' takes an optional case name for the local context
  5119 introduction rule (default "that").
  5120 
  5121 * Isar: removed obsolete 'concl is' patterns.  INCOMPATIBILITY, use
  5122 explicit (is "_ ==> ?foo") in the rare cases where this still happens
  5123 to occur.
  5124 
  5125 * Pure: syntax "CONST name" produces a fully internalized constant
  5126 according to the current context.  This is particularly useful for
  5127 syntax translations that should refer to internal constant
  5128 representations independently of name spaces.
  5129 
  5130 * Pure: syntax constant for foo (binder "FOO ") is called "foo_binder"
  5131 instead of "FOO ". This allows multiple binder declarations to coexist
  5132 in the same context.  INCOMPATIBILITY.
  5133 
  5134 * Isar/locales: 'notation' provides a robust interface to the 'syntax'
  5135 primitive that also works in a locale context (both for constants and
  5136 fixed variables). Type declaration and internal syntactic representation
  5137 of given constants retrieved from the context. Likewise, the
  5138 'no_notation' command allows to remove given syntax annotations from the
  5139 current context.
  5140 
  5141 * Isar/locales: new derived specification elements 'axiomatization',
  5142 'definition', 'abbreviation', which support type-inference, admit
  5143 object-level specifications (equality, equivalence).  See also the
  5144 isar-ref manual.  Examples:
  5145 
  5146   axiomatization
  5147     eq  (infix "===" 50) where
  5148     eq_refl: "x === x" and eq_subst: "x === y ==> P x ==> P y"
  5149 
  5150   definition "f x y = x + y + 1"
  5151   definition g where "g x = f x x"
  5152 
  5153   abbreviation
  5154     neq  (infix "=!=" 50) where
  5155     "x =!= y == ~ (x === y)"
  5156 
  5157 These specifications may be also used in a locale context.  Then the
  5158 constants being introduced depend on certain fixed parameters, and the
  5159 constant name is qualified by the locale base name.  An internal
  5160 abbreviation takes care for convenient input and output, making the
  5161 parameters implicit and using the original short name.  See also
  5162 src/HOL/ex/Abstract_NAT.thy for an example of deriving polymorphic
  5163 entities from a monomorphic theory.
  5164 
  5165 Presently, abbreviations are only available 'in' a target locale, but
  5166 not inherited by general import expressions.  Also note that
  5167 'abbreviation' may be used as a type-safe replacement for 'syntax' +
  5168 'translations' in common applications.  The "no_abbrevs" print mode
  5169 prevents folding of abbreviations in term output.
  5170 
  5171 Concrete syntax is attached to specified constants in internal form,
  5172 independently of name spaces.  The parse tree representation is
  5173 slightly different -- use 'notation' instead of raw 'syntax', and
  5174 'translations' with explicit "CONST" markup to accommodate this.
  5175 
  5176 * Pure/Isar: unified syntax for new-style specification mechanisms
  5177 (e.g.  'definition', 'abbreviation', or 'inductive' in HOL) admits
  5178 full type inference and dummy patterns ("_").  For example:
  5179 
  5180   definition "K x _ = x"
  5181 
  5182   inductive conj for A B
  5183   where "A ==> B ==> conj A B"
  5184 
  5185 * Pure: command 'print_abbrevs' prints all constant abbreviations of
  5186 the current context.  Print mode "no_abbrevs" prevents inversion of
  5187 abbreviations on output.
  5188 
  5189 * Isar/locales: improved parameter handling: use of locales "var" and
  5190 "struct" no longer necessary; - parameter renamings are no longer
  5191 required to be injective.  For example, this allows to define
  5192 endomorphisms as locale endom = homom mult mult h.
  5193 
  5194 * Isar/locales: changed the way locales with predicates are defined.
  5195 Instead of accumulating the specification, the imported expression is
  5196 now an interpretation.  INCOMPATIBILITY: different normal form of
  5197 locale expressions.  In particular, in interpretations of locales with
  5198 predicates, goals repesenting already interpreted fragments are not
  5199 removed automatically.  Use methods `intro_locales' and
  5200 `unfold_locales'; see below.
  5201 
  5202 * Isar/locales: new methods `intro_locales' and `unfold_locales'
  5203 provide backward reasoning on locales predicates.  The methods are
  5204 aware of interpretations and discharge corresponding goals.
  5205 `intro_locales' is less aggressive then `unfold_locales' and does not
  5206 unfold predicates to assumptions.
  5207 
  5208 * Isar/locales: the order in which locale fragments are accumulated
  5209 has changed.  This enables to override declarations from fragments due
  5210 to interpretations -- for example, unwanted simp rules.
  5211 
  5212 * Isar/locales: interpretation in theories and proof contexts has been
  5213 extended.  One may now specify (and prove) equations, which are
  5214 unfolded in interpreted theorems.  This is useful for replacing
  5215 defined concepts (constants depending on locale parameters) by
  5216 concepts already existing in the target context.  Example:
  5217 
  5218   interpretation partial_order ["op <= :: [int, int] => bool"]
  5219     where "partial_order.less (op <=) (x::int) y = (x < y)"
  5220 
  5221 Typically, the constant `partial_order.less' is created by a
  5222 definition specification element in the context of locale
  5223 partial_order.
  5224 
  5225 * Method "induct": improved internal context management to support
  5226 local fixes and defines on-the-fly. Thus explicit meta-level
  5227 connectives !!  and ==> are rarely required anymore in inductive goals
  5228 (using object-logic connectives for this purpose has been long
  5229 obsolete anyway). Common proof patterns are explained in
  5230 src/HOL/Induct/Common_Patterns.thy, see also
  5231 src/HOL/Isar_examples/Puzzle.thy and src/HOL/Lambda for realistic
  5232 examples.
  5233 
  5234 * Method "induct": improved handling of simultaneous goals. Instead of
  5235 introducing object-level conjunction, the statement is now split into
  5236 several conclusions, while the corresponding symbolic cases are nested
  5237 accordingly. INCOMPATIBILITY, proofs need to be structured explicitly,
  5238 see src/HOL/Induct/Common_Patterns.thy, for example.
  5239 
  5240 * Method "induct": mutual induction rules are now specified as a list
  5241 of rule sharing the same induction cases. HOL packages usually provide
  5242 foo_bar.inducts for mutually defined items foo and bar (e.g. inductive
  5243 predicates/sets or datatypes). INCOMPATIBILITY, users need to specify
  5244 mutual induction rules differently, i.e. like this:
  5245 
  5246   (induct rule: foo_bar.inducts)
  5247   (induct set: foo bar)
  5248   (induct pred: foo bar)
  5249   (induct type: foo bar)
  5250 
  5251 The ML function ProjectRule.projections turns old-style rules into the
  5252 new format.
  5253 
  5254 * Method "coinduct": dual of induction, see
  5255 src/HOL/Library/Coinductive_List.thy for various examples.
  5256 
  5257 * Method "cases", "induct", "coinduct": the ``(open)'' option is
  5258 considered a legacy feature.
  5259 
  5260 * Attribute "symmetric" produces result with standardized schematic
  5261 variables (index 0).  Potential INCOMPATIBILITY.
  5262 
  5263 * Simplifier: by default the simplifier trace only shows top level
  5264 rewrites now. That is, trace_simp_depth_limit is set to 1 by
  5265 default. Thus there is less danger of being flooded by the trace. The
  5266 trace indicates where parts have been suppressed.
  5267   
  5268 * Provers/classical: removed obsolete classical version of elim_format
  5269 attribute; classical elim/dest rules are now treated uniformly when
  5270 manipulating the claset.
  5271 
  5272 * Provers/classical: stricter checks to ensure that supplied intro,
  5273 dest and elim rules are well-formed; dest and elim rules must have at
  5274 least one premise.
  5275 
  5276 * Provers/classical: attributes dest/elim/intro take an optional
  5277 weight argument for the rule (just as the Pure versions).  Weights are
  5278 ignored by automated tools, but determine the search order of single
  5279 rule steps.
  5280 
  5281 * Syntax: input syntax now supports dummy variable binding "%_. b",
  5282 where the body does not mention the bound variable.  Note that dummy
  5283 patterns implicitly depend on their context of bounds, which makes
  5284 "{_. _}" match any set comprehension as expected.  Potential
  5285 INCOMPATIBILITY -- parse translations need to cope with syntactic
  5286 constant "_idtdummy" in the binding position.
  5287 
  5288 * Syntax: removed obsolete syntactic constant "_K" and its associated
  5289 parse translation.  INCOMPATIBILITY -- use dummy abstraction instead,
  5290 for example "A -> B" => "Pi A (%_. B)".
  5291 
  5292 * Pure: 'class_deps' command visualizes the subclass relation, using
  5293 the graph browser tool.
  5294 
  5295 * Pure: 'print_theory' now suppresses certain internal declarations by
  5296 default; use '!' option for full details.
  5297 
  5298 
  5299 *** HOL ***
  5300 
  5301 * Method "metis" proves goals by applying the Metis general-purpose
  5302 resolution prover (see also http://gilith.com/software/metis/).
  5303 Examples are in the directory MetisExamples.  WARNING: the
  5304 Isabelle/HOL-Metis integration does not yet work properly with
  5305 multi-threading.
  5306   
  5307 * Command 'sledgehammer' invokes external automatic theorem provers as
  5308 background processes.  It generates calls to the "metis" method if
  5309 successful. These can be pasted into the proof.  Users do not have to
  5310 wait for the automatic provers to return.  WARNING: does not really
  5311 work with multi-threading.
  5312 
  5313 * New "auto_quickcheck" feature tests outermost goal statements for
  5314 potential counter-examples.  Controlled by ML references
  5315 auto_quickcheck (default true) and auto_quickcheck_time_limit (default
  5316 5000 milliseconds).  Fails silently if statements is outside of
  5317 executable fragment, or any other codgenerator problem occurs.
  5318 
  5319 * New constant "undefined" with axiom "undefined x = undefined".
  5320 
  5321 * Added class "HOL.eq", allowing for code generation with polymorphic
  5322 equality.
  5323 
  5324 * Some renaming of class constants due to canonical name prefixing in
  5325 the new 'class' package:
  5326 
  5327     HOL.abs ~> HOL.abs_class.abs
  5328     HOL.divide ~> HOL.divide_class.divide
  5329     0 ~> HOL.zero_class.zero
  5330     1 ~> HOL.one_class.one
  5331     op + ~> HOL.plus_class.plus
  5332     op - ~> HOL.minus_class.minus
  5333     uminus ~> HOL.minus_class.uminus
  5334     op * ~> HOL.times_class.times
  5335     op < ~> HOL.ord_class.less
  5336     op <= > HOL.ord_class.less_eq
  5337     Nat.power ~> Power.power_class.power
  5338     Nat.size ~> Nat.size_class.size
  5339     Numeral.number_of ~> Numeral.number_class.number_of
  5340     FixedPoint.Inf ~> Lattices.complete_lattice_class.Inf
  5341     FixedPoint.Sup ~> Lattices.complete_lattice_class.Sup
  5342     Orderings.min ~> Orderings.ord_class.min
  5343     Orderings.max ~> Orderings.ord_class.max
  5344     Divides.op div ~> Divides.div_class.div
  5345     Divides.op mod ~> Divides.div_class.mod
  5346     Divides.op dvd ~> Divides.div_class.dvd
  5347 
  5348 INCOMPATIBILITY.  Adaptions may be required in the following cases:
  5349 
  5350 a) User-defined constants using any of the names "plus", "minus",
  5351 "times", "less" or "less_eq". The standard syntax translations for
  5352 "+", "-" and "*" may go wrong.  INCOMPATIBILITY: use more specific
  5353 names.
  5354 
  5355 b) Variables named "plus", "minus", "times", "less", "less_eq"
  5356 INCOMPATIBILITY: use more specific names.
  5357 
  5358 c) Permutative equations (e.g. "a + b = b + a")
  5359 Since the change of names also changes the order of terms, permutative
  5360 rewrite rules may get applied in a different order. Experience shows
  5361 that this is rarely the case (only two adaptions in the whole Isabelle
  5362 distribution).  INCOMPATIBILITY: rewrite proofs
  5363 
  5364 d) ML code directly refering to constant names
  5365 This in general only affects hand-written proof tactics, simprocs and
  5366 so on.  INCOMPATIBILITY: grep your sourcecode and replace names.
  5367 Consider using @{const_name} antiquotation.
  5368 
  5369 * New class "default" with associated constant "default".
  5370 
  5371 * Function "sgn" is now overloaded and available on int, real, complex
  5372 (and other numeric types), using class "sgn".  Two possible defs of
  5373 sgn are given as equational assumptions in the classes sgn_if and
  5374 sgn_div_norm; ordered_idom now also inherits from sgn_if.
  5375 INCOMPATIBILITY.
  5376 
  5377 * Locale "partial_order" now unified with class "order" (cf. theory
  5378 Orderings), added parameter "less".  INCOMPATIBILITY.
  5379 
  5380 * Renamings in classes "order" and "linorder": facts "refl", "trans" and
  5381 "cases" to "order_refl", "order_trans" and "linorder_cases", to avoid
  5382 clashes with HOL "refl" and "trans".  INCOMPATIBILITY.
  5383 
  5384 * Classes "order" and "linorder": potential INCOMPATIBILITY due to
  5385 changed order of proof goals in instance proofs.
  5386 
  5387 * The transitivity reasoner for partial and linear orders is set up
  5388 for classes "order" and "linorder".  Instances of the reasoner are available
  5389 in all contexts importing or interpreting the corresponding locales.
  5390 Method "order" invokes the reasoner separately; the reasoner
  5391 is also integrated with the Simplifier as a solver.  Diagnostic
  5392 command 'print_orders' shows the available instances of the reasoner
  5393 in the current context.
  5394 
  5395 * Localized monotonicity predicate in theory "Orderings"; integrated
  5396 lemmas max_of_mono and min_of_mono with this predicate.
  5397 INCOMPATIBILITY.
  5398 
  5399 * Formulation of theorem "dense" changed slightly due to integration
  5400 with new class dense_linear_order.
  5401 
  5402 * Uniform lattice theory development in HOL.
  5403 
  5404     constants "meet" and "join" now named "inf" and "sup"
  5405     constant "Meet" now named "Inf"
  5406 
  5407     classes "meet_semilorder" and "join_semilorder" now named
  5408       "lower_semilattice" and "upper_semilattice"
  5409     class "lorder" now named "lattice"
  5410     class "comp_lat" now named "complete_lattice"
  5411 
  5412     Instantiation of lattice classes allows explicit definitions
  5413     for "inf" and "sup" operations (or "Inf" and "Sup" for complete lattices).
  5414 
  5415   INCOMPATIBILITY.  Theorem renames:
  5416 
  5417     meet_left_le            ~> inf_le1
  5418     meet_right_le           ~> inf_le2
  5419     join_left_le            ~> sup_ge1
  5420     join_right_le           ~> sup_ge2
  5421     meet_join_le            ~> inf_sup_ord
  5422     le_meetI                ~> le_infI
  5423     join_leI                ~> le_supI
  5424     le_meet                 ~> le_inf_iff
  5425     le_join                 ~> ge_sup_conv
  5426     meet_idempotent         ~> inf_idem
  5427     join_idempotent         ~> sup_idem
  5428     meet_comm               ~> inf_commute
  5429     join_comm               ~> sup_commute
  5430     meet_leI1               ~> le_infI1
  5431     meet_leI2               ~> le_infI2
  5432     le_joinI1               ~> le_supI1
  5433     le_joinI2               ~> le_supI2
  5434     meet_assoc              ~> inf_assoc
  5435     join_assoc              ~> sup_assoc
  5436     meet_left_comm          ~> inf_left_commute
  5437     meet_left_idempotent    ~> inf_left_idem
  5438     join_left_comm          ~> sup_left_commute
  5439     join_left_idempotent    ~> sup_left_idem
  5440     meet_aci                ~> inf_aci
  5441     join_aci                ~> sup_aci
  5442     le_def_meet             ~> le_iff_inf
  5443     le_def_join             ~> le_iff_sup
  5444     join_absorp2            ~> sup_absorb2
  5445     join_absorp1            ~> sup_absorb1
  5446     meet_absorp1            ~> inf_absorb1
  5447     meet_absorp2            ~> inf_absorb2
  5448     meet_join_absorp        ~> inf_sup_absorb
  5449     join_meet_absorp        ~> sup_inf_absorb
  5450     distrib_join_le         ~> distrib_sup_le
  5451     distrib_meet_le         ~> distrib_inf_le
  5452 
  5453     add_meet_distrib_left   ~> add_inf_distrib_left
  5454     add_join_distrib_left   ~> add_sup_distrib_left
  5455     is_join_neg_meet        ~> is_join_neg_inf
  5456     is_meet_neg_join        ~> is_meet_neg_sup
  5457     add_meet_distrib_right  ~> add_inf_distrib_right
  5458     add_join_distrib_right  ~> add_sup_distrib_right
  5459     add_meet_join_distribs  ~> add_sup_inf_distribs
  5460     join_eq_neg_meet        ~> sup_eq_neg_inf
  5461     meet_eq_neg_join        ~> inf_eq_neg_sup
  5462     add_eq_meet_join        ~> add_eq_inf_sup
  5463     meet_0_imp_0            ~> inf_0_imp_0
  5464     join_0_imp_0            ~> sup_0_imp_0
  5465     meet_0_eq_0             ~> inf_0_eq_0
  5466     join_0_eq_0             ~> sup_0_eq_0
  5467     neg_meet_eq_join        ~> neg_inf_eq_sup
  5468     neg_join_eq_meet        ~> neg_sup_eq_inf
  5469     join_eq_if              ~> sup_eq_if
  5470 
  5471     mono_meet               ~> mono_inf
  5472     mono_join               ~> mono_sup
  5473     meet_bool_eq            ~> inf_bool_eq
  5474     join_bool_eq            ~> sup_bool_eq
  5475     meet_fun_eq             ~> inf_fun_eq
  5476     join_fun_eq             ~> sup_fun_eq
  5477     meet_set_eq             ~> inf_set_eq
  5478     join_set_eq             ~> sup_set_eq
  5479     meet1_iff               ~> inf1_iff
  5480     meet2_iff               ~> inf2_iff
  5481     meet1I                  ~> inf1I
  5482     meet2I                  ~> inf2I
  5483     meet1D1                 ~> inf1D1
  5484     meet2D1                 ~> inf2D1
  5485     meet1D2                 ~> inf1D2
  5486     meet2D2                 ~> inf2D2
  5487     meet1E                  ~> inf1E
  5488     meet2E                  ~> inf2E
  5489     join1_iff               ~> sup1_iff
  5490     join2_iff               ~> sup2_iff
  5491     join1I1                 ~> sup1I1
  5492     join2I1                 ~> sup2I1
  5493     join1I1                 ~> sup1I1
  5494     join2I2                 ~> sup1I2
  5495     join1CI                 ~> sup1CI
  5496     join2CI                 ~> sup2CI
  5497     join1E                  ~> sup1E
  5498     join2E                  ~> sup2E
  5499 
  5500     is_meet_Meet            ~> is_meet_Inf
  5501     Meet_bool_def           ~> Inf_bool_def
  5502     Meet_fun_def            ~> Inf_fun_def
  5503     Meet_greatest           ~> Inf_greatest
  5504     Meet_lower              ~> Inf_lower
  5505     Meet_set_def            ~> Inf_set_def
  5506 
  5507     Sup_def                 ~> Sup_Inf
  5508     Sup_bool_eq             ~> Sup_bool_def
  5509     Sup_fun_eq              ~> Sup_fun_def
  5510     Sup_set_eq              ~> Sup_set_def
  5511 
  5512     listsp_meetI            ~> listsp_infI
  5513     listsp_meet_eq          ~> listsp_inf_eq
  5514 
  5515     meet_min                ~> inf_min
  5516     join_max                ~> sup_max
  5517 
  5518 * Added syntactic class "size"; overloaded constant "size" now has
  5519 type "'a::size ==> bool"
  5520 
  5521 * Internal reorganisation of `size' of datatypes: size theorems
  5522 "foo.size" are no longer subsumed by "foo.simps" (but are still
  5523 simplification rules by default!); theorems "prod.size" now named
  5524 "*.size".
  5525 
  5526 * Class "div" now inherits from class "times" rather than "type".
  5527 INCOMPATIBILITY.
  5528 
  5529 * HOL/Finite_Set: "name-space" locales Lattice, Distrib_lattice,
  5530 Linorder etc.  have disappeared; operations defined in terms of
  5531 fold_set now are named Inf_fin, Sup_fin.  INCOMPATIBILITY.
  5532 
  5533 * HOL/Nat: neq0_conv no longer declared as iff.  INCOMPATIBILITY.
  5534 
  5535 * HOL-Word: New extensive library and type for generic, fixed size
  5536 machine words, with arithmetic, bit-wise, shifting and rotating
  5537 operations, reflection into int, nat, and bool lists, automation for
  5538 linear arithmetic (by automatic reflection into nat or int), including
  5539 lemmas on overflow and monotonicity.  Instantiated to all appropriate
  5540 arithmetic type classes, supporting automatic simplification of
  5541 numerals on all operations.
  5542 
  5543 * Library/Boolean_Algebra: locales for abstract boolean algebras.
  5544 
  5545 * Library/Numeral_Type: numbers as types, e.g. TYPE(32).
  5546 
  5547 * Code generator library theories:
  5548   - Code_Integer represents HOL integers by big integer literals in target
  5549     languages.
  5550   - Code_Char represents HOL characters by character literals in target
  5551     languages.
  5552   - Code_Char_chr like Code_Char, but also offers treatment of character
  5553     codes; includes Code_Integer.
  5554   - Executable_Set allows to generate code for finite sets using lists.
  5555   - Executable_Rat implements rational numbers as triples (sign, enumerator,
  5556     denominator).
  5557   - Executable_Real implements a subset of real numbers, namly those
  5558     representable by rational numbers.
  5559   - Efficient_Nat implements natural numbers by integers, which in general will
  5560     result in higher efficency; pattern matching with 0/Suc is eliminated;
  5561     includes Code_Integer.
  5562   - Code_Index provides an additional datatype index which is mapped to
  5563     target-language built-in integers.
  5564   - Code_Message provides an additional datatype message_string which is isomorphic to
  5565     strings; messages are mapped to target-language strings.
  5566 
  5567 * New package for inductive predicates
  5568 
  5569   An n-ary predicate p with m parameters z_1, ..., z_m can now be defined via
  5570 
  5571     inductive
  5572       p :: "U_1 => ... => U_m => T_1 => ... => T_n => bool"
  5573       for z_1 :: U_1 and ... and z_n :: U_m
  5574     where
  5575       rule_1: "... ==> p z_1 ... z_m t_1_1 ... t_1_n"
  5576     | ...
  5577 
  5578   with full support for type-inference, rather than
  5579 
  5580     consts s :: "U_1 => ... => U_m => (T_1 * ... * T_n) set"
  5581 
  5582     abbreviation p :: "U_1 => ... => U_m => T_1 => ... => T_n => bool"
  5583     where "p z_1 ... z_m x_1 ... x_n == (x_1, ..., x_n) : s z_1 ... z_m"
  5584 
  5585     inductive "s z_1 ... z_m"
  5586     intros
  5587       rule_1: "... ==> (t_1_1, ..., t_1_n) : s z_1 ... z_m"
  5588       ...
  5589 
  5590   For backward compatibility, there is a wrapper allowing inductive
  5591   sets to be defined with the new package via
  5592 
  5593     inductive_set
  5594       s :: "U_1 => ... => U_m => (T_1 * ... * T_n) set"
  5595       for z_1 :: U_1 and ... and z_n :: U_m
  5596     where
  5597       rule_1: "... ==> (t_1_1, ..., t_1_n) : s z_1 ... z_m"
  5598     | ...
  5599 
  5600   or
  5601 
  5602     inductive_set
  5603       s :: "U_1 => ... => U_m => (T_1 * ... * T_n) set"
  5604       and p :: "U_1 => ... => U_m => T_1 => ... => T_n => bool"
  5605       for z_1 :: U_1 and ... and z_n :: U_m
  5606     where
  5607       "p z_1 ... z_m x_1 ... x_n == (x_1, ..., x_n) : s z_1 ... z_m"
  5608     | rule_1: "... ==> p z_1 ... z_m t_1_1 ... t_1_n"
  5609     | ...
  5610 
  5611   if the additional syntax "p ..." is required.
  5612 
  5613   Numerous examples can be found in the subdirectories src/HOL/Auth,
  5614   src/HOL/Bali, src/HOL/Induct, and src/HOL/MicroJava.
  5615 
  5616   INCOMPATIBILITIES:
  5617 
  5618   - Since declaration and definition of inductive sets or predicates
  5619     is no longer separated, abbreviations involving the newly
  5620     introduced sets or predicates must be specified together with the
  5621     introduction rules after the 'where' keyword (see above), rather
  5622     than before the actual inductive definition.
  5623 
  5624   - The variables in induction and elimination rules are now
  5625     quantified in the order of their occurrence in the introduction
  5626     rules, rather than in alphabetical order. Since this may break
  5627     some proofs, these proofs either have to be repaired, e.g. by
  5628     reordering the variables a_i_1 ... a_i_{k_i} in Isar 'case'
  5629     statements of the form
  5630 
  5631       case (rule_i a_i_1 ... a_i_{k_i})
  5632 
  5633     or the old order of quantification has to be restored by explicitly adding
  5634     meta-level quantifiers in the introduction rules, i.e.
  5635 
  5636       | rule_i: "!!a_i_1 ... a_i_{k_i}. ... ==> p z_1 ... z_m t_i_1 ... t_i_n"
  5637 
  5638   - The format of the elimination rules is now
  5639 
  5640       p z_1 ... z_m x_1 ... x_n ==>
  5641         (!!a_1_1 ... a_1_{k_1}. x_1 = t_1_1 ==> ... ==> x_n = t_1_n ==> ... ==> P)
  5642         ==> ... ==> P
  5643 
  5644     for predicates and
  5645 
  5646       (x_1, ..., x_n) : s z_1 ... z_m ==>
  5647         (!!a_1_1 ... a_1_{k_1}. x_1 = t_1_1 ==> ... ==> x_n = t_1_n ==> ... ==> P)
  5648         ==> ... ==> P
  5649 
  5650     for sets rather than
  5651 
  5652       x : s z_1 ... z_m ==>
  5653         (!!a_1_1 ... a_1_{k_1}. x = (t_1_1, ..., t_1_n) ==> ... ==> P)
  5654         ==> ... ==> P
  5655 
  5656     This may require terms in goals to be expanded to n-tuples
  5657     (e.g. using case_tac or simplification with the split_paired_all
  5658     rule) before the above elimination rule is applicable.
  5659 
  5660   - The elimination or case analysis rules for (mutually) inductive
  5661     sets or predicates are now called "p_1.cases" ... "p_k.cases". The
  5662     list of rules "p_1_..._p_k.elims" is no longer available.
  5663 
  5664 * New package "function"/"fun" for general recursive functions,
  5665 supporting mutual and nested recursion, definitions in local contexts,
  5666 more general pattern matching and partiality. See HOL/ex/Fundefs.thy
  5667 for small examples, and the separate tutorial on the function
  5668 package. The old recdef "package" is still available as before, but
  5669 users are encouraged to use the new package.
  5670 
  5671 * Method "lexicographic_order" automatically synthesizes termination
  5672 relations as lexicographic combinations of size measures. 
  5673 
  5674 * Case-expressions allow arbitrary constructor-patterns (including
  5675 "_") and take their order into account, like in functional
  5676 programming.  Internally, this is translated into nested
  5677 case-expressions; missing cases are added and mapped to the predefined
  5678 constant "undefined". In complicated cases printing may no longer show
  5679 the original input but the internal form. Lambda-abstractions allow
  5680 the same form of pattern matching: "% pat1 => e1 | ..." is an
  5681 abbreviation for "%x. case x of pat1 => e1 | ..." where x is a new
  5682 variable.
  5683 
  5684 * IntDef: The constant "int :: nat => int" has been removed; now "int"
  5685 is an abbreviation for "of_nat :: nat => int". The simplification
  5686 rules for "of_nat" have been changed to work like "int" did
  5687 previously.  Potential INCOMPATIBILITY:
  5688   - "of_nat (Suc m)" simplifies to "1 + of_nat m" instead of "of_nat m + 1"
  5689   - of_nat_diff and of_nat_mult are no longer default simp rules
  5690 
  5691 * Method "algebra" solves polynomial equations over (semi)rings using
  5692 Groebner bases. The (semi)ring structure is defined by locales and the
  5693 tool setup depends on that generic context. Installing the method for
  5694 a specific type involves instantiating the locale and possibly adding
  5695 declarations for computation on the coefficients.  The method is
  5696 already instantiated for natural numbers and for the axiomatic class
  5697 of idoms with numerals.  See also the paper by Chaieb and Wenzel at
  5698 CALCULEMUS 2007 for the general principles underlying this
  5699 architecture of context-aware proof-tools.
  5700 
  5701 * Method "ferrack" implements quantifier elimination over
  5702 special-purpose dense linear orders using locales (analogous to
  5703 "algebra"). The method is already installed for class
  5704 {ordered_field,recpower,number_ring} which subsumes real, hyperreal,
  5705 rat, etc.
  5706 
  5707 * Former constant "List.op @" now named "List.append".  Use ML
  5708 antiquotations @{const_name List.append} or @{term " ... @ ... "} to
  5709 circumvent possible incompatibilities when working on ML level.
  5710 
  5711 * primrec: missing cases mapped to "undefined" instead of "arbitrary".
  5712 
  5713 * New function listsum :: 'a list => 'a for arbitrary monoids.
  5714 Special syntax: "SUM x <- xs. f x" (and latex variants)
  5715 
  5716 * New syntax for Haskell-like list comprehension (input only), eg.
  5717 [(x,y). x <- xs, y <- ys, x ~= y], see also src/HOL/List.thy.
  5718 
  5719 * The special syntax for function "filter" has changed from [x :
  5720 xs. P] to [x <- xs. P] to avoid an ambiguity caused by list
  5721 comprehension syntax, and for uniformity.  INCOMPATIBILITY.
  5722 
  5723 * [a..b] is now defined for arbitrary linear orders.  It used to be
  5724 defined on nat only, as an abbreviation for [a..<Suc b]
  5725 INCOMPATIBILITY.
  5726 
  5727 * Renamed lemma "set_take_whileD"  to "set_takeWhileD".
  5728 
  5729 * New functions "sorted" and "sort" in src/HOL/List.thy.
  5730 
  5731 * New lemma collection field_simps (an extension of ring_simps) for
  5732 manipulating (in)equations involving division. Multiplies with all
  5733 denominators that can be proved to be non-zero (in equations) or
  5734 positive/negative (in inequations).
  5735 
  5736 * Lemma collections ring_eq_simps, group_eq_simps and ring_distrib
  5737 have been improved and renamed to ring_simps, group_simps and
  5738 ring_distribs.  Removed lemmas field_xyz in theory Ring_and_Field
  5739 because they were subsumed by lemmas xyz.  INCOMPATIBILITY.
  5740 
  5741 * Theory Library/Commutative_Ring: switched from recdef to function
  5742 package; constants add, mul, pow now curried.  Infix syntax for
  5743 algebraic operations.
  5744 
  5745 * Dropped redundant lemma def_imp_eq in favor of meta_eq_to_obj_eq.
  5746 INCOMPATIBILITY.
  5747 
  5748 * Dropped redundant lemma if_def2 in favor of if_bool_eq_conj.
  5749 INCOMPATIBILITY.
  5750 
  5751 * HOL/records: generalised field-update to take a function on the
  5752 field rather than the new value: r(|A := x|) is translated to A_update
  5753 (K x) r The K-combinator that is internally used is called K_record.
  5754 INCOMPATIBILITY: Usage of the plain update functions has to be
  5755 adapted.
  5756  
  5757 * Class "semiring_0" now contains annihilation axioms x * 0 = 0 and 0
  5758 * x = 0, which are required for a semiring.  Richer structures do not
  5759 inherit from semiring_0 anymore, because this property is a theorem
  5760 there, not an axiom.  INCOMPATIBILITY: In instances of semiring_0,
  5761 there is more to prove, but this is mostly trivial.
  5762 
  5763 * Class "recpower" is generalized to arbitrary monoids, not just
  5764 commutative semirings.  INCOMPATIBILITY: may need to incorporate
  5765 commutativity or semiring properties additionally.
  5766 
  5767 * Constant "List.list_all2" in List.thy now uses authentic syntax.
  5768 INCOMPATIBILITY: translations containing list_all2 may go wrong,
  5769 better use 'abbreviation'.
  5770 
  5771 * Renamed constant "List.op mem" to "List.member".  INCOMPATIBILITY.
  5772 
  5773 * Numeral syntax: type 'bin' which was a mere type copy of 'int' has
  5774 been abandoned in favour of plain 'int'.  INCOMPATIBILITY --
  5775 significant changes for setting up numeral syntax for types:
  5776   - New constants Numeral.pred and Numeral.succ instead
  5777       of former Numeral.bin_pred and Numeral.bin_succ.
  5778   - Use integer operations instead of bin_add, bin_mult and so on.
  5779   - Numeral simplification theorems named Numeral.numeral_simps instead of Bin_simps.
  5780   - ML structure Bin_Simprocs now named Int_Numeral_Base_Simprocs.
  5781 
  5782 See src/HOL/Integ/IntArith.thy for an example setup.
  5783 
  5784 * Command 'normal_form' computes the normal form of a term that may
  5785 contain free variables.  For example ``normal_form "rev [a, b, c]"''
  5786 produces ``[b, c, a]'' (without proof).  This command is suitable for
  5787 heavy-duty computations because the functions are compiled to ML
  5788 first.  Correspondingly, a method "normalization" is provided.  See
  5789 further src/HOL/ex/NormalForm.thy and src/Tools/nbe.ML.
  5790 
  5791 * Alternative iff syntax "A <-> B" for equality on bool (with priority
  5792 25 like -->); output depends on the "iff" print_mode, the default is
  5793 "A = B" (with priority 50).
  5794 
  5795 * Relations less (<) and less_eq (<=) are also available on type bool.
  5796 Modified syntax to disallow nesting without explicit parentheses,
  5797 e.g. "(x < y) < z" or "x < (y < z)", but NOT "x < y < z".  Potential
  5798 INCOMPATIBILITY.
  5799 
  5800 * "LEAST x:A. P" expands to "LEAST x. x:A & P" (input only).
  5801 
  5802 * Relation composition operator "op O" now has precedence 75 and binds
  5803 stronger than union and intersection. INCOMPATIBILITY.
  5804 
  5805 * The old set interval syntax "{m..n(}" (and relatives) has been
  5806 removed.  Use "{m..<n}" (and relatives) instead.
  5807 
  5808 * In the context of the assumption "~(s = t)" the Simplifier rewrites
  5809 "t = s" to False (by simproc "neq").  INCOMPATIBILITY, consider using
  5810 ``declare [[simproc del: neq]]''.
  5811 
  5812 * Simplifier: "m dvd n" where m and n are numbers is evaluated to
  5813 True/False.
  5814 
  5815 * Theorem Cons_eq_map_conv no longer declared as "simp".
  5816 
  5817 * Theorem setsum_mult renamed to setsum_right_distrib.
  5818 
  5819 * Prefer ex1I over ex_ex1I in single-step reasoning, e.g. by the
  5820 ``rule'' method.
  5821 
  5822 * Reimplemented methods "sat" and "satx", with several improvements:
  5823 goals no longer need to be stated as "<prems> ==> False", equivalences
  5824 (i.e. "=" on type bool) are handled, variable names of the form
  5825 "lit_<n>" are no longer reserved, significant speedup.
  5826 
  5827 * Methods "sat" and "satx" can now replay MiniSat proof traces.
  5828 zChaff is still supported as well.
  5829 
  5830 * 'inductive' and 'datatype': provide projections of mutual rules,
  5831 bundled as foo_bar.inducts;
  5832 
  5833 * Library: moved theories Parity, GCD, Binomial, Infinite_Set to
  5834 Library.
  5835 
  5836 * Library: moved theory Accessible_Part to main HOL.
  5837 
  5838 * Library: added theory Coinductive_List of potentially infinite lists
  5839 as greatest fixed-point.
  5840 
  5841 * Library: added theory AssocList which implements (finite) maps as
  5842 association lists.
  5843 
  5844 * Method "evaluation" solves goals (i.e. a boolean expression)
  5845 efficiently by compiling it to ML.  The goal is "proved" (via an
  5846 oracle) if it evaluates to True.
  5847 
  5848 * Linear arithmetic now splits certain operators (e.g. min, max, abs)
  5849 also when invoked by the simplifier.  This results in the Simplifier
  5850 being more powerful on arithmetic goals.  INCOMPATIBILITY.
  5851 Configuration option fast_arith_split_limit=0 recovers the old
  5852 behavior.
  5853 
  5854 * Support for hex (0x20) and binary (0b1001) numerals.
  5855 
  5856 * New method: reify eqs (t), where eqs are equations for an
  5857 interpretation I :: 'a list => 'b => 'c and t::'c is an optional
  5858 parameter, computes a term s::'b and a list xs::'a list and proves the
  5859 theorem I xs s = t. This is also known as reification or quoting. The
  5860 resulting theorem is applied to the subgoal to substitute t with I xs
  5861 s.  If t is omitted, the subgoal itself is reified.
  5862 
  5863 * New method: reflection corr_thm eqs (t). The parameters eqs and (t)
  5864 are as explained above. corr_thm is a theorem for I vs (f t) = I vs t,
  5865 where f is supposed to be a computable function (in the sense of code
  5866 generattion). The method uses reify to compute s and xs as above then
  5867 applies corr_thm and uses normalization by evaluation to "prove" f s =
  5868 r and finally gets the theorem t = r, which is again applied to the
  5869 subgoal. An Example is available in src/HOL/ex/ReflectionEx.thy.
  5870 
  5871 * Reflection: Automatic reification now handels binding, an example is
  5872 available in src/HOL/ex/ReflectionEx.thy
  5873 
  5874 * HOL-Statespace: ``State Spaces: The Locale Way'' introduces a
  5875 command 'statespace' that is similar to 'record', but introduces an
  5876 abstract specification based on the locale infrastructure instead of
  5877 HOL types.  This leads to extra flexibility in composing state spaces,
  5878 in particular multiple inheritance and renaming of components.
  5879 
  5880 
  5881 *** HOL-Complex ***
  5882 
  5883 * Hyperreal: Functions root and sqrt are now defined on negative real
  5884 inputs so that root n (- x) = - root n x and sqrt (- x) = - sqrt x.
  5885 Nonnegativity side conditions have been removed from many lemmas, so
  5886 that more subgoals may now be solved by simplification; potential
  5887 INCOMPATIBILITY.
  5888 
  5889 * Real: new type classes formalize real normed vector spaces and
  5890 algebras, using new overloaded constants scaleR :: real => 'a => 'a
  5891 and norm :: 'a => real.
  5892 
  5893 * Real: constant of_real :: real => 'a::real_algebra_1 injects from
  5894 reals into other types. The overloaded constant Reals :: 'a set is now
  5895 defined as range of_real; potential INCOMPATIBILITY.
  5896 
  5897 * Real: proper support for ML code generation, including 'quickcheck'.
  5898 Reals are implemented as arbitrary precision rationals.
  5899 
  5900 * Hyperreal: Several constants that previously worked only for the
  5901 reals have been generalized, so they now work over arbitrary vector
  5902 spaces. Type annotations may need to be added in some cases; potential
  5903 INCOMPATIBILITY.
  5904 
  5905   Infinitesimal  :: ('a::real_normed_vector) star set
  5906   HFinite        :: ('a::real_normed_vector) star set
  5907   HInfinite      :: ('a::real_normed_vector) star set
  5908   approx         :: ('a::real_normed_vector) star => 'a star => bool
  5909   monad          :: ('a::real_normed_vector) star => 'a star set
  5910   galaxy         :: ('a::real_normed_vector) star => 'a star set
  5911   (NS)LIMSEQ     :: [nat => 'a::real_normed_vector, 'a] => bool
  5912   (NS)convergent :: (nat => 'a::real_normed_vector) => bool
  5913   (NS)Bseq       :: (nat => 'a::real_normed_vector) => bool
  5914   (NS)Cauchy     :: (nat => 'a::real_normed_vector) => bool
  5915   (NS)LIM        :: ['a::real_normed_vector => 'b::real_normed_vector, 'a, 'b] => bool
  5916   is(NS)Cont     :: ['a::real_normed_vector => 'b::real_normed_vector, 'a] => bool
  5917   deriv          :: ['a::real_normed_field => 'a, 'a, 'a] => bool
  5918   sgn            :: 'a::real_normed_vector => 'a
  5919   exp            :: 'a::{recpower,real_normed_field,banach} => 'a
  5920 
  5921 * Complex: Some complex-specific constants are now abbreviations for
  5922 overloaded ones: complex_of_real = of_real, cmod = norm, hcmod =
  5923 hnorm.  Other constants have been entirely removed in favor of the
  5924 polymorphic versions (INCOMPATIBILITY):
  5925 
  5926   approx        <-- capprox
  5927   HFinite       <-- CFinite
  5928   HInfinite     <-- CInfinite
  5929   Infinitesimal <-- CInfinitesimal
  5930   monad         <-- cmonad
  5931   galaxy        <-- cgalaxy
  5932   (NS)LIM       <-- (NS)CLIM, (NS)CRLIM
  5933   is(NS)Cont    <-- is(NS)Contc, is(NS)contCR
  5934   (ns)deriv     <-- (ns)cderiv
  5935 
  5936 
  5937 *** HOL-Algebra ***
  5938 
  5939 * Formalisation of ideals and the quotient construction over rings.
  5940 
  5941 * Order and lattice theory no longer based on records.
  5942 INCOMPATIBILITY.
  5943 
  5944 * Renamed lemmas least_carrier -> least_closed and greatest_carrier ->
  5945 greatest_closed.  INCOMPATIBILITY.
  5946 
  5947 * Method algebra is now set up via an attribute.  For examples see
  5948 Ring.thy.  INCOMPATIBILITY: the method is now weaker on combinations
  5949 of algebraic structures.
  5950 
  5951 * Renamed theory CRing to Ring.
  5952 
  5953 
  5954 *** HOL-Nominal ***
  5955 
  5956 * Substantial, yet incomplete support for nominal datatypes (binding
  5957 structures) based on HOL-Nominal logic.  See src/HOL/Nominal and
  5958 src/HOL/Nominal/Examples.  Prospective users should consult
  5959 http://isabelle.in.tum.de/nominal/
  5960 
  5961 
  5962 *** ML ***
  5963 
  5964 * ML basics: just one true type int, which coincides with IntInf.int
  5965 (even on SML/NJ).
  5966 
  5967 * ML within Isar: antiquotations allow to embed statically-checked
  5968 formal entities in the source, referring to the context available at
  5969 compile-time.  For example:
  5970 
  5971 ML {* @{sort "{zero,one}"} *}
  5972 ML {* @{typ "'a => 'b"} *}
  5973 ML {* @{term "%x. x"} *}
  5974 ML {* @{prop "x == y"} *}
  5975 ML {* @{ctyp "'a => 'b"} *}
  5976 ML {* @{cterm "%x. x"} *}
  5977 ML {* @{cprop "x == y"} *}
  5978 ML {* @{thm asm_rl} *}
  5979 ML {* @{thms asm_rl} *}
  5980 ML {* @{type_name c} *}
  5981 ML {* @{type_syntax c} *}
  5982 ML {* @{const_name c} *}
  5983 ML {* @{const_syntax c} *}
  5984 ML {* @{context} *}
  5985 ML {* @{theory} *}
  5986 ML {* @{theory Pure} *}
  5987 ML {* @{theory_ref} *}
  5988 ML {* @{theory_ref Pure} *}
  5989 ML {* @{simpset} *}
  5990 ML {* @{claset} *}
  5991 ML {* @{clasimpset} *}
  5992 
  5993 The same works for sources being ``used'' within an Isar context.
  5994 
  5995 * ML in Isar: improved error reporting; extra verbosity with
  5996 ML_Context.trace enabled.
  5997 
  5998 * Pure/General/table.ML: the join operations now works via exceptions
  5999 DUP/SAME instead of type option. This is simpler in simple cases, and
  6000 admits slightly more efficient complex applications.
  6001 
  6002 * Pure: 'advanced' translation functions (parse_translation etc.) now
  6003 use Context.generic instead of just theory.
  6004 
  6005 * Pure: datatype Context.generic joins theory/Proof.context and
  6006 provides some facilities for code that works in either kind of
  6007 context, notably GenericDataFun for uniform theory and proof data.
  6008 
  6009 * Pure: simplified internal attribute type, which is now always
  6010 Context.generic * thm -> Context.generic * thm. Global (theory) vs.
  6011 local (Proof.context) attributes have been discontinued, while
  6012 minimizing code duplication. Thm.rule_attribute and
  6013 Thm.declaration_attribute build canonical attributes; see also structure
  6014 Context for further operations on Context.generic, notably
  6015 GenericDataFun. INCOMPATIBILITY, need to adapt attribute type
  6016 declarations and definitions.
  6017 
  6018 * Context data interfaces (Theory/Proof/GenericDataFun): removed
  6019 name/print, uninitialized data defaults to ad-hoc copy of empty value,
  6020 init only required for impure data. INCOMPATIBILITY: empty really need
  6021 to be empty (no dependencies on theory content!)
  6022 
  6023 * Pure/kernel: consts certification ignores sort constraints given in
  6024 signature declarations. (This information is not relevant to the
  6025 logic, but only for type inference.) SIGNIFICANT INTERNAL CHANGE,
  6026 potential INCOMPATIBILITY.
  6027 
  6028 * Pure: axiomatic type classes are now purely definitional, with
  6029 explicit proofs of class axioms and super class relations performed
  6030 internally. See Pure/axclass.ML for the main internal interfaces --
  6031 notably AxClass.define_class supercedes AxClass.add_axclass, and
  6032 AxClass.axiomatize_class/classrel/arity supersede
  6033 Sign.add_classes/classrel/arities.
  6034 
  6035 * Pure/Isar: Args/Attrib parsers operate on Context.generic --
  6036 global/local versions on theory vs. Proof.context have been
  6037 discontinued; Attrib.syntax and Method.syntax have been adapted
  6038 accordingly.  INCOMPATIBILITY, need to adapt parser expressions for
  6039 attributes, methods, etc.
  6040 
  6041 * Pure: several functions of signature "... -> theory -> theory * ..."
  6042 have been reoriented to "... -> theory -> ... * theory" in order to
  6043 allow natural usage in combination with the ||>, ||>>, |-> and
  6044 fold_map combinators.
  6045 
  6046 * Pure: official theorem names (closed derivations) and additional
  6047 comments (tags) are now strictly separate.  Name hints -- which are
  6048 maintained as tags -- may be attached any time without affecting the
  6049 derivation.
  6050 
  6051 * Pure: primitive rule lift_rule now takes goal cterm instead of an
  6052 actual goal state (thm).  Use Thm.lift_rule (Thm.cprem_of st i) to
  6053 achieve the old behaviour.
  6054 
  6055 * Pure: the "Goal" constant is now called "prop", supporting a
  6056 slightly more general idea of ``protecting'' meta-level rule
  6057 statements.
  6058 
  6059 * Pure: Logic.(un)varify only works in a global context, which is now
  6060 enforced instead of silently assumed.  INCOMPATIBILITY, may use
  6061 Logic.legacy_(un)varify as temporary workaround.
  6062 
  6063 * Pure: structure Name provides scalable operations for generating
  6064 internal variable names, notably Name.variants etc.  This replaces
  6065 some popular functions from term.ML:
  6066 
  6067   Term.variant		->  Name.variant
  6068   Term.variantlist	->  Name.variant_list
  6069   Term.invent_names	->  Name.invent_list
  6070 
  6071 Note that low-level renaming rarely occurs in new code -- operations
  6072 from structure Variable are used instead (see below).
  6073 
  6074 * Pure: structure Variable provides fundamental operations for proper
  6075 treatment of fixed/schematic variables in a context.  For example,
  6076 Variable.import introduces fixes for schematics of given facts and
  6077 Variable.export reverses the effect (up to renaming) -- this replaces
  6078 various freeze_thaw operations.
  6079 
  6080 * Pure: structure Goal provides simple interfaces for
  6081 init/conclude/finish and tactical prove operations (replacing former
  6082 Tactic.prove).  Goal.prove is the canonical way to prove results
  6083 within a given context; Goal.prove_global is a degraded version for
  6084 theory level goals, including a global Drule.standard.  Note that
  6085 OldGoals.prove_goalw_cterm has long been obsolete, since it is
  6086 ill-behaved in a local proof context (e.g. with local fixes/assumes or
  6087 in a locale context).
  6088 
  6089 * Pure/Syntax: generic interfaces for parsing (Syntax.parse_term etc.)
  6090 and type checking (Syntax.check_term etc.), with common combinations
  6091 (Syntax.read_term etc.). These supersede former Sign.read_term etc.
  6092 which are considered legacy and await removal.
  6093 
  6094 * Pure/Syntax: generic interfaces for type unchecking
  6095 (Syntax.uncheck_terms etc.) and unparsing (Syntax.unparse_term etc.),
  6096 with common combinations (Syntax.pretty_term, Syntax.string_of_term
  6097 etc.).  Former Sign.pretty_term, Sign.string_of_term etc. are still
  6098 available for convenience, but refer to the very same operations using
  6099 a mere theory instead of a full context.
  6100 
  6101 * Isar: simplified treatment of user-level errors, using exception
  6102 ERROR of string uniformly.  Function error now merely raises ERROR,
  6103 without any side effect on output channels.  The Isar toplevel takes
  6104 care of proper display of ERROR exceptions.  ML code may use plain
  6105 handle/can/try; cat_error may be used to concatenate errors like this:
  6106 
  6107   ... handle ERROR msg => cat_error msg "..."
  6108 
  6109 Toplevel ML code (run directly or through the Isar toplevel) may be
  6110 embedded into the Isar toplevel with exception display/debug like
  6111 this:
  6112 
  6113   Isar.toplevel (fn () => ...)
  6114 
  6115 INCOMPATIBILITY, removed special transform_error facilities, removed
  6116 obsolete variants of user-level exceptions (ERROR_MESSAGE,
  6117 Context.PROOF, ProofContext.CONTEXT, Proof.STATE, ProofHistory.FAIL)
  6118 -- use plain ERROR instead.
  6119 
  6120 * Isar: theory setup now has type (theory -> theory), instead of a
  6121 list.  INCOMPATIBILITY, may use #> to compose setup functions.
  6122 
  6123 * Isar: ML toplevel pretty printer for type Proof.context, subject to
  6124 ProofContext.debug/verbose flags.
  6125 
  6126 * Isar: Toplevel.theory_to_proof admits transactions that modify the
  6127 theory before entering a proof state.  Transactions now always see a
  6128 quasi-functional intermediate checkpoint, both in interactive and
  6129 batch mode.
  6130 
  6131 * Isar: simplified interfaces for outer syntax.  Renamed
  6132 OuterSyntax.add_keywords to OuterSyntax.keywords.  Removed
  6133 OuterSyntax.add_parsers -- this functionality is now included in
  6134 OuterSyntax.command etc.  INCOMPATIBILITY.
  6135 
  6136 * Simplifier: the simpset of a running simplification process now
  6137 contains a proof context (cf. Simplifier.the_context), which is the
  6138 very context that the initial simpset has been retrieved from (by
  6139 simpset_of/local_simpset_of).  Consequently, all plug-in components
  6140 (solver, looper etc.) may depend on arbitrary proof data.
  6141 
  6142 * Simplifier.inherit_context inherits the proof context (plus the
  6143 local bounds) of the current simplification process; any simproc
  6144 etc. that calls the Simplifier recursively should do this!  Removed
  6145 former Simplifier.inherit_bounds, which is already included here --
  6146 INCOMPATIBILITY.  Tools based on low-level rewriting may even have to
  6147 specify an explicit context using Simplifier.context/theory_context.
  6148 
  6149 * Simplifier/Classical Reasoner: more abstract interfaces
  6150 change_simpset/claset for modifying the simpset/claset reference of a
  6151 theory; raw versions simpset/claset_ref etc. have been discontinued --
  6152 INCOMPATIBILITY.
  6153 
  6154 * Provers: more generic wrt. syntax of object-logics, avoid hardwired
  6155 "Trueprop" etc.
  6156 
  6157 
  6158 *** System ***
  6159 
  6160 * settings: the default heap location within ISABELLE_HOME_USER now
  6161 includes ISABELLE_IDENTIFIER.  This simplifies use of multiple
  6162 Isabelle installations.
  6163 
  6164 * isabelle-process: option -S (secure mode) disables some critical
  6165 operations, notably runtime compilation and evaluation of ML source
  6166 code.
  6167 
  6168 * Basic Isabelle mode for jEdit, see Isabelle/lib/jedit/.
  6169 
  6170 * Support for parallel execution, using native multicore support of
  6171 Poly/ML 5.1.  The theory loader exploits parallelism when processing
  6172 independent theories, according to the given theory header
  6173 specifications. The maximum number of worker threads is specified via
  6174 usedir option -M or the "max-threads" setting in Proof General. A
  6175 speedup factor of 1.5--3.5 can be expected on a 4-core machine, and up
  6176 to 6 on a 8-core machine.  User-code needs to observe certain
  6177 guidelines for thread-safe programming, see appendix A in the Isar
  6178 Implementation manual.
  6179 
  6180 
  6181 
  6182 New in Isabelle2005 (October 2005)
  6183 ----------------------------------
  6184 
  6185 *** General ***
  6186 
  6187 * Theory headers: the new header syntax for Isar theories is
  6188 
  6189   theory <name>
  6190   imports <theory1> ... <theoryN>
  6191   uses <file1> ... <fileM>
  6192   begin
  6193 
  6194 where the 'uses' part is optional.  The previous syntax
  6195 
  6196   theory <name> = <theory1> + ... + <theoryN>:
  6197 
  6198 will disappear in the next release.  Use isatool fixheaders to convert
  6199 existing theory files.  Note that there is no change in ancient
  6200 non-Isar theories now, but these will disappear soon.
  6201 
  6202 * Theory loader: parent theories can now also be referred to via
  6203 relative and absolute paths.
  6204 
  6205 * Command 'find_theorems' searches for a list of criteria instead of a
  6206 list of constants. Known criteria are: intro, elim, dest, name:string,
  6207 simp:term, and any term. Criteria can be preceded by '-' to select
  6208 theorems that do not match. Intro, elim, dest select theorems that
  6209 match the current goal, name:s selects theorems whose fully qualified
  6210 name contain s, and simp:term selects all simplification rules whose
  6211 lhs match term.  Any other term is interpreted as pattern and selects
  6212 all theorems matching the pattern. Available in ProofGeneral under
  6213 'ProofGeneral -> Find Theorems' or C-c C-f.  Example:
  6214 
  6215   C-c C-f (100) "(_::nat) + _ + _" intro -name: "HOL."
  6216 
  6217 prints the last 100 theorems matching the pattern "(_::nat) + _ + _",
  6218 matching the current goal as introduction rule and not having "HOL."
  6219 in their name (i.e. not being defined in theory HOL).
  6220 
  6221 * Command 'thms_containing' has been discontinued in favour of
  6222 'find_theorems'; INCOMPATIBILITY.
  6223 
  6224 * Communication with Proof General is now 8bit clean, which means that
  6225 Unicode text in UTF-8 encoding may be used within theory texts (both
  6226 formal and informal parts).  Cf. option -U of the Isabelle Proof
  6227 General interface.  Here are some simple examples (cf. src/HOL/ex):
  6228 
  6229   http://isabelle.in.tum.de/library/HOL/ex/Hebrew.html
  6230   http://isabelle.in.tum.de/library/HOL/ex/Chinese.html
  6231 
  6232 * Improved efficiency of the Simplifier and, to a lesser degree, the
  6233 Classical Reasoner.  Typical big applications run around 2 times
  6234 faster.
  6235 
  6236 
  6237 *** Document preparation ***
  6238 
  6239 * Commands 'display_drafts' and 'print_drafts' perform simple output
  6240 of raw sources.  Only those symbols that do not require additional