NEWS
author wenzelm
Mon Sep 03 11:09:25 2012 +0200 (2012-09-03)
changeset 49072 747835eb2782
parent 49000 0cebcbeac4c7
child 49145 0ee5983e3d59
permissions -rw-r--r--
"isabelle logo" produces EPS and PDF format simultaneously;
more robust invocation of epstopdf: avoid filter mode;
     1 Isabelle NEWS -- history user-relevant changes
     2 ==============================================
     3 
     4 New in this Isabelle version
     5 ----------------------------
     6 
     7 *** General ***
     8 
     9 * Command 'ML_file' evaluates ML text from a file directly within the
    10 theory, without any predeclaration via 'uses' in the theory header.
    11 
    12 * Discontinued obsolete method fastsimp / tactic fast_simp_tac, which
    13 is called fastforce / fast_force_tac already since Isabelle2011-1.
    14 
    15 * Updated and extended "isar-ref" manual, reduced remaining material
    16 in old "ref" manual.
    17 
    18 
    19 *** Pure ***
    20 
    21 * Code generation for Haskell: restrict unqualified imports from
    22 Haskell Prelude to a small set of fundamental operations.
    23 
    24 * Command "export_code": relative file names are interpreted
    25 relatively to master directory of current theory rather than
    26 the rather arbitrary current working directory.
    27 INCOMPATIBILITY.
    28 
    29 * Discontinued obsolete attribute "COMP".  Potential INCOMPATIBILITY,
    30 use regular rule composition via "OF" / "THEN", or explicit proof
    31 structure instead.  Note that Isabelle/ML provides a variety of
    32 operators like COMP, INCR_COMP, COMP_INCR, which need to be applied
    33 with some care where this is really required.
    34 
    35 * Command 'typ' supports an additional variant with explicit sort
    36 constraint, to infer and check the most general type conforming to a
    37 given given sort.  Example (in HOL):
    38 
    39   typ "_ * _ * bool * unit" :: finite
    40 
    41 
    42 *** HOL ***
    43 
    44 * HOL/Codatatype: New (co)datatype package with support for mixed,
    45 nested recursion and interesting non-free datatypes.
    46 
    47 * HOL/Ordinals_and_Cardinals: Theories of ordinals and cardinals
    48 (supersedes the AFP entry of the same name).
    49 
    50 * Library/Debug.thy and Library/Parallel.thy: debugging and parallel
    51 execution for code generated towards Isabelle/ML.
    52 
    53 * Simproc "finite_Collect" rewrites set comprehensions into pointfree
    54 expressions.
    55 
    56 * Quickcheck:
    57 
    58   - added an optimisation for equality premises.
    59     It is switched on by default, and can be switched off by setting
    60     the configuration quickcheck_optimise_equality to false.
    61 
    62 * The SMT solver Z3 has now by default a restricted set of directly
    63 supported features. For the full set of features (div/mod, nonlinear
    64 arithmetic, datatypes/records) with potential proof reconstruction
    65 failures, enable the configuration option "z3_with_extensions".
    66 Minor INCOMPATIBILITY.
    67 
    68 * Sledgehammer:
    69 
    70   - Added MaSh relevance filter based on machine-learning; see the
    71     Sledgehammer manual for details.
    72   - Rationalized type encodings ("type_enc" option).
    73   - Renamed options:
    74       max_relevant ~> max_facts
    75       relevance_thresholds ~> fact_thresholds
    76 
    77 
    78 *** Document preparation ***
    79 
    80 * Default for \<euro> is now based on eurosym package, instead of
    81 slightly exotic babel/greek.
    82 
    83 * Document variant NAME may use different LaTeX entry point
    84 document/root_NAME.tex if that file exists, instead of the common
    85 document/root.tex.
    86 
    87 * Simplified custom document/build script, instead of old-style
    88 document/IsaMakefile.  Minor INCOMPATIBILITY.
    89 
    90 
    91 *** ML ***
    92 
    93 * Renamed Position.str_of to Position.here to emphasize that this is a
    94 formal device to inline positions into message text, but not
    95 necessarily printing visible text.
    96 
    97 
    98 *** System ***
    99 
   100 * The ML system is configured as regular component, and no longer
   101 picked up from some surrounding directory.  Potential INCOMPATIBILITY
   102 for home-made configurations.
   103 
   104 * The "isabelle logo" tool produces EPS and PDF format simultaneously.
   105 Minor INCOMPATIBILITY in command-line options.
   106 
   107 * Advanced support for Isabelle sessions and build management, see
   108 "system" manual for the chapter of that name, especially the "isabelle
   109 build" tool and its examples.  INCOMPATIBILITY, isabelle usedir /
   110 mkdir / make are rendered obsolete.
   111 
   112 * Discontinued obsolete "isabelle makeall".
   113 
   114 * Discontinued obsolete IsaMakefile and ROOT.ML files from the
   115 Isabelle distribution, except for rudimentary src/HOL/IsaMakefile that
   116 provides some traditional targets that invoke "isabelle build".  Note
   117 that this is inefficient!  Applications of Isabelle/HOL involving
   118 "isabelle make" should be upgraded to use "isabelle build" directly.
   119 
   120 * Discontinued obsolete Isabelle/build script, it is superseded by the
   121 regular isabelle build tool.  For example:
   122 
   123   isabelle build -s -b HOLCF
   124 
   125 * The "isabelle options" tool prints Isabelle system options, as
   126 required for "isabelle build", for example.
   127 
   128 * The "isabelle mkroot" tool prepares session root directories for use
   129 with "isabelle build", similar to former "isabelle mkdir" for
   130 "isabelle usedir".
   131 
   132 * The "isabelle components" tool helps to resolve add-on components
   133 that are not bundled, or referenced from a bare-bones repository
   134 version of Isabelle.
   135 
   136 * Discontinued support for Poly/ML 5.2.1, which was the last version
   137 without exception positions and advanced ML compiler/toplevel
   138 configuration.
   139 
   140 * Discontinued special treatment of Proof General -- no longer guess
   141 PROOFGENERAL_HOME based on accidental file-system layout.  Minor
   142 INCOMPATIBILITY: provide PROOFGENERAL_HOME and PROOFGENERAL_OPTIONS
   143 settings manually, or use a Proof General version that has been
   144 bundled as Isabelle component.
   145 
   146 
   147 New in Isabelle2012 (May 2012)
   148 ------------------------------
   149 
   150 *** General ***
   151 
   152 * Prover IDE (PIDE) improvements:
   153 
   154   - more robust Sledgehammer integration (as before the sledgehammer
   155     command-line needs to be typed into the source buffer)
   156   - markup for bound variables
   157   - markup for types of term variables (displayed as tooltips)
   158   - support for user-defined Isar commands within the running session
   159   - improved support for Unicode outside original 16bit range
   160     e.g. glyph for \<A> (thanks to jEdit 4.5.1)
   161 
   162 * Forward declaration of outer syntax keywords within the theory
   163 header -- minor INCOMPATIBILITY for user-defined commands.  Allow new
   164 commands to be used in the same theory where defined.
   165 
   166 * Auxiliary contexts indicate block structure for specifications with
   167 additional parameters and assumptions.  Such unnamed contexts may be
   168 nested within other targets, like 'theory', 'locale', 'class',
   169 'instantiation' etc.  Results from the local context are generalized
   170 accordingly and applied to the enclosing target context.  Example:
   171 
   172   context
   173     fixes x y z :: 'a
   174     assumes xy: "x = y" and yz: "y = z"
   175   begin
   176 
   177   lemma my_trans: "x = z" using xy yz by simp
   178 
   179   end
   180 
   181   thm my_trans
   182 
   183 The most basic application is to factor-out context elements of
   184 several fixes/assumes/shows theorem statements, e.g. see
   185 ~~/src/HOL/Isar_Examples/Group_Context.thy
   186 
   187 Any other local theory specification element works within the "context
   188 ... begin ... end" block as well.
   189 
   190 * Bundled declarations associate attributed fact expressions with a
   191 given name in the context.  These may be later included in other
   192 contexts.  This allows to manage context extensions casually, without
   193 the logical dependencies of locales and locale interpretation.  See
   194 commands 'bundle', 'include', 'including' etc. in the isar-ref manual.
   195 
   196 * Commands 'lemmas' and 'theorems' allow local variables using 'for'
   197 declaration, and results are standardized before being stored.  Thus
   198 old-style "standard" after instantiation or composition of facts
   199 becomes obsolete.  Minor INCOMPATIBILITY, due to potential change of
   200 indices of schematic variables.
   201 
   202 * Rule attributes in local theory declarations (e.g. locale or class)
   203 are now statically evaluated: the resulting theorem is stored instead
   204 of the original expression.  INCOMPATIBILITY in rare situations, where
   205 the historic accident of dynamic re-evaluation in interpretations
   206 etc. was exploited.
   207 
   208 * New tutorial "Programming and Proving in Isabelle/HOL"
   209 ("prog-prove").  It completely supersedes "A Tutorial Introduction to
   210 Structured Isar Proofs" ("isar-overview"), which has been removed.  It
   211 also supersedes "Isabelle/HOL, A Proof Assistant for Higher-Order
   212 Logic" as the recommended beginners tutorial, but does not cover all
   213 of the material of that old tutorial.
   214 
   215 * Updated and extended reference manuals: "isar-ref",
   216 "implementation", "system"; reduced remaining material in old "ref"
   217 manual.
   218 
   219 
   220 *** Pure ***
   221 
   222 * Command 'definition' no longer exports the foundational "raw_def"
   223 into the user context.  Minor INCOMPATIBILITY, may use the regular
   224 "def" result with attribute "abs_def" to imitate the old version.
   225 
   226 * Attribute "abs_def" turns an equation of the form "f x y == t" into
   227 "f == %x y. t", which ensures that "simp" or "unfold" steps always
   228 expand it.  This also works for object-logic equality.  (Formerly
   229 undocumented feature.)
   230 
   231 * Sort constraints are now propagated in simultaneous statements, just
   232 like type constraints.  INCOMPATIBILITY in rare situations, where
   233 distinct sorts used to be assigned accidentally.  For example:
   234 
   235   lemma "P (x::'a::foo)" and "Q (y::'a::bar)"  -- "now illegal"
   236 
   237   lemma "P (x::'a)" and "Q (y::'a::bar)"
   238     -- "now uniform 'a::bar instead of default sort for first occurrence (!)"
   239 
   240 * Rule composition via attribute "OF" (or ML functions OF/MRS) is more
   241 tolerant against multiple unifiers, as long as the final result is
   242 unique.  (As before, rules are composed in canonical right-to-left
   243 order to accommodate newly introduced premises.)
   244 
   245 * Renamed some inner syntax categories:
   246 
   247     num ~> num_token
   248     xnum ~> xnum_token
   249     xstr ~> str_token
   250 
   251 Minor INCOMPATIBILITY.  Note that in practice "num_const" or
   252 "num_position" etc. are mainly used instead (which also include
   253 position information via constraints).
   254 
   255 * Simplified configuration options for syntax ambiguity: see
   256 "syntax_ambiguity_warning" and "syntax_ambiguity_limit" in isar-ref
   257 manual.  Minor INCOMPATIBILITY.
   258 
   259 * Discontinued configuration option "syntax_positions": atomic terms
   260 in parse trees are always annotated by position constraints.
   261 
   262 * Old code generator for SML and its commands 'code_module',
   263 'code_library', 'consts_code', 'types_code' have been discontinued.
   264 Use commands of the generic code generator instead.  INCOMPATIBILITY.
   265 
   266 * Redundant attribute "code_inline" has been discontinued. Use
   267 "code_unfold" instead.  INCOMPATIBILITY.
   268 
   269 * Dropped attribute "code_unfold_post" in favor of the its dual
   270 "code_abbrev", which yields a common pattern in definitions like
   271 
   272   definition [code_abbrev]: "f = t"
   273 
   274 INCOMPATIBILITY.
   275 
   276 * Obsolete 'types' command has been discontinued.  Use 'type_synonym'
   277 instead.  INCOMPATIBILITY.
   278 
   279 * Discontinued old "prems" fact, which used to refer to the accidental
   280 collection of foundational premises in the context (already marked as
   281 legacy since Isabelle2011).
   282 
   283 
   284 *** HOL ***
   285 
   286 * Type 'a set is now a proper type constructor (just as before
   287 Isabelle2008).  Definitions mem_def and Collect_def have disappeared.
   288 Non-trivial INCOMPATIBILITY.  For developments keeping predicates and
   289 sets separate, it is often sufficient to rephrase some set S that has
   290 been accidentally used as predicates by "%x. x : S", and some
   291 predicate P that has been accidentally used as set by "{x. P x}".
   292 Corresponding proofs in a first step should be pruned from any
   293 tinkering with former theorems mem_def and Collect_def as far as
   294 possible.
   295 
   296 For developments which deliberately mix predicates and sets, a
   297 planning step is necessary to determine what should become a predicate
   298 and what a set.  It can be helpful to carry out that step in
   299 Isabelle2011-1 before jumping right into the current release.
   300 
   301 * Code generation by default implements sets as container type rather
   302 than predicates.  INCOMPATIBILITY.
   303 
   304 * New type synonym 'a rel = ('a * 'a) set
   305 
   306 * The representation of numerals has changed.  Datatype "num"
   307 represents strictly positive binary numerals, along with functions
   308 "numeral :: num => 'a" and "neg_numeral :: num => 'a" to represent
   309 positive and negated numeric literals, respectively.  See also
   310 definitions in ~~/src/HOL/Num.thy.  Potential INCOMPATIBILITY, some
   311 user theories may require adaptations as follows:
   312 
   313   - Theorems with number_ring or number_semiring constraints: These
   314     classes are gone; use comm_ring_1 or comm_semiring_1 instead.
   315 
   316   - Theories defining numeric types: Remove number, number_semiring,
   317     and number_ring instances. Defer all theorems about numerals until
   318     after classes one and semigroup_add have been instantiated.
   319 
   320   - Numeral-only simp rules: Replace each rule having a "number_of v"
   321     pattern with two copies, one for numeral and one for neg_numeral.
   322 
   323   - Theorems about subclasses of semiring_1 or ring_1: These classes
   324     automatically support numerals now, so more simp rules and
   325     simprocs may now apply within the proof.
   326 
   327   - Definitions and theorems using old constructors Pls/Min/Bit0/Bit1:
   328     Redefine using other integer operations.
   329 
   330 * Transfer: New package intended to generalize the existing
   331 "descending" method and related theorem attributes from the Quotient
   332 package.  (Not all functionality is implemented yet, but future
   333 development will focus on Transfer as an eventual replacement for the
   334 corresponding parts of the Quotient package.)
   335 
   336   - transfer_rule attribute: Maintains a collection of transfer rules,
   337     which relate constants at two different types. Transfer rules may
   338     relate different type instances of the same polymorphic constant,
   339     or they may relate an operation on a raw type to a corresponding
   340     operation on an abstract type (quotient or subtype). For example:
   341 
   342     ((A ===> B) ===> list_all2 A ===> list_all2 B) map map
   343     (cr_int ===> cr_int ===> cr_int) (%(x,y) (u,v). (x+u, y+v)) plus_int
   344 
   345   - transfer method: Replaces a subgoal on abstract types with an
   346     equivalent subgoal on the corresponding raw types. Constants are
   347     replaced with corresponding ones according to the transfer rules.
   348     Goals are generalized over all free variables by default; this is
   349     necessary for variables whose types change, but can be overridden
   350     for specific variables with e.g. "transfer fixing: x y z".  The
   351     variant transfer' method allows replacing a subgoal with one that
   352     is logically stronger (rather than equivalent).
   353 
   354   - relator_eq attribute: Collects identity laws for relators of
   355     various type constructors, e.g. "list_all2 (op =) = (op =)".  The
   356     transfer method uses these lemmas to infer transfer rules for
   357     non-polymorphic constants on the fly.
   358 
   359   - transfer_prover method: Assists with proving a transfer rule for a
   360     new constant, provided the constant is defined in terms of other
   361     constants that already have transfer rules. It should be applied
   362     after unfolding the constant definitions.
   363 
   364   - HOL/ex/Transfer_Int_Nat.thy: Example theory demonstrating transfer
   365     from type nat to type int.
   366 
   367 * Lifting: New package intended to generalize the quotient_definition
   368 facility of the Quotient package; designed to work with Transfer.
   369 
   370   - lift_definition command: Defines operations on an abstract type in
   371     terms of a corresponding operation on a representation
   372     type.  Example syntax:
   373 
   374     lift_definition dlist_insert :: "'a => 'a dlist => 'a dlist"
   375       is List.insert
   376 
   377     Users must discharge a respectfulness proof obligation when each
   378     constant is defined. (For a type copy, i.e. a typedef with UNIV,
   379     the proof is discharged automatically.) The obligation is
   380     presented in a user-friendly, readable form; a respectfulness
   381     theorem in the standard format and a transfer rule are generated
   382     by the package.
   383 
   384   - Integration with code_abstype: For typedefs (e.g. subtypes
   385     corresponding to a datatype invariant, such as dlist),
   386     lift_definition generates a code certificate theorem and sets up
   387     code generation for each constant.
   388 
   389   - setup_lifting command: Sets up the Lifting package to work with a
   390     user-defined type. The user must provide either a quotient theorem
   391     or a type_definition theorem.  The package configures transfer
   392     rules for equality and quantifiers on the type, and sets up the
   393     lift_definition command to work with the type.
   394 
   395   - Usage examples: See Quotient_Examples/Lift_DList.thy,
   396     Quotient_Examples/Lift_RBT.thy, Quotient_Examples/Lift_FSet.thy,
   397     Word/Word.thy and Library/Float.thy.
   398 
   399 * Quotient package:
   400 
   401   - The 'quotient_type' command now supports a 'morphisms' option with
   402     rep and abs functions, similar to typedef.
   403 
   404   - 'quotient_type' sets up new types to work with the Lifting and
   405     Transfer packages, as with 'setup_lifting'.
   406 
   407   - The 'quotient_definition' command now requires the user to prove a
   408     respectfulness property at the point where the constant is
   409     defined, similar to lift_definition; INCOMPATIBILITY.
   410 
   411   - Renamed predicate 'Quotient' to 'Quotient3', and renamed theorems
   412     accordingly, INCOMPATIBILITY.
   413 
   414 * New diagnostic command 'find_unused_assms' to find potentially
   415 superfluous assumptions in theorems using Quickcheck.
   416 
   417 * Quickcheck:
   418 
   419   - Quickcheck returns variable assignments as counterexamples, which
   420     allows to reveal the underspecification of functions under test.
   421     For example, refuting "hd xs = x", it presents the variable
   422     assignment xs = [] and x = a1 as a counterexample, assuming that
   423     any property is false whenever "hd []" occurs in it.
   424 
   425     These counterexample are marked as potentially spurious, as
   426     Quickcheck also returns "xs = []" as a counterexample to the
   427     obvious theorem "hd xs = hd xs".
   428 
   429     After finding a potentially spurious counterexample, Quickcheck
   430     continues searching for genuine ones.
   431 
   432     By default, Quickcheck shows potentially spurious and genuine
   433     counterexamples. The option "genuine_only" sets quickcheck to only
   434     show genuine counterexamples.
   435 
   436   - The command 'quickcheck_generator' creates random and exhaustive
   437     value generators for a given type and operations.
   438 
   439     It generates values by using the operations as if they were
   440     constructors of that type.
   441 
   442   - Support for multisets.
   443 
   444   - Added "use_subtype" options.
   445 
   446   - Added "quickcheck_locale" configuration to specify how to process
   447     conjectures in a locale context.
   448 
   449 * Nitpick: Fixed infinite loop caused by the 'peephole_optim' option
   450 and affecting 'rat' and 'real'.
   451 
   452 * Sledgehammer:
   453   - Integrated more tightly with SPASS, as described in the ITP 2012
   454     paper "More SPASS with Isabelle".
   455   - Made it try "smt" as a fallback if "metis" fails or times out.
   456   - Added support for the following provers: Alt-Ergo (via Why3 and
   457     TFF1), iProver, iProver-Eq.
   458   - Sped up the minimizer.
   459   - Added "lam_trans", "uncurry_aliases", and "minimize" options.
   460   - Renamed "slicing" ("no_slicing") option to "slice" ("dont_slice").
   461   - Renamed "sound" option to "strict".
   462 
   463 * Metis: Added possibility to specify lambda translations scheme as a
   464 parenthesized argument (e.g., "by (metis (lifting) ...)").
   465 
   466 * SMT: Renamed "smt_fixed" option to "smt_read_only_certificates".
   467 
   468 * Command 'try0': Renamed from 'try_methods'. INCOMPATIBILITY.
   469 
   470 * New "case_product" attribute to generate a case rule doing multiple
   471 case distinctions at the same time.  E.g.
   472 
   473   list.exhaust [case_product nat.exhaust]
   474 
   475 produces a rule which can be used to perform case distinction on both
   476 a list and a nat.
   477 
   478 * New "eventually_elim" method as a generalized variant of the
   479 eventually_elim* rules.  Supports structured proofs.
   480 
   481 * Typedef with implicit set definition is considered legacy.  Use
   482 "typedef (open)" form instead, which will eventually become the
   483 default.
   484 
   485 * Record: code generation can be switched off manually with
   486 
   487   declare [[record_coden = false]]  -- "default true"
   488 
   489 * Datatype: type parameters allow explicit sort constraints.
   490 
   491 * Concrete syntax for case expressions includes constraints for source
   492 positions, and thus produces Prover IDE markup for its bindings.
   493 INCOMPATIBILITY for old-style syntax translations that augment the
   494 pattern notation; e.g. see src/HOL/HOLCF/One.thy for translations of
   495 one_case.
   496 
   497 * Clarified attribute "mono_set": pure declaration without modifying
   498 the result of the fact expression.
   499 
   500 * More default pred/set conversions on a couple of relation operations
   501 and predicates.  Added powers of predicate relations.  Consolidation
   502 of some relation theorems:
   503 
   504   converse_def ~> converse_unfold
   505   rel_comp_def ~> relcomp_unfold
   506   symp_def ~> (modified, use symp_def and sym_def instead)
   507   transp_def ~> transp_trans
   508   Domain_def ~> Domain_unfold
   509   Range_def ~> Domain_converse [symmetric]
   510 
   511 Generalized theorems INF_INT_eq, INF_INT_eq2, SUP_UN_eq, SUP_UN_eq2.
   512 
   513 See theory "Relation" for examples for making use of pred/set
   514 conversions by means of attributes "to_set" and "to_pred".
   515 
   516 INCOMPATIBILITY.
   517 
   518 * Renamed facts about the power operation on relations, i.e., relpow
   519 to match the constant's name:
   520 
   521   rel_pow_1 ~> relpow_1
   522   rel_pow_0_I ~> relpow_0_I
   523   rel_pow_Suc_I ~> relpow_Suc_I
   524   rel_pow_Suc_I2 ~> relpow_Suc_I2
   525   rel_pow_0_E ~> relpow_0_E
   526   rel_pow_Suc_E ~> relpow_Suc_E
   527   rel_pow_E ~> relpow_E
   528   rel_pow_Suc_D2 ~> relpow_Suc_D2
   529   rel_pow_Suc_E2 ~> relpow_Suc_E2
   530   rel_pow_Suc_D2' ~> relpow_Suc_D2'
   531   rel_pow_E2 ~> relpow_E2
   532   rel_pow_add ~> relpow_add
   533   rel_pow_commute ~> relpow
   534   rel_pow_empty ~> relpow_empty:
   535   rtrancl_imp_UN_rel_pow ~> rtrancl_imp_UN_relpow
   536   rel_pow_imp_rtrancl ~> relpow_imp_rtrancl
   537   rtrancl_is_UN_rel_pow ~> rtrancl_is_UN_relpow
   538   rtrancl_imp_rel_pow ~> rtrancl_imp_relpow
   539   rel_pow_fun_conv ~> relpow_fun_conv
   540   rel_pow_finite_bounded1 ~> relpow_finite_bounded1
   541   rel_pow_finite_bounded ~> relpow_finite_bounded
   542   rtrancl_finite_eq_rel_pow ~> rtrancl_finite_eq_relpow
   543   trancl_finite_eq_rel_pow ~> trancl_finite_eq_relpow
   544   single_valued_rel_pow ~> single_valued_relpow
   545 
   546 INCOMPATIBILITY.
   547 
   548 * Theory Relation: Consolidated constant name for relation composition
   549 and corresponding theorem names:
   550 
   551   - Renamed constant rel_comp to relcomp.
   552 
   553   - Dropped abbreviation pred_comp. Use relcompp instead.
   554 
   555   - Renamed theorems:
   556 
   557     rel_compI ~> relcompI
   558     rel_compEpair ~> relcompEpair
   559     rel_compE ~> relcompE
   560     pred_comp_rel_comp_eq ~> relcompp_relcomp_eq
   561     rel_comp_empty1 ~> relcomp_empty1
   562     rel_comp_mono ~> relcomp_mono
   563     rel_comp_subset_Sigma ~> relcomp_subset_Sigma
   564     rel_comp_distrib ~> relcomp_distrib
   565     rel_comp_distrib2 ~> relcomp_distrib2
   566     rel_comp_UNION_distrib ~> relcomp_UNION_distrib
   567     rel_comp_UNION_distrib2 ~> relcomp_UNION_distrib2
   568     single_valued_rel_comp ~> single_valued_relcomp
   569     rel_comp_def ~> relcomp_unfold
   570     converse_rel_comp ~> converse_relcomp
   571     pred_compI ~> relcomppI
   572     pred_compE ~> relcomppE
   573     pred_comp_bot1 ~> relcompp_bot1
   574     pred_comp_bot2 ~> relcompp_bot2
   575     transp_pred_comp_less_eq ~> transp_relcompp_less_eq
   576     pred_comp_mono ~> relcompp_mono
   577     pred_comp_distrib ~> relcompp_distrib
   578     pred_comp_distrib2 ~> relcompp_distrib2
   579     converse_pred_comp ~> converse_relcompp
   580 
   581     finite_rel_comp ~> finite_relcomp
   582 
   583     set_rel_comp ~> set_relcomp
   584 
   585 INCOMPATIBILITY.
   586 
   587 * Theory Divides: Discontinued redundant theorems about div and mod.
   588 INCOMPATIBILITY, use the corresponding generic theorems instead.
   589 
   590   DIVISION_BY_ZERO ~> div_by_0, mod_by_0
   591   zdiv_self ~> div_self
   592   zmod_self ~> mod_self
   593   zdiv_zero ~> div_0
   594   zmod_zero ~> mod_0
   595   zdiv_zmod_equality ~> div_mod_equality2
   596   zdiv_zmod_equality2 ~> div_mod_equality
   597   zmod_zdiv_trivial ~> mod_div_trivial
   598   zdiv_zminus_zminus ~> div_minus_minus
   599   zmod_zminus_zminus ~> mod_minus_minus
   600   zdiv_zminus2 ~> div_minus_right
   601   zmod_zminus2 ~> mod_minus_right
   602   zdiv_minus1_right ~> div_minus1_right
   603   zmod_minus1_right ~> mod_minus1_right
   604   zdvd_mult_div_cancel ~> dvd_mult_div_cancel
   605   zmod_zmult1_eq ~> mod_mult_right_eq
   606   zpower_zmod ~> power_mod
   607   zdvd_zmod ~> dvd_mod
   608   zdvd_zmod_imp_zdvd ~> dvd_mod_imp_dvd
   609   mod_mult_distrib ~> mult_mod_left
   610   mod_mult_distrib2 ~> mult_mod_right
   611 
   612 * Removed redundant theorems nat_mult_2 and nat_mult_2_right; use
   613 generic mult_2 and mult_2_right instead. INCOMPATIBILITY.
   614 
   615 * Finite_Set.fold now qualified.  INCOMPATIBILITY.
   616 
   617 * Consolidated theorem names concerning fold combinators:
   618 
   619   inf_INFI_fold_inf ~> inf_INF_fold_inf
   620   sup_SUPR_fold_sup ~> sup_SUP_fold_sup
   621   INFI_fold_inf ~> INF_fold_inf
   622   SUPR_fold_sup ~> SUP_fold_sup
   623   union_set ~> union_set_fold
   624   minus_set ~> minus_set_fold
   625   INFI_set_fold ~> INF_set_fold
   626   SUPR_set_fold ~> SUP_set_fold
   627   INF_code ~> INF_set_foldr
   628   SUP_code ~> SUP_set_foldr
   629   foldr.simps ~> foldr.simps (in point-free formulation)
   630   foldr_fold_rev ~> foldr_conv_fold
   631   foldl_fold ~> foldl_conv_fold
   632   foldr_foldr ~> foldr_conv_foldl
   633   foldl_foldr ~> foldl_conv_foldr
   634   fold_set_remdups ~> fold_set_fold_remdups
   635   fold_set ~> fold_set_fold
   636   fold1_set ~> fold1_set_fold
   637 
   638 INCOMPATIBILITY.
   639 
   640 * Dropped rarely useful theorems concerning fold combinators:
   641 foldl_apply, foldl_fun_comm, foldl_rev, fold_weak_invariant,
   642 rev_foldl_cons, fold_set_remdups, fold_set, fold_set1,
   643 concat_conv_foldl, foldl_weak_invariant, foldl_invariant,
   644 foldr_invariant, foldl_absorb0, foldl_foldr1_lemma, foldl_foldr1,
   645 listsum_conv_fold, listsum_foldl, sort_foldl_insort, foldl_assoc,
   646 foldr_conv_foldl, start_le_sum, elem_le_sum, sum_eq_0_conv.
   647 INCOMPATIBILITY.  For the common phrases "%xs. List.foldr plus xs 0"
   648 and "List.foldl plus 0", prefer "List.listsum".  Otherwise it can be
   649 useful to boil down "List.foldr" and "List.foldl" to "List.fold" by
   650 unfolding "foldr_conv_fold" and "foldl_conv_fold".
   651 
   652 * Dropped lemmas minus_set_foldr, union_set_foldr, union_coset_foldr,
   653 inter_coset_foldr, Inf_fin_set_foldr, Sup_fin_set_foldr,
   654 Min_fin_set_foldr, Max_fin_set_foldr, Inf_set_foldr, Sup_set_foldr,
   655 INF_set_foldr, SUP_set_foldr.  INCOMPATIBILITY.  Prefer corresponding
   656 lemmas over fold rather than foldr, or make use of lemmas
   657 fold_conv_foldr and fold_rev.
   658 
   659 * Congruence rules Option.map_cong and Option.bind_cong for recursion
   660 through option types.
   661 
   662 * "Transitive_Closure.ntrancl": bounded transitive closure on
   663 relations.
   664 
   665 * Constant "Set.not_member" now qualified.  INCOMPATIBILITY.
   666 
   667 * Theory Int: Discontinued many legacy theorems specific to type int.
   668 INCOMPATIBILITY, use the corresponding generic theorems instead.
   669 
   670   zminus_zminus ~> minus_minus
   671   zminus_0 ~> minus_zero
   672   zminus_zadd_distrib ~> minus_add_distrib
   673   zadd_commute ~> add_commute
   674   zadd_assoc ~> add_assoc
   675   zadd_left_commute ~> add_left_commute
   676   zadd_ac ~> add_ac
   677   zmult_ac ~> mult_ac
   678   zadd_0 ~> add_0_left
   679   zadd_0_right ~> add_0_right
   680   zadd_zminus_inverse2 ~> left_minus
   681   zmult_zminus ~> mult_minus_left
   682   zmult_commute ~> mult_commute
   683   zmult_assoc ~> mult_assoc
   684   zadd_zmult_distrib ~> left_distrib
   685   zadd_zmult_distrib2 ~> right_distrib
   686   zdiff_zmult_distrib ~> left_diff_distrib
   687   zdiff_zmult_distrib2 ~> right_diff_distrib
   688   zmult_1 ~> mult_1_left
   689   zmult_1_right ~> mult_1_right
   690   zle_refl ~> order_refl
   691   zle_trans ~> order_trans
   692   zle_antisym ~> order_antisym
   693   zle_linear ~> linorder_linear
   694   zless_linear ~> linorder_less_linear
   695   zadd_left_mono ~> add_left_mono
   696   zadd_strict_right_mono ~> add_strict_right_mono
   697   zadd_zless_mono ~> add_less_le_mono
   698   int_0_less_1 ~> zero_less_one
   699   int_0_neq_1 ~> zero_neq_one
   700   zless_le ~> less_le
   701   zpower_zadd_distrib ~> power_add
   702   zero_less_zpower_abs_iff ~> zero_less_power_abs_iff
   703   zero_le_zpower_abs ~> zero_le_power_abs
   704 
   705 * Theory Deriv: Renamed
   706 
   707   DERIV_nonneg_imp_nonincreasing ~> DERIV_nonneg_imp_nondecreasing
   708 
   709 * Theory Library/Multiset: Improved code generation of multisets.
   710 
   711 * Theory HOL/Library/Set_Algebras: Addition and multiplication on sets
   712 are expressed via type classes again. The special syntax
   713 \<oplus>/\<otimes> has been replaced by plain +/*. Removed constant
   714 setsum_set, which is now subsumed by Big_Operators.setsum.
   715 INCOMPATIBILITY.
   716 
   717 * Theory HOL/Library/Diagonalize has been removed. INCOMPATIBILITY,
   718 use theory HOL/Library/Nat_Bijection instead.
   719 
   720 * Theory HOL/Library/RBT_Impl: Backing implementation of red-black
   721 trees is now inside a type class context.  Names of affected
   722 operations and lemmas have been prefixed by rbt_.  INCOMPATIBILITY for
   723 theories working directly with raw red-black trees, adapt the names as
   724 follows:
   725 
   726   Operations:
   727   bulkload -> rbt_bulkload
   728   del_from_left -> rbt_del_from_left
   729   del_from_right -> rbt_del_from_right
   730   del -> rbt_del
   731   delete -> rbt_delete
   732   ins -> rbt_ins
   733   insert -> rbt_insert
   734   insertw -> rbt_insert_with
   735   insert_with_key -> rbt_insert_with_key
   736   map_entry -> rbt_map_entry
   737   lookup -> rbt_lookup
   738   sorted -> rbt_sorted
   739   tree_greater -> rbt_greater
   740   tree_less -> rbt_less
   741   tree_less_symbol -> rbt_less_symbol
   742   union -> rbt_union
   743   union_with -> rbt_union_with
   744   union_with_key -> rbt_union_with_key
   745 
   746   Lemmas:
   747   balance_left_sorted -> balance_left_rbt_sorted
   748   balance_left_tree_greater -> balance_left_rbt_greater
   749   balance_left_tree_less -> balance_left_rbt_less
   750   balance_right_sorted -> balance_right_rbt_sorted
   751   balance_right_tree_greater -> balance_right_rbt_greater
   752   balance_right_tree_less -> balance_right_rbt_less
   753   balance_sorted -> balance_rbt_sorted
   754   balance_tree_greater -> balance_rbt_greater
   755   balance_tree_less -> balance_rbt_less
   756   bulkload_is_rbt -> rbt_bulkload_is_rbt
   757   combine_sorted -> combine_rbt_sorted
   758   combine_tree_greater -> combine_rbt_greater
   759   combine_tree_less -> combine_rbt_less
   760   delete_in_tree -> rbt_delete_in_tree
   761   delete_is_rbt -> rbt_delete_is_rbt
   762   del_from_left_tree_greater -> rbt_del_from_left_rbt_greater
   763   del_from_left_tree_less -> rbt_del_from_left_rbt_less
   764   del_from_right_tree_greater -> rbt_del_from_right_rbt_greater
   765   del_from_right_tree_less -> rbt_del_from_right_rbt_less
   766   del_in_tree -> rbt_del_in_tree
   767   del_inv1_inv2 -> rbt_del_inv1_inv2
   768   del_sorted -> rbt_del_rbt_sorted
   769   del_tree_greater -> rbt_del_rbt_greater
   770   del_tree_less -> rbt_del_rbt_less
   771   dom_lookup_Branch -> dom_rbt_lookup_Branch
   772   entries_lookup -> entries_rbt_lookup
   773   finite_dom_lookup -> finite_dom_rbt_lookup
   774   insert_sorted -> rbt_insert_rbt_sorted
   775   insertw_is_rbt -> rbt_insertw_is_rbt
   776   insertwk_is_rbt -> rbt_insertwk_is_rbt
   777   insertwk_sorted -> rbt_insertwk_rbt_sorted
   778   insertw_sorted -> rbt_insertw_rbt_sorted
   779   ins_sorted -> ins_rbt_sorted
   780   ins_tree_greater -> ins_rbt_greater
   781   ins_tree_less -> ins_rbt_less
   782   is_rbt_sorted -> is_rbt_rbt_sorted
   783   lookup_balance -> rbt_lookup_balance
   784   lookup_bulkload -> rbt_lookup_rbt_bulkload
   785   lookup_delete -> rbt_lookup_rbt_delete
   786   lookup_Empty -> rbt_lookup_Empty
   787   lookup_from_in_tree -> rbt_lookup_from_in_tree
   788   lookup_in_tree -> rbt_lookup_in_tree
   789   lookup_ins -> rbt_lookup_ins
   790   lookup_insert -> rbt_lookup_rbt_insert
   791   lookup_insertw -> rbt_lookup_rbt_insertw
   792   lookup_insertwk -> rbt_lookup_rbt_insertwk
   793   lookup_keys -> rbt_lookup_keys
   794   lookup_map -> rbt_lookup_map
   795   lookup_map_entry -> rbt_lookup_rbt_map_entry
   796   lookup_tree_greater -> rbt_lookup_rbt_greater
   797   lookup_tree_less -> rbt_lookup_rbt_less
   798   lookup_union -> rbt_lookup_rbt_union
   799   map_entry_color_of -> rbt_map_entry_color_of
   800   map_entry_inv1 -> rbt_map_entry_inv1
   801   map_entry_inv2 -> rbt_map_entry_inv2
   802   map_entry_is_rbt -> rbt_map_entry_is_rbt
   803   map_entry_sorted -> rbt_map_entry_rbt_sorted
   804   map_entry_tree_greater -> rbt_map_entry_rbt_greater
   805   map_entry_tree_less -> rbt_map_entry_rbt_less
   806   map_tree_greater -> map_rbt_greater
   807   map_tree_less -> map_rbt_less
   808   map_sorted -> map_rbt_sorted
   809   paint_sorted -> paint_rbt_sorted
   810   paint_lookup -> paint_rbt_lookup
   811   paint_tree_greater -> paint_rbt_greater
   812   paint_tree_less -> paint_rbt_less
   813   sorted_entries -> rbt_sorted_entries
   814   tree_greater_eq_trans -> rbt_greater_eq_trans
   815   tree_greater_nit -> rbt_greater_nit
   816   tree_greater_prop -> rbt_greater_prop
   817   tree_greater_simps -> rbt_greater_simps
   818   tree_greater_trans -> rbt_greater_trans
   819   tree_less_eq_trans -> rbt_less_eq_trans
   820   tree_less_nit -> rbt_less_nit
   821   tree_less_prop -> rbt_less_prop
   822   tree_less_simps -> rbt_less_simps
   823   tree_less_trans -> rbt_less_trans
   824   tree_ord_props -> rbt_ord_props
   825   union_Branch -> rbt_union_Branch
   826   union_is_rbt -> rbt_union_is_rbt
   827   unionw_is_rbt -> rbt_unionw_is_rbt
   828   unionwk_is_rbt -> rbt_unionwk_is_rbt
   829   unionwk_sorted -> rbt_unionwk_rbt_sorted
   830 
   831 * Theory HOL/Library/Float: Floating point numbers are now defined as
   832 a subset of the real numbers.  All operations are defined using the
   833 lifing-framework and proofs use the transfer method.  INCOMPATIBILITY.
   834 
   835   Changed Operations:
   836   float_abs -> abs
   837   float_nprt -> nprt
   838   float_pprt -> pprt
   839   pow2 -> use powr
   840   round_down -> float_round_down
   841   round_up -> float_round_up
   842   scale -> exponent
   843 
   844   Removed Operations:
   845   ceiling_fl, lb_mult, lb_mod, ub_mult, ub_mod
   846 
   847   Renamed Lemmas:
   848   abs_float_def -> Float.compute_float_abs
   849   bitlen_ge0 -> bitlen_nonneg
   850   bitlen.simps -> Float.compute_bitlen
   851   float_components -> Float_mantissa_exponent
   852   float_divl.simps -> Float.compute_float_divl
   853   float_divr.simps -> Float.compute_float_divr
   854   float_eq_odd -> mult_powr_eq_mult_powr_iff
   855   float_power -> real_of_float_power
   856   lapprox_posrat_def -> Float.compute_lapprox_posrat
   857   lapprox_rat.simps -> Float.compute_lapprox_rat
   858   le_float_def' -> Float.compute_float_le
   859   le_float_def -> less_eq_float.rep_eq
   860   less_float_def' -> Float.compute_float_less
   861   less_float_def -> less_float.rep_eq
   862   normfloat_def -> Float.compute_normfloat
   863   normfloat_imp_odd_or_zero -> mantissa_not_dvd and mantissa_noteq_0
   864   normfloat -> normfloat_def
   865   normfloat_unique -> use normfloat_def
   866   number_of_float_Float -> Float.compute_float_numeral, Float.compute_float_neg_numeral
   867   one_float_def -> Float.compute_float_one
   868   plus_float_def -> Float.compute_float_plus
   869   rapprox_posrat_def -> Float.compute_rapprox_posrat
   870   rapprox_rat.simps -> Float.compute_rapprox_rat
   871   real_of_float_0 -> zero_float.rep_eq
   872   real_of_float_1 -> one_float.rep_eq
   873   real_of_float_abs -> abs_float.rep_eq
   874   real_of_float_add -> plus_float.rep_eq
   875   real_of_float_minus -> uminus_float.rep_eq
   876   real_of_float_mult -> times_float.rep_eq
   877   real_of_float_simp -> Float.rep_eq
   878   real_of_float_sub -> minus_float.rep_eq
   879   round_down.simps -> Float.compute_float_round_down
   880   round_up.simps -> Float.compute_float_round_up
   881   times_float_def -> Float.compute_float_times
   882   uminus_float_def -> Float.compute_float_uminus
   883   zero_float_def -> Float.compute_float_zero
   884 
   885   Lemmas not necessary anymore, use the transfer method:
   886   bitlen_B0, bitlen_B1, bitlen_ge1, bitlen_Min, bitlen_Pls, float_divl,
   887   float_divr, float_le_simp, float_less1_mantissa_bound,
   888   float_less_simp, float_less_zero, float_le_zero,
   889   float_pos_less1_e_neg, float_pos_m_pos, float_split, float_split2,
   890   floor_pos_exp, lapprox_posrat, lapprox_posrat_bottom, lapprox_rat,
   891   lapprox_rat_bottom, normalized_float, rapprox_posrat,
   892   rapprox_posrat_le1, rapprox_rat, real_of_float_ge0_exp,
   893   real_of_float_neg_exp, real_of_float_nge0_exp, round_down floor_fl,
   894   round_up, zero_le_float, zero_less_float
   895 
   896 * New theory HOL/Library/DAList provides an abstract type for
   897 association lists with distinct keys.
   898 
   899 * Session HOL/IMP: Added new theory of abstract interpretation of
   900 annotated commands.
   901 
   902 * Session HOL-Import: Re-implementation from scratch is faster,
   903 simpler, and more scalable.  Requires a proof bundle, which is
   904 available as an external component.  Discontinued old (and mostly
   905 dead) Importer for HOL4 and HOL Light.  INCOMPATIBILITY.
   906 
   907 * Session HOL-Word: Discontinued many redundant theorems specific to
   908 type 'a word. INCOMPATIBILITY, use the corresponding generic theorems
   909 instead.
   910 
   911   word_sub_alt ~> word_sub_wi
   912   word_add_alt ~> word_add_def
   913   word_mult_alt ~> word_mult_def
   914   word_minus_alt ~> word_minus_def
   915   word_0_alt ~> word_0_wi
   916   word_1_alt ~> word_1_wi
   917   word_add_0 ~> add_0_left
   918   word_add_0_right ~> add_0_right
   919   word_mult_1 ~> mult_1_left
   920   word_mult_1_right ~> mult_1_right
   921   word_add_commute ~> add_commute
   922   word_add_assoc ~> add_assoc
   923   word_add_left_commute ~> add_left_commute
   924   word_mult_commute ~> mult_commute
   925   word_mult_assoc ~> mult_assoc
   926   word_mult_left_commute ~> mult_left_commute
   927   word_left_distrib ~> left_distrib
   928   word_right_distrib ~> right_distrib
   929   word_left_minus ~> left_minus
   930   word_diff_0_right ~> diff_0_right
   931   word_diff_self ~> diff_self
   932   word_sub_def ~> diff_minus
   933   word_diff_minus ~> diff_minus
   934   word_add_ac ~> add_ac
   935   word_mult_ac ~> mult_ac
   936   word_plus_ac0 ~> add_0_left add_0_right add_ac
   937   word_times_ac1 ~> mult_1_left mult_1_right mult_ac
   938   word_order_trans ~> order_trans
   939   word_order_refl ~> order_refl
   940   word_order_antisym ~> order_antisym
   941   word_order_linear ~> linorder_linear
   942   lenw1_zero_neq_one ~> zero_neq_one
   943   word_number_of_eq ~> number_of_eq
   944   word_of_int_add_hom ~> wi_hom_add
   945   word_of_int_sub_hom ~> wi_hom_sub
   946   word_of_int_mult_hom ~> wi_hom_mult
   947   word_of_int_minus_hom ~> wi_hom_neg
   948   word_of_int_succ_hom ~> wi_hom_succ
   949   word_of_int_pred_hom ~> wi_hom_pred
   950   word_of_int_0_hom ~> word_0_wi
   951   word_of_int_1_hom ~> word_1_wi
   952 
   953 * Session HOL-Word: New proof method "word_bitwise" for splitting
   954 machine word equalities and inequalities into logical circuits,
   955 defined in HOL/Word/WordBitwise.thy.  Supports addition, subtraction,
   956 multiplication, shifting by constants, bitwise operators and numeric
   957 constants.  Requires fixed-length word types, not 'a word.  Solves
   958 many standard word identities outright and converts more into first
   959 order problems amenable to blast or similar.  See also examples in
   960 HOL/Word/Examples/WordExamples.thy.
   961 
   962 * Session HOL-Probability: Introduced the type "'a measure" to
   963 represent measures, this replaces the records 'a algebra and 'a
   964 measure_space.  The locales based on subset_class now have two
   965 locale-parameters the space \<Omega> and the set of measurable sets M.
   966 The product of probability spaces uses now the same constant as the
   967 finite product of sigma-finite measure spaces "PiM :: ('i => 'a)
   968 measure".  Most constants are defined now outside of locales and gain
   969 an additional parameter, like null_sets, almost_eventually or \<mu>'.
   970 Measure space constructions for distributions and densities now got
   971 their own constants distr and density.  Instead of using locales to
   972 describe measure spaces with a finite space, the measure count_space
   973 and point_measure is introduced.  INCOMPATIBILITY.
   974 
   975   Renamed constants:
   976   measure -> emeasure
   977   finite_measure.\<mu>' -> measure
   978   product_algebra_generator -> prod_algebra
   979   product_prob_space.emb -> prod_emb
   980   product_prob_space.infprod_algebra -> PiM
   981 
   982   Removed locales:
   983   completeable_measure_space
   984   finite_measure_space
   985   finite_prob_space
   986   finite_product_finite_prob_space
   987   finite_product_sigma_algebra
   988   finite_sigma_algebra
   989   measure_space
   990   pair_finite_prob_space
   991   pair_finite_sigma_algebra
   992   pair_finite_space
   993   pair_sigma_algebra
   994   product_sigma_algebra
   995 
   996   Removed constants:
   997   conditional_space
   998   distribution -> use distr measure, or distributed predicate
   999   image_space
  1000   joint_distribution -> use distr measure, or distributed predicate
  1001   pair_measure_generator
  1002   product_prob_space.infprod_algebra -> use PiM
  1003   subvimage
  1004 
  1005   Replacement theorems:
  1006   finite_additivity_sufficient -> ring_of_sets.countably_additiveI_finite
  1007   finite_measure.empty_measure -> measure_empty
  1008   finite_measure.finite_continuity_from_above -> finite_measure.finite_Lim_measure_decseq
  1009   finite_measure.finite_continuity_from_below -> finite_measure.finite_Lim_measure_incseq
  1010   finite_measure.finite_measure_countably_subadditive -> finite_measure.finite_measure_subadditive_countably
  1011   finite_measure.finite_measure_eq -> finite_measure.emeasure_eq_measure
  1012   finite_measure.finite_measure -> finite_measure.emeasure_finite
  1013   finite_measure.finite_measure_finite_singleton -> finite_measure.finite_measure_eq_setsum_singleton
  1014   finite_measure.positive_measure' -> measure_nonneg
  1015   finite_measure.real_measure -> finite_measure.emeasure_real
  1016   finite_product_prob_space.finite_measure_times -> finite_product_prob_space.finite_measure_PiM_emb
  1017   finite_product_sigma_algebra.in_P -> sets_PiM_I_finite
  1018   finite_product_sigma_algebra.P_empty -> space_PiM_empty, sets_PiM_empty
  1019   information_space.conditional_entropy_eq -> information_space.conditional_entropy_simple_distributed
  1020   information_space.conditional_entropy_positive -> information_space.conditional_entropy_nonneg_simple
  1021   information_space.conditional_mutual_information_eq_mutual_information -> information_space.conditional_mutual_information_eq_mutual_information_simple
  1022   information_space.conditional_mutual_information_generic_positive -> information_space.conditional_mutual_information_nonneg_simple
  1023   information_space.conditional_mutual_information_positive -> information_space.conditional_mutual_information_nonneg_simple
  1024   information_space.entropy_commute -> information_space.entropy_commute_simple
  1025   information_space.entropy_eq -> information_space.entropy_simple_distributed
  1026   information_space.entropy_generic_eq -> information_space.entropy_simple_distributed
  1027   information_space.entropy_positive -> information_space.entropy_nonneg_simple
  1028   information_space.entropy_uniform_max -> information_space.entropy_uniform
  1029   information_space.KL_eq_0_imp -> information_space.KL_eq_0_iff_eq
  1030   information_space.KL_eq_0 -> information_space.KL_same_eq_0
  1031   information_space.KL_ge_0 -> information_space.KL_nonneg
  1032   information_space.mutual_information_eq -> information_space.mutual_information_simple_distributed
  1033   information_space.mutual_information_positive -> information_space.mutual_information_nonneg_simple
  1034   Int_stable_cuboids -> Int_stable_atLeastAtMost
  1035   Int_stable_product_algebra_generator -> positive_integral
  1036   measure_preserving -> equality "distr M N f = N" "f : measurable M N"
  1037   measure_space.additive -> emeasure_additive
  1038   measure_space.AE_iff_null_set -> AE_iff_null
  1039   measure_space.almost_everywhere_def -> eventually_ae_filter
  1040   measure_space.almost_everywhere_vimage -> AE_distrD
  1041   measure_space.continuity_from_above -> INF_emeasure_decseq
  1042   measure_space.continuity_from_above_Lim -> Lim_emeasure_decseq
  1043   measure_space.continuity_from_below_Lim -> Lim_emeasure_incseq
  1044   measure_space.continuity_from_below -> SUP_emeasure_incseq
  1045   measure_space_density -> emeasure_density
  1046   measure_space.density_is_absolutely_continuous -> absolutely_continuousI_density
  1047   measure_space.integrable_vimage -> integrable_distr
  1048   measure_space.integral_translated_density -> integral_density
  1049   measure_space.integral_vimage -> integral_distr
  1050   measure_space.measure_additive -> plus_emeasure
  1051   measure_space.measure_compl -> emeasure_compl
  1052   measure_space.measure_countable_increasing -> emeasure_countable_increasing
  1053   measure_space.measure_countably_subadditive -> emeasure_subadditive_countably
  1054   measure_space.measure_decseq -> decseq_emeasure
  1055   measure_space.measure_Diff -> emeasure_Diff
  1056   measure_space.measure_Diff_null_set -> emeasure_Diff_null_set
  1057   measure_space.measure_eq_0 -> emeasure_eq_0
  1058   measure_space.measure_finitely_subadditive -> emeasure_subadditive_finite
  1059   measure_space.measure_finite_singleton -> emeasure_eq_setsum_singleton
  1060   measure_space.measure_incseq -> incseq_emeasure
  1061   measure_space.measure_insert -> emeasure_insert
  1062   measure_space.measure_mono -> emeasure_mono
  1063   measure_space.measure_not_negative -> emeasure_not_MInf
  1064   measure_space.measure_preserving_Int_stable -> measure_eqI_generator_eq
  1065   measure_space.measure_setsum -> setsum_emeasure
  1066   measure_space.measure_setsum_split -> setsum_emeasure_cover
  1067   measure_space.measure_space_vimage -> emeasure_distr
  1068   measure_space.measure_subadditive_finite -> emeasure_subadditive_finite
  1069   measure_space.measure_subadditive -> subadditive
  1070   measure_space.measure_top -> emeasure_space
  1071   measure_space.measure_UN_eq_0 -> emeasure_UN_eq_0
  1072   measure_space.measure_Un_null_set -> emeasure_Un_null_set
  1073   measure_space.positive_integral_translated_density -> positive_integral_density
  1074   measure_space.positive_integral_vimage -> positive_integral_distr
  1075   measure_space.real_continuity_from_above -> Lim_measure_decseq
  1076   measure_space.real_continuity_from_below -> Lim_measure_incseq
  1077   measure_space.real_measure_countably_subadditive -> measure_subadditive_countably
  1078   measure_space.real_measure_Diff -> measure_Diff
  1079   measure_space.real_measure_finite_Union -> measure_finite_Union
  1080   measure_space.real_measure_setsum_singleton -> measure_eq_setsum_singleton
  1081   measure_space.real_measure_subadditive -> measure_subadditive
  1082   measure_space.real_measure_Union -> measure_Union
  1083   measure_space.real_measure_UNION -> measure_UNION
  1084   measure_space.simple_function_vimage -> simple_function_comp
  1085   measure_space.simple_integral_vimage -> simple_integral_distr
  1086   measure_space.simple_integral_vimage -> simple_integral_distr
  1087   measure_unique_Int_stable -> measure_eqI_generator_eq
  1088   measure_unique_Int_stable_vimage -> measure_eqI_generator_eq
  1089   pair_sigma_algebra.measurable_cut_fst -> sets_Pair1
  1090   pair_sigma_algebra.measurable_cut_snd -> sets_Pair2
  1091   pair_sigma_algebra.measurable_pair_image_fst -> measurable_Pair1
  1092   pair_sigma_algebra.measurable_pair_image_snd -> measurable_Pair2
  1093   pair_sigma_algebra.measurable_product_swap -> measurable_pair_swap_iff
  1094   pair_sigma_algebra.pair_sigma_algebra_measurable -> measurable_pair_swap
  1095   pair_sigma_algebra.pair_sigma_algebra_swap_measurable -> measurable_pair_swap'
  1096   pair_sigma_algebra.sets_swap -> sets_pair_swap
  1097   pair_sigma_finite.measure_cut_measurable_fst -> pair_sigma_finite.measurable_emeasure_Pair1
  1098   pair_sigma_finite.measure_cut_measurable_snd -> pair_sigma_finite.measurable_emeasure_Pair2
  1099   pair_sigma_finite.measure_preserving_swap -> pair_sigma_finite.distr_pair_swap
  1100   pair_sigma_finite.pair_measure_alt2 -> pair_sigma_finite.emeasure_pair_measure_alt2
  1101   pair_sigma_finite.pair_measure_alt -> pair_sigma_finite.emeasure_pair_measure_alt
  1102   pair_sigma_finite.pair_measure_times -> pair_sigma_finite.emeasure_pair_measure_Times
  1103   prob_space.indep_distribution_eq_measure -> prob_space.indep_vars_iff_distr_eq_PiM
  1104   prob_space.indep_var_distributionD -> prob_space.indep_var_distribution_eq
  1105   prob_space.measure_space_1 -> prob_space.emeasure_space_1
  1106   prob_space.prob_space_vimage -> prob_space_distr
  1107   prob_space.random_variable_restrict -> measurable_restrict
  1108   prob_space_unique_Int_stable -> measure_eqI_prob_space
  1109   product_algebraE -> prod_algebraE_all
  1110   product_algebra_generator_der -> prod_algebra_eq_finite
  1111   product_algebra_generator_into_space -> prod_algebra_sets_into_space
  1112   product_algebraI -> sets_PiM_I_finite
  1113   product_measure_exists -> product_sigma_finite.sigma_finite
  1114   product_prob_space.finite_index_eq_finite_product -> product_prob_space.sets_PiM_generator
  1115   product_prob_space.finite_measure_infprod_emb_Pi -> product_prob_space.measure_PiM_emb
  1116   product_prob_space.infprod_spec -> product_prob_space.emeasure_PiM_emb_not_empty
  1117   product_prob_space.measurable_component -> measurable_component_singleton
  1118   product_prob_space.measurable_emb -> measurable_prod_emb
  1119   product_prob_space.measurable_into_infprod_algebra -> measurable_PiM_single
  1120   product_prob_space.measurable_singleton_infprod -> measurable_component_singleton
  1121   product_prob_space.measure_emb -> emeasure_prod_emb
  1122   product_prob_space.measure_preserving_restrict -> product_prob_space.distr_restrict
  1123   product_sigma_algebra.product_algebra_into_space -> space_closed
  1124   product_sigma_finite.measure_fold -> product_sigma_finite.distr_merge
  1125   product_sigma_finite.measure_preserving_component_singelton -> product_sigma_finite.distr_singleton
  1126   product_sigma_finite.measure_preserving_merge -> product_sigma_finite.distr_merge
  1127   sequence_space.measure_infprod -> sequence_space.measure_PiM_countable
  1128   sets_product_algebra -> sets_PiM
  1129   sigma_algebra.measurable_sigma -> measurable_measure_of
  1130   sigma_finite_measure.disjoint_sigma_finite -> sigma_finite_disjoint
  1131   sigma_finite_measure.RN_deriv_vimage -> sigma_finite_measure.RN_deriv_distr
  1132   sigma_product_algebra_sigma_eq -> sigma_prod_algebra_sigma_eq
  1133   space_product_algebra -> space_PiM
  1134 
  1135 * Session HOL-TPTP: support to parse and import TPTP problems (all
  1136 languages) into Isabelle/HOL.
  1137 
  1138 
  1139 *** FOL ***
  1140 
  1141 * New "case_product" attribute (see HOL).
  1142 
  1143 
  1144 *** ZF ***
  1145 
  1146 * Greater support for structured proofs involving induction or case
  1147 analysis.
  1148 
  1149 * Much greater use of mathematical symbols.
  1150 
  1151 * Removal of many ML theorem bindings.  INCOMPATIBILITY.
  1152 
  1153 
  1154 *** ML ***
  1155 
  1156 * Antiquotation @{keyword "name"} produces a parser for outer syntax
  1157 from a minor keyword introduced via theory header declaration.
  1158 
  1159 * Antiquotation @{command_spec "name"} produces the
  1160 Outer_Syntax.command_spec from a major keyword introduced via theory
  1161 header declaration; it can be passed to Outer_Syntax.command etc.
  1162 
  1163 * Local_Theory.define no longer hard-wires default theorem name
  1164 "foo_def", but retains the binding as given.  If that is Binding.empty
  1165 / Attrib.empty_binding, the result is not registered as user-level
  1166 fact.  The Local_Theory.define_internal variant allows to specify a
  1167 non-empty name (used for the foundation in the background theory),
  1168 while omitting the fact binding in the user-context.  Potential
  1169 INCOMPATIBILITY for derived definitional packages: need to specify
  1170 naming policy for primitive definitions more explicitly.
  1171 
  1172 * Renamed Thm.capply to Thm.apply, and Thm.cabs to Thm.lambda in
  1173 conformance with similar operations in structure Term and Logic.
  1174 
  1175 * Antiquotation @{attributes [...]} embeds attribute source
  1176 representation into the ML text, which is particularly useful with
  1177 declarations like Local_Theory.note.
  1178 
  1179 * Structure Proof_Context follows standard naming scheme.  Old
  1180 ProofContext has been discontinued.  INCOMPATIBILITY.
  1181 
  1182 * Refined Local_Theory.declaration {syntax, pervasive}, with subtle
  1183 change of semantics: update is applied to auxiliary local theory
  1184 context as well.
  1185 
  1186 * Modernized some old-style infix operations:
  1187 
  1188   addeqcongs    ~> Simplifier.add_eqcong
  1189   deleqcongs    ~> Simplifier.del_eqcong
  1190   addcongs      ~> Simplifier.add_cong
  1191   delcongs      ~> Simplifier.del_cong
  1192   setmksimps    ~> Simplifier.set_mksimps
  1193   setmkcong     ~> Simplifier.set_mkcong
  1194   setmksym      ~> Simplifier.set_mksym
  1195   setmkeqTrue   ~> Simplifier.set_mkeqTrue
  1196   settermless   ~> Simplifier.set_termless
  1197   setsubgoaler  ~> Simplifier.set_subgoaler
  1198   addsplits     ~> Splitter.add_split
  1199   delsplits     ~> Splitter.del_split
  1200 
  1201 
  1202 *** System ***
  1203 
  1204 * USER_HOME settings variable points to cross-platform user home
  1205 directory, which coincides with HOME on POSIX systems only.  Likewise,
  1206 the Isabelle path specification "~" now expands to $USER_HOME, instead
  1207 of former $HOME.  A different default for USER_HOME may be set
  1208 explicitly in shell environment, before Isabelle settings are
  1209 evaluated.  Minor INCOMPATIBILITY: need to adapt Isabelle path where
  1210 the generic user home was intended.
  1211 
  1212 * ISABELLE_HOME_WINDOWS refers to ISABELLE_HOME in windows file name
  1213 notation, which is useful for the jEdit file browser, for example.
  1214 
  1215 * ISABELLE_JDK_HOME settings variable points to JDK with javac and jar
  1216 (not just JRE).
  1217 
  1218 
  1219 
  1220 New in Isabelle2011-1 (October 2011)
  1221 ------------------------------------
  1222 
  1223 *** General ***
  1224 
  1225 * Improved Isabelle/jEdit Prover IDE (PIDE), which can be invoked as
  1226 "isabelle jedit" or "ISABELLE_HOME/Isabelle" on the command line.
  1227 
  1228   - Management of multiple theory files directly from the editor
  1229     buffer store -- bypassing the file-system (no requirement to save
  1230     files for checking).
  1231 
  1232   - Markup of formal entities within the text buffer, with semantic
  1233     highlighting, tooltips and hyperlinks to jump to defining source
  1234     positions.
  1235 
  1236   - Improved text rendering, with sub/superscripts in the source
  1237     buffer (including support for copy/paste wrt. output panel, HTML
  1238     theory output and other non-Isabelle text boxes).
  1239 
  1240   - Refined scheduling of proof checking and printing of results,
  1241     based on interactive editor view.  (Note: jEdit folding and
  1242     narrowing allows to restrict buffer perspectives explicitly.)
  1243 
  1244   - Reduced CPU performance requirements, usable on machines with few
  1245     cores.
  1246 
  1247   - Reduced memory requirements due to pruning of unused document
  1248     versions (garbage collection).
  1249 
  1250 See also ~~/src/Tools/jEdit/README.html for further information,
  1251 including some remaining limitations.
  1252 
  1253 * Theory loader: source files are exclusively located via the master
  1254 directory of each theory node (where the .thy file itself resides).
  1255 The global load path (such as src/HOL/Library) has been discontinued.
  1256 Note that the path element ~~ may be used to reference theories in the
  1257 Isabelle home folder -- for instance, "~~/src/HOL/Library/FuncSet".
  1258 INCOMPATIBILITY.
  1259 
  1260 * Theory loader: source files are identified by content via SHA1
  1261 digests.  Discontinued former path/modtime identification and optional
  1262 ISABELLE_FILE_IDENT plugin scripts.
  1263 
  1264 * Parallelization of nested Isar proofs is subject to
  1265 Goal.parallel_proofs_threshold (default 100).  See also isabelle
  1266 usedir option -Q.
  1267 
  1268 * Name space: former unsynchronized references are now proper
  1269 configuration options, with more conventional names:
  1270 
  1271   long_names   ~> names_long
  1272   short_names  ~> names_short
  1273   unique_names ~> names_unique
  1274 
  1275 Minor INCOMPATIBILITY, need to declare options in context like this:
  1276 
  1277   declare [[names_unique = false]]
  1278 
  1279 * Literal facts `prop` may contain dummy patterns, e.g. `_ = _`.  Note
  1280 that the result needs to be unique, which means fact specifications
  1281 may have to be refined after enriching a proof context.
  1282 
  1283 * Attribute "case_names" has been refined: the assumptions in each case
  1284 can be named now by following the case name with [name1 name2 ...].
  1285 
  1286 * Isabelle/Isar reference manual has been updated and extended:
  1287   - "Synopsis" provides a catalog of main Isar language concepts.
  1288   - Formal references in syntax diagrams, via @{rail} antiquotation.
  1289   - Updated material from classic "ref" manual, notably about
  1290     "Classical Reasoner".
  1291 
  1292 
  1293 *** HOL ***
  1294 
  1295 * Class bot and top require underlying partial order rather than
  1296 preorder: uniqueness of bot and top is guaranteed.  INCOMPATIBILITY.
  1297 
  1298 * Class complete_lattice: generalized a couple of lemmas from sets;
  1299 generalized theorems INF_cong and SUP_cong.  New type classes for
  1300 complete boolean algebras and complete linear orders.  Lemmas
  1301 Inf_less_iff, less_Sup_iff, INF_less_iff, less_SUP_iff now reside in
  1302 class complete_linorder.
  1303 
  1304 Changed proposition of lemmas Inf_bool_def, Sup_bool_def, Inf_fun_def,
  1305 Sup_fun_def, Inf_apply, Sup_apply.
  1306 
  1307 Removed redundant lemmas (the right hand side gives hints how to
  1308 replace them for (metis ...), or (simp only: ...) proofs):
  1309 
  1310   Inf_singleton ~> Inf_insert [where A="{}", unfolded Inf_empty inf_top_right]
  1311   Sup_singleton ~> Sup_insert [where A="{}", unfolded Sup_empty sup_bot_right]
  1312   Inf_binary ~> Inf_insert, Inf_empty, and inf_top_right
  1313   Sup_binary ~> Sup_insert, Sup_empty, and sup_bot_right
  1314   Int_eq_Inter ~> Inf_insert, Inf_empty, and inf_top_right
  1315   Un_eq_Union ~> Sup_insert, Sup_empty, and sup_bot_right
  1316   Inter_def ~> INF_def, image_def
  1317   Union_def ~> SUP_def, image_def
  1318   INT_eq ~> INF_def, and image_def
  1319   UN_eq ~> SUP_def, and image_def
  1320   INF_subset ~> INF_superset_mono [OF _ order_refl]
  1321 
  1322 More consistent and comprehensive names:
  1323 
  1324   INTER_eq_Inter_image ~> INF_def
  1325   UNION_eq_Union_image ~> SUP_def
  1326   INFI_def ~> INF_def
  1327   SUPR_def ~> SUP_def
  1328   INF_leI ~> INF_lower
  1329   INF_leI2 ~> INF_lower2
  1330   le_INFI ~> INF_greatest
  1331   le_SUPI ~> SUP_upper
  1332   le_SUPI2 ~> SUP_upper2
  1333   SUP_leI ~> SUP_least
  1334   INFI_bool_eq ~> INF_bool_eq
  1335   SUPR_bool_eq ~> SUP_bool_eq
  1336   INFI_apply ~> INF_apply
  1337   SUPR_apply ~> SUP_apply
  1338   INTER_def ~> INTER_eq
  1339   UNION_def ~> UNION_eq
  1340 
  1341 INCOMPATIBILITY.
  1342 
  1343 * Renamed theory Complete_Lattice to Complete_Lattices.
  1344 INCOMPATIBILITY.
  1345 
  1346 * Theory Complete_Lattices: lemmas Inf_eq_top_iff, INF_eq_top_iff,
  1347 INF_image, Inf_insert, INF_top, Inf_top_conv, INF_top_conv, SUP_bot,
  1348 Sup_bot_conv, SUP_bot_conv, Sup_eq_top_iff, SUP_eq_top_iff, SUP_image,
  1349 Sup_insert are now declared as [simp].  INCOMPATIBILITY.
  1350 
  1351 * Theory Lattice: lemmas compl_inf_bot, compl_le_comp_iff,
  1352 compl_sup_top, inf_idem, inf_left_idem, inf_sup_absorb, sup_idem,
  1353 sup_inf_absob, sup_left_idem are now declared as [simp].  Minor
  1354 INCOMPATIBILITY.
  1355 
  1356 * Added syntactic classes "inf" and "sup" for the respective
  1357 constants.  INCOMPATIBILITY: Changes in the argument order of the
  1358 (mostly internal) locale predicates for some derived classes.
  1359 
  1360 * Theorem collections ball_simps and bex_simps do not contain theorems
  1361 referring to UNION any longer; these have been moved to collection
  1362 UN_ball_bex_simps.  INCOMPATIBILITY.
  1363 
  1364 * Theory Archimedean_Field: floor now is defined as parameter of a
  1365 separate type class floor_ceiling.
  1366 
  1367 * Theory Finite_Set: more coherent development of fold_set locales:
  1368 
  1369     locale fun_left_comm ~> locale comp_fun_commute
  1370     locale fun_left_comm_idem ~> locale comp_fun_idem
  1371 
  1372 Both use point-free characterization; interpretation proofs may need
  1373 adjustment.  INCOMPATIBILITY.
  1374 
  1375 * Theory Limits: Type "'a net" has been renamed to "'a filter", in
  1376 accordance with standard mathematical terminology. INCOMPATIBILITY.
  1377 
  1378 * Theory Complex_Main: The locale interpretations for the
  1379 bounded_linear and bounded_bilinear locales have been removed, in
  1380 order to reduce the number of duplicate lemmas. Users must use the
  1381 original names for distributivity theorems, potential INCOMPATIBILITY.
  1382 
  1383   divide.add ~> add_divide_distrib
  1384   divide.diff ~> diff_divide_distrib
  1385   divide.setsum ~> setsum_divide_distrib
  1386   mult.add_right ~> right_distrib
  1387   mult.diff_right ~> right_diff_distrib
  1388   mult_right.setsum ~> setsum_right_distrib
  1389   mult_left.diff ~> left_diff_distrib
  1390 
  1391 * Theory Complex_Main: Several redundant theorems have been removed or
  1392 replaced by more general versions. INCOMPATIBILITY.
  1393 
  1394   real_diff_def ~> minus_real_def
  1395   real_divide_def ~> divide_real_def
  1396   real_less_def ~> less_le
  1397   real_abs_def ~> abs_real_def
  1398   real_sgn_def ~> sgn_real_def
  1399   real_mult_commute ~> mult_commute
  1400   real_mult_assoc ~> mult_assoc
  1401   real_mult_1 ~> mult_1_left
  1402   real_add_mult_distrib ~> left_distrib
  1403   real_zero_not_eq_one ~> zero_neq_one
  1404   real_mult_inverse_left ~> left_inverse
  1405   INVERSE_ZERO ~> inverse_zero
  1406   real_le_refl ~> order_refl
  1407   real_le_antisym ~> order_antisym
  1408   real_le_trans ~> order_trans
  1409   real_le_linear ~> linear
  1410   real_le_eq_diff ~> le_iff_diff_le_0
  1411   real_add_left_mono ~> add_left_mono
  1412   real_mult_order ~> mult_pos_pos
  1413   real_mult_less_mono2 ~> mult_strict_left_mono
  1414   real_of_int_real_of_nat ~> real_of_int_of_nat_eq
  1415   real_0_le_divide_iff ~> zero_le_divide_iff
  1416   realpow_two_disj ~> power2_eq_iff
  1417   real_squared_diff_one_factored ~> square_diff_one_factored
  1418   realpow_two_diff ~> square_diff_square_factored
  1419   reals_complete2 ~> complete_real
  1420   real_sum_squared_expand ~> power2_sum
  1421   exp_ln_eq ~> ln_unique
  1422   expi_add ~> exp_add
  1423   expi_zero ~> exp_zero
  1424   lemma_DERIV_subst ~> DERIV_cong
  1425   LIMSEQ_Zfun_iff ~> tendsto_Zfun_iff
  1426   LIMSEQ_const ~> tendsto_const
  1427   LIMSEQ_norm ~> tendsto_norm
  1428   LIMSEQ_add ~> tendsto_add
  1429   LIMSEQ_minus ~> tendsto_minus
  1430   LIMSEQ_minus_cancel ~> tendsto_minus_cancel
  1431   LIMSEQ_diff ~> tendsto_diff
  1432   bounded_linear.LIMSEQ ~> bounded_linear.tendsto
  1433   bounded_bilinear.LIMSEQ ~> bounded_bilinear.tendsto
  1434   LIMSEQ_mult ~> tendsto_mult
  1435   LIMSEQ_inverse ~> tendsto_inverse
  1436   LIMSEQ_divide ~> tendsto_divide
  1437   LIMSEQ_pow ~> tendsto_power
  1438   LIMSEQ_setsum ~> tendsto_setsum
  1439   LIMSEQ_setprod ~> tendsto_setprod
  1440   LIMSEQ_norm_zero ~> tendsto_norm_zero_iff
  1441   LIMSEQ_rabs_zero ~> tendsto_rabs_zero_iff
  1442   LIMSEQ_imp_rabs ~> tendsto_rabs
  1443   LIMSEQ_add_minus ~> tendsto_add [OF _ tendsto_minus]
  1444   LIMSEQ_add_const ~> tendsto_add [OF _ tendsto_const]
  1445   LIMSEQ_diff_const ~> tendsto_diff [OF _ tendsto_const]
  1446   LIMSEQ_Complex ~> tendsto_Complex
  1447   LIM_ident ~> tendsto_ident_at
  1448   LIM_const ~> tendsto_const
  1449   LIM_add ~> tendsto_add
  1450   LIM_add_zero ~> tendsto_add_zero
  1451   LIM_minus ~> tendsto_minus
  1452   LIM_diff ~> tendsto_diff
  1453   LIM_norm ~> tendsto_norm
  1454   LIM_norm_zero ~> tendsto_norm_zero
  1455   LIM_norm_zero_cancel ~> tendsto_norm_zero_cancel
  1456   LIM_norm_zero_iff ~> tendsto_norm_zero_iff
  1457   LIM_rabs ~> tendsto_rabs
  1458   LIM_rabs_zero ~> tendsto_rabs_zero
  1459   LIM_rabs_zero_cancel ~> tendsto_rabs_zero_cancel
  1460   LIM_rabs_zero_iff ~> tendsto_rabs_zero_iff
  1461   LIM_compose ~> tendsto_compose
  1462   LIM_mult ~> tendsto_mult
  1463   LIM_scaleR ~> tendsto_scaleR
  1464   LIM_of_real ~> tendsto_of_real
  1465   LIM_power ~> tendsto_power
  1466   LIM_inverse ~> tendsto_inverse
  1467   LIM_sgn ~> tendsto_sgn
  1468   isCont_LIM_compose ~> isCont_tendsto_compose
  1469   bounded_linear.LIM ~> bounded_linear.tendsto
  1470   bounded_linear.LIM_zero ~> bounded_linear.tendsto_zero
  1471   bounded_bilinear.LIM ~> bounded_bilinear.tendsto
  1472   bounded_bilinear.LIM_prod_zero ~> bounded_bilinear.tendsto_zero
  1473   bounded_bilinear.LIM_left_zero ~> bounded_bilinear.tendsto_left_zero
  1474   bounded_bilinear.LIM_right_zero ~> bounded_bilinear.tendsto_right_zero
  1475   LIM_inverse_fun ~> tendsto_inverse [OF tendsto_ident_at]
  1476 
  1477 * Theory Complex_Main: The definition of infinite series was
  1478 generalized.  Now it is defined on the type class {topological_space,
  1479 comm_monoid_add}.  Hence it is useable also for extended real numbers.
  1480 
  1481 * Theory Complex_Main: The complex exponential function "expi" is now
  1482 a type-constrained abbreviation for "exp :: complex => complex"; thus
  1483 several polymorphic lemmas about "exp" are now applicable to "expi".
  1484 
  1485 * Code generation:
  1486 
  1487   - Theory Library/Code_Char_ord provides native ordering of
  1488     characters in the target language.
  1489 
  1490   - Commands code_module and code_library are legacy, use export_code
  1491     instead.
  1492 
  1493   - Method "evaluation" is legacy, use method "eval" instead.
  1494 
  1495   - Legacy evaluator "SML" is deactivated by default.  May be
  1496     reactivated by the following theory command:
  1497 
  1498       setup {* Value.add_evaluator ("SML", Codegen.eval_term) *}
  1499 
  1500 * Declare ext [intro] by default.  Rare INCOMPATIBILITY.
  1501 
  1502 * New proof method "induction" that gives induction hypotheses the
  1503 name "IH", thus distinguishing them from further hypotheses that come
  1504 from rule induction.  The latter are still called "hyps".  Method
  1505 "induction" is a thin wrapper around "induct" and follows the same
  1506 syntax.
  1507 
  1508 * Method "fastsimp" has been renamed to "fastforce", but "fastsimp" is
  1509 still available as a legacy feature for some time.
  1510 
  1511 * Nitpick:
  1512   - Added "need" and "total_consts" options.
  1513   - Reintroduced "show_skolems" option by popular demand.
  1514   - Renamed attribute: nitpick_def ~> nitpick_unfold.
  1515     INCOMPATIBILITY.
  1516 
  1517 * Sledgehammer:
  1518   - Use quasi-sound (and efficient) translations by default.
  1519   - Added support for the following provers: E-ToFoF, LEO-II,
  1520     Satallax, SNARK, Waldmeister, and Z3 with TPTP syntax.
  1521   - Automatically preplay and minimize proofs before showing them if
  1522     this can be done within reasonable time.
  1523   - sledgehammer available_provers ~> sledgehammer supported_provers.
  1524     INCOMPATIBILITY.
  1525   - Added "preplay_timeout", "slicing", "type_enc", "sound",
  1526     "max_mono_iters", and "max_new_mono_instances" options.
  1527   - Removed "explicit_apply" and "full_types" options as well as "Full
  1528     Types" Proof General menu item. INCOMPATIBILITY.
  1529 
  1530 * Metis:
  1531   - Removed "metisF" -- use "metis" instead. INCOMPATIBILITY.
  1532   - Obsoleted "metisFT" -- use "metis (full_types)" instead.
  1533     INCOMPATIBILITY.
  1534 
  1535 * Command 'try':
  1536   - Renamed 'try_methods' and added "simp:", "intro:", "dest:", and
  1537     "elim:" options. INCOMPATIBILITY.
  1538   - Introduced 'try' that not only runs 'try_methods' but also
  1539     'solve_direct', 'sledgehammer', 'quickcheck', and 'nitpick'.
  1540 
  1541 * Quickcheck:
  1542   - Added "eval" option to evaluate terms for the found counterexample
  1543     (currently only supported by the default (exhaustive) tester).
  1544   - Added post-processing of terms to obtain readable counterexamples
  1545     (currently only supported by the default (exhaustive) tester).
  1546   - New counterexample generator quickcheck[narrowing] enables
  1547     narrowing-based testing.  Requires the Glasgow Haskell compiler
  1548     with its installation location defined in the Isabelle settings
  1549     environment as ISABELLE_GHC.
  1550   - Removed quickcheck tester "SML" based on the SML code generator
  1551     (formly in HOL/Library).
  1552 
  1553 * Function package: discontinued option "tailrec".  INCOMPATIBILITY,
  1554 use 'partial_function' instead.
  1555 
  1556 * Theory Library/Extended_Reals replaces now the positive extended
  1557 reals found in probability theory. This file is extended by
  1558 Multivariate_Analysis/Extended_Real_Limits.
  1559 
  1560 * Theory Library/Old_Recdef: old 'recdef' package has been moved here,
  1561 from where it must be imported explicitly if it is really required.
  1562 INCOMPATIBILITY.
  1563 
  1564 * Theory Library/Wfrec: well-founded recursion combinator "wfrec" has
  1565 been moved here.  INCOMPATIBILITY.
  1566 
  1567 * Theory Library/Saturated provides type of numbers with saturated
  1568 arithmetic.
  1569 
  1570 * Theory Library/Product_Lattice defines a pointwise ordering for the
  1571 product type 'a * 'b, and provides instance proofs for various order
  1572 and lattice type classes.
  1573 
  1574 * Theory Library/Countable now provides the "countable_datatype" proof
  1575 method for proving "countable" class instances for datatypes.
  1576 
  1577 * Theory Library/Cset_Monad allows do notation for computable sets
  1578 (cset) via the generic monad ad-hoc overloading facility.
  1579 
  1580 * Library: Theories of common data structures are split into theories
  1581 for implementation, an invariant-ensuring type, and connection to an
  1582 abstract type. INCOMPATIBILITY.
  1583 
  1584   - RBT is split into RBT and RBT_Mapping.
  1585   - AssocList is split and renamed into AList and AList_Mapping.
  1586   - DList is split into DList_Impl, DList, and DList_Cset.
  1587   - Cset is split into Cset and List_Cset.
  1588 
  1589 * Theory Library/Nat_Infinity has been renamed to
  1590 Library/Extended_Nat, with name changes of the following types and
  1591 constants:
  1592 
  1593   type inat   ~> type enat
  1594   Fin         ~> enat
  1595   Infty       ~> infinity (overloaded)
  1596   iSuc        ~> eSuc
  1597   the_Fin     ~> the_enat
  1598 
  1599 Every theorem name containing "inat", "Fin", "Infty", or "iSuc" has
  1600 been renamed accordingly. INCOMPATIBILITY.
  1601 
  1602 * Session Multivariate_Analysis: The euclidean_space type class now
  1603 fixes a constant "Basis :: 'a set" consisting of the standard
  1604 orthonormal basis for the type. Users now have the option of
  1605 quantifying over this set instead of using the "basis" function, e.g.
  1606 "ALL x:Basis. P x" vs "ALL i<DIM('a). P (basis i)".
  1607 
  1608 * Session Multivariate_Analysis: Type "('a, 'b) cart" has been renamed
  1609 to "('a, 'b) vec" (the syntax "'a ^ 'b" remains unaffected). Constants
  1610 "Cart_nth" and "Cart_lambda" have been respectively renamed to
  1611 "vec_nth" and "vec_lambda"; theorems mentioning those names have
  1612 changed to match. Definition theorems for overloaded constants now use
  1613 the standard "foo_vec_def" naming scheme. A few other theorems have
  1614 been renamed as follows (INCOMPATIBILITY):
  1615 
  1616   Cart_eq          ~> vec_eq_iff
  1617   dist_nth_le_cart ~> dist_vec_nth_le
  1618   tendsto_vector   ~> vec_tendstoI
  1619   Cauchy_vector    ~> vec_CauchyI
  1620 
  1621 * Session Multivariate_Analysis: Several duplicate theorems have been
  1622 removed, and other theorems have been renamed or replaced with more
  1623 general versions. INCOMPATIBILITY.
  1624 
  1625   finite_choice ~> finite_set_choice
  1626   eventually_conjI ~> eventually_conj
  1627   eventually_and ~> eventually_conj_iff
  1628   eventually_false ~> eventually_False
  1629   setsum_norm ~> norm_setsum
  1630   Lim_sequentially ~> LIMSEQ_def
  1631   Lim_ident_at ~> LIM_ident
  1632   Lim_const ~> tendsto_const
  1633   Lim_cmul ~> tendsto_scaleR [OF tendsto_const]
  1634   Lim_neg ~> tendsto_minus
  1635   Lim_add ~> tendsto_add
  1636   Lim_sub ~> tendsto_diff
  1637   Lim_mul ~> tendsto_scaleR
  1638   Lim_vmul ~> tendsto_scaleR [OF _ tendsto_const]
  1639   Lim_null_norm ~> tendsto_norm_zero_iff [symmetric]
  1640   Lim_linear ~> bounded_linear.tendsto
  1641   Lim_component ~> tendsto_euclidean_component
  1642   Lim_component_cart ~> tendsto_vec_nth
  1643   Lim_inner ~> tendsto_inner [OF tendsto_const]
  1644   dot_lsum ~> inner_setsum_left
  1645   dot_rsum ~> inner_setsum_right
  1646   continuous_cmul ~> continuous_scaleR [OF continuous_const]
  1647   continuous_neg ~> continuous_minus
  1648   continuous_sub ~> continuous_diff
  1649   continuous_vmul ~> continuous_scaleR [OF _ continuous_const]
  1650   continuous_mul ~> continuous_scaleR
  1651   continuous_inv ~> continuous_inverse
  1652   continuous_at_within_inv ~> continuous_at_within_inverse
  1653   continuous_at_inv ~> continuous_at_inverse
  1654   continuous_at_norm ~> continuous_norm [OF continuous_at_id]
  1655   continuous_at_infnorm ~> continuous_infnorm [OF continuous_at_id]
  1656   continuous_at_component ~> continuous_component [OF continuous_at_id]
  1657   continuous_on_neg ~> continuous_on_minus
  1658   continuous_on_sub ~> continuous_on_diff
  1659   continuous_on_cmul ~> continuous_on_scaleR [OF continuous_on_const]
  1660   continuous_on_vmul ~> continuous_on_scaleR [OF _ continuous_on_const]
  1661   continuous_on_mul ~> continuous_on_scaleR
  1662   continuous_on_mul_real ~> continuous_on_mult
  1663   continuous_on_inner ~> continuous_on_inner [OF continuous_on_const]
  1664   continuous_on_norm ~> continuous_on_norm [OF continuous_on_id]
  1665   continuous_on_inverse ~> continuous_on_inv
  1666   uniformly_continuous_on_neg ~> uniformly_continuous_on_minus
  1667   uniformly_continuous_on_sub ~> uniformly_continuous_on_diff
  1668   subset_interior ~> interior_mono
  1669   subset_closure ~> closure_mono
  1670   closure_univ ~> closure_UNIV
  1671   real_arch_lt ~> reals_Archimedean2
  1672   real_arch ~> reals_Archimedean3
  1673   real_abs_norm ~> abs_norm_cancel
  1674   real_abs_sub_norm ~> norm_triangle_ineq3
  1675   norm_cauchy_schwarz_abs ~> Cauchy_Schwarz_ineq2
  1676 
  1677 * Session HOL-Probability:
  1678   - Caratheodory's extension lemma is now proved for ring_of_sets.
  1679   - Infinite products of probability measures are now available.
  1680   - Sigma closure is independent, if the generator is independent
  1681   - Use extended reals instead of positive extended
  1682     reals. INCOMPATIBILITY.
  1683 
  1684 * Session HOLCF: Discontinued legacy theorem names, INCOMPATIBILITY.
  1685 
  1686   expand_fun_below ~> fun_below_iff
  1687   below_fun_ext ~> fun_belowI
  1688   expand_cfun_eq ~> cfun_eq_iff
  1689   ext_cfun ~> cfun_eqI
  1690   expand_cfun_below ~> cfun_below_iff
  1691   below_cfun_ext ~> cfun_belowI
  1692   monofun_fun_fun ~> fun_belowD
  1693   monofun_fun_arg ~> monofunE
  1694   monofun_lub_fun ~> adm_monofun [THEN admD]
  1695   cont_lub_fun ~> adm_cont [THEN admD]
  1696   cont2cont_Rep_CFun ~> cont2cont_APP
  1697   cont_Rep_CFun_app ~> cont_APP_app
  1698   cont_Rep_CFun_app_app ~> cont_APP_app_app
  1699   cont_cfun_fun ~> cont_Rep_cfun1 [THEN contE]
  1700   cont_cfun_arg ~> cont_Rep_cfun2 [THEN contE]
  1701   contlub_cfun ~> lub_APP [symmetric]
  1702   contlub_LAM ~> lub_LAM [symmetric]
  1703   thelubI ~> lub_eqI
  1704   UU_I ~> bottomI
  1705   lift_distinct1 ~> lift.distinct(1)
  1706   lift_distinct2 ~> lift.distinct(2)
  1707   Def_not_UU ~> lift.distinct(2)
  1708   Def_inject ~> lift.inject
  1709   below_UU_iff ~> below_bottom_iff
  1710   eq_UU_iff ~> eq_bottom_iff
  1711 
  1712 
  1713 *** Document preparation ***
  1714 
  1715 * Antiquotation @{rail} layouts railroad syntax diagrams, see also
  1716 isar-ref manual, both for description and actual application of the
  1717 same.
  1718 
  1719 * Antiquotation @{value} evaluates the given term and presents its
  1720 result.
  1721 
  1722 * Antiquotations: term style "isub" provides ad-hoc conversion of
  1723 variables x1, y23 into subscripted form x\<^isub>1,
  1724 y\<^isub>2\<^isub>3.
  1725 
  1726 * Predefined LaTeX macros for Isabelle symbols \<bind> and \<then>
  1727 (e.g. see ~~/src/HOL/Library/Monad_Syntax.thy).
  1728 
  1729 * Localized \isabellestyle switch can be used within blocks or groups
  1730 like this:
  1731 
  1732   \isabellestyle{it}  %preferred default
  1733   {\isabellestylett @{text "typewriter stuff"}}
  1734 
  1735 * Discontinued special treatment of hard tabulators.  Implicit
  1736 tab-width is now defined as 1.  Potential INCOMPATIBILITY for visual
  1737 layouts.
  1738 
  1739 
  1740 *** ML ***
  1741 
  1742 * The inner syntax of sort/type/term/prop supports inlined YXML
  1743 representations within quoted string tokens.  By encoding logical
  1744 entities via Term_XML (in ML or Scala) concrete syntax can be
  1745 bypassed, which is particularly useful for producing bits of text
  1746 under external program control.
  1747 
  1748 * Antiquotations for ML and document preparation are managed as theory
  1749 data, which requires explicit setup.
  1750 
  1751 * Isabelle_Process.is_active allows tools to check if the official
  1752 process wrapper is running (Isabelle/Scala/jEdit) or the old TTY loop
  1753 (better known as Proof General).
  1754 
  1755 * Structure Proof_Context follows standard naming scheme.  Old
  1756 ProofContext is still available for some time as legacy alias.
  1757 
  1758 * Structure Timing provides various operations for timing; supersedes
  1759 former start_timing/end_timing etc.
  1760 
  1761 * Path.print is the official way to show file-system paths to users
  1762 (including quotes etc.).
  1763 
  1764 * Inner syntax: identifiers in parse trees of generic categories
  1765 "logic", "aprop", "idt" etc. carry position information (disguised as
  1766 type constraints).  Occasional INCOMPATIBILITY with non-compliant
  1767 translations that choke on unexpected type constraints.  Positions can
  1768 be stripped in ML translations via Syntax.strip_positions /
  1769 Syntax.strip_positions_ast, or via the syntax constant
  1770 "_strip_positions" within parse trees.  As last resort, positions can
  1771 be disabled via the configuration option Syntax.positions, which is
  1772 called "syntax_positions" in Isar attribute syntax.
  1773 
  1774 * Discontinued special status of various ML structures that contribute
  1775 to structure Syntax (Ast, Lexicon, Mixfix, Parser, Printer etc.): less
  1776 pervasive content, no inclusion in structure Syntax.  INCOMPATIBILITY,
  1777 refer directly to Ast.Constant, Lexicon.is_identifier,
  1778 Syntax_Trans.mk_binder_tr etc.
  1779 
  1780 * Typed print translation: discontinued show_sorts argument, which is
  1781 already available via context of "advanced" translation.
  1782 
  1783 * Refined PARALLEL_GOALS tactical: degrades gracefully for schematic
  1784 goal states; body tactic needs to address all subgoals uniformly.
  1785 
  1786 * Slightly more special eq_list/eq_set, with shortcut involving
  1787 pointer equality (assumes that eq relation is reflexive).
  1788 
  1789 * Classical tactics use proper Proof.context instead of historic types
  1790 claset/clasimpset.  Old-style declarations like addIs, addEs, addDs
  1791 operate directly on Proof.context.  Raw type claset retains its use as
  1792 snapshot of the classical context, which can be recovered via
  1793 (put_claset HOL_cs) etc.  Type clasimpset has been discontinued.
  1794 INCOMPATIBILITY, classical tactics and derived proof methods require
  1795 proper Proof.context.
  1796 
  1797 
  1798 *** System ***
  1799 
  1800 * Discontinued support for Poly/ML 5.2, which was the last version
  1801 without proper multithreading and TimeLimit implementation.
  1802 
  1803 * Discontinued old lib/scripts/polyml-platform, which has been
  1804 obsolete since Isabelle2009-2.
  1805 
  1806 * Various optional external tools are referenced more robustly and
  1807 uniformly by explicit Isabelle settings as follows:
  1808 
  1809   ISABELLE_CSDP   (formerly CSDP_EXE)
  1810   ISABELLE_GHC    (formerly EXEC_GHC or GHC_PATH)
  1811   ISABELLE_OCAML  (formerly EXEC_OCAML)
  1812   ISABELLE_SWIPL  (formerly EXEC_SWIPL)
  1813   ISABELLE_YAP    (formerly EXEC_YAP)
  1814 
  1815 Note that automated detection from the file-system or search path has
  1816 been discontinued.  INCOMPATIBILITY.
  1817 
  1818 * Scala layer provides JVM method invocation service for static
  1819 methods of type (String)String, see Invoke_Scala.method in ML.  For
  1820 example:
  1821 
  1822   Invoke_Scala.method "java.lang.System.getProperty" "java.home"
  1823 
  1824 Together with YXML.string_of_body/parse_body and XML.Encode/Decode
  1825 this allows to pass structured values between ML and Scala.
  1826 
  1827 * The IsabelleText fonts includes some further glyphs to support the
  1828 Prover IDE.  Potential INCOMPATIBILITY: users who happen to have
  1829 installed a local copy (which is normally *not* required) need to
  1830 delete or update it from ~~/lib/fonts/.
  1831 
  1832 
  1833 
  1834 New in Isabelle2011 (January 2011)
  1835 ----------------------------------
  1836 
  1837 *** General ***
  1838 
  1839 * Experimental Prover IDE based on Isabelle/Scala and jEdit (see
  1840 src/Tools/jEdit).  This also serves as IDE for Isabelle/ML, with
  1841 useful tooltips and hyperlinks produced from its static analysis.  The
  1842 bundled component provides an executable Isabelle tool that can be run
  1843 like this:
  1844 
  1845   Isabelle2011/bin/isabelle jedit
  1846 
  1847 * Significantly improved Isabelle/Isar implementation manual.
  1848 
  1849 * System settings: ISABELLE_HOME_USER now includes ISABELLE_IDENTIFIER
  1850 (and thus refers to something like $HOME/.isabelle/Isabelle2011),
  1851 while the default heap location within that directory lacks that extra
  1852 suffix.  This isolates multiple Isabelle installations from each
  1853 other, avoiding problems with old settings in new versions.
  1854 INCOMPATIBILITY, need to copy/upgrade old user settings manually.
  1855 
  1856 * Source files are always encoded as UTF-8, instead of old-fashioned
  1857 ISO-Latin-1.  INCOMPATIBILITY.  Isabelle LaTeX documents might require
  1858 the following package declarations:
  1859 
  1860   \usepackage[utf8]{inputenc}
  1861   \usepackage{textcomp}
  1862 
  1863 * Explicit treatment of UTF-8 sequences as Isabelle symbols, such that
  1864 a Unicode character is treated as a single symbol, not a sequence of
  1865 non-ASCII bytes as before.  Since Isabelle/ML string literals may
  1866 contain symbols without further backslash escapes, Unicode can now be
  1867 used here as well.  Recall that Symbol.explode in ML provides a
  1868 consistent view on symbols, while raw explode (or String.explode)
  1869 merely give a byte-oriented representation.
  1870 
  1871 * Theory loader: source files are primarily located via the master
  1872 directory of each theory node (where the .thy file itself resides).
  1873 The global load path is still partially available as legacy feature.
  1874 Minor INCOMPATIBILITY due to subtle change in file lookup: use
  1875 explicit paths, relatively to the theory.
  1876 
  1877 * Special treatment of ML file names has been discontinued.
  1878 Historically, optional extensions .ML or .sml were added on demand --
  1879 at the cost of clarity of file dependencies.  Recall that Isabelle/ML
  1880 files exclusively use the .ML extension.  Minor INCOMPATIBILTY.
  1881 
  1882 * Various options that affect pretty printing etc. are now properly
  1883 handled within the context via configuration options, instead of
  1884 unsynchronized references or print modes.  There are both ML Config.T
  1885 entities and Isar declaration attributes to access these.
  1886 
  1887   ML (Config.T)                 Isar (attribute)
  1888 
  1889   eta_contract                  eta_contract
  1890   show_brackets                 show_brackets
  1891   show_sorts                    show_sorts
  1892   show_types                    show_types
  1893   show_question_marks           show_question_marks
  1894   show_consts                   show_consts
  1895   show_abbrevs                  show_abbrevs
  1896 
  1897   Syntax.ast_trace              syntax_ast_trace
  1898   Syntax.ast_stat               syntax_ast_stat
  1899   Syntax.ambiguity_level        syntax_ambiguity_level
  1900 
  1901   Goal_Display.goals_limit      goals_limit
  1902   Goal_Display.show_main_goal   show_main_goal
  1903 
  1904   Method.rule_trace             rule_trace
  1905 
  1906   Thy_Output.display            thy_output_display
  1907   Thy_Output.quotes             thy_output_quotes
  1908   Thy_Output.indent             thy_output_indent
  1909   Thy_Output.source             thy_output_source
  1910   Thy_Output.break              thy_output_break
  1911 
  1912 Note that corresponding "..._default" references in ML may only be
  1913 changed globally at the ROOT session setup, but *not* within a theory.
  1914 The option "show_abbrevs" supersedes the former print mode
  1915 "no_abbrevs" with inverted meaning.
  1916 
  1917 * More systematic naming of some configuration options.
  1918 INCOMPATIBILITY.
  1919 
  1920   trace_simp  ~>  simp_trace
  1921   debug_simp  ~>  simp_debug
  1922 
  1923 * Support for real valued configuration options, using simplistic
  1924 floating-point notation that coincides with the inner syntax for
  1925 float_token.
  1926 
  1927 * Support for real valued preferences (with approximative PGIP type):
  1928 front-ends need to accept "pgint" values in float notation.
  1929 INCOMPATIBILITY.
  1930 
  1931 * The IsabelleText font now includes Cyrillic, Hebrew, Arabic from
  1932 DejaVu Sans.
  1933 
  1934 * Discontinued support for Poly/ML 5.0 and 5.1 versions.
  1935 
  1936 
  1937 *** Pure ***
  1938 
  1939 * Command 'type_synonym' (with single argument) replaces somewhat
  1940 outdated 'types', which is still available as legacy feature for some
  1941 time.
  1942 
  1943 * Command 'nonterminal' (with 'and' separated list of arguments)
  1944 replaces somewhat outdated 'nonterminals'.  INCOMPATIBILITY.
  1945 
  1946 * Command 'notepad' replaces former 'example_proof' for
  1947 experimentation in Isar without any result.  INCOMPATIBILITY.
  1948 
  1949 * Locale interpretation commands 'interpret' and 'sublocale' accept
  1950 lists of equations to map definitions in a locale to appropriate
  1951 entities in the context of the interpretation.  The 'interpretation'
  1952 command already provided this functionality.
  1953 
  1954 * Diagnostic command 'print_dependencies' prints the locale instances
  1955 that would be activated if the specified expression was interpreted in
  1956 the current context.  Variant "print_dependencies!" assumes a context
  1957 without interpretations.
  1958 
  1959 * Diagnostic command 'print_interps' prints interpretations in proofs
  1960 in addition to interpretations in theories.
  1961 
  1962 * Discontinued obsolete 'global' and 'local' commands to manipulate
  1963 the theory name space.  Rare INCOMPATIBILITY.  The ML functions
  1964 Sign.root_path and Sign.local_path may be applied directly where this
  1965 feature is still required for historical reasons.
  1966 
  1967 * Discontinued obsolete 'constdefs' command.  INCOMPATIBILITY, use
  1968 'definition' instead.
  1969 
  1970 * The "prems" fact, which refers to the accidental collection of
  1971 foundational premises in the context, is now explicitly marked as
  1972 legacy feature and will be discontinued soon.  Consider using "assms"
  1973 of the head statement or reference facts by explicit names.
  1974 
  1975 * Document antiquotations @{class} and @{type} print classes and type
  1976 constructors.
  1977 
  1978 * Document antiquotation @{file} checks file/directory entries within
  1979 the local file system.
  1980 
  1981 
  1982 *** HOL ***
  1983 
  1984 * Coercive subtyping: functions can be declared as coercions and type
  1985 inference will add them as necessary upon input of a term.  Theory
  1986 Complex_Main declares real :: nat => real and real :: int => real as
  1987 coercions. A coercion function f is declared like this:
  1988 
  1989   declare [[coercion f]]
  1990 
  1991 To lift coercions through type constructors (e.g. from nat => real to
  1992 nat list => real list), map functions can be declared, e.g.
  1993 
  1994   declare [[coercion_map map]]
  1995 
  1996 Currently coercion inference is activated only in theories including
  1997 real numbers, i.e. descendants of Complex_Main.  This is controlled by
  1998 the configuration option "coercion_enabled", e.g. it can be enabled in
  1999 other theories like this:
  2000 
  2001   declare [[coercion_enabled]]
  2002 
  2003 * Command 'partial_function' provides basic support for recursive
  2004 function definitions over complete partial orders.  Concrete instances
  2005 are provided for i) the option type, ii) tail recursion on arbitrary
  2006 types, and iii) the heap monad of Imperative_HOL.  See
  2007 src/HOL/ex/Fundefs.thy and src/HOL/Imperative_HOL/ex/Linked_Lists.thy
  2008 for examples.
  2009 
  2010 * Function package: f.psimps rules are no longer implicitly declared
  2011 as [simp].  INCOMPATIBILITY.
  2012 
  2013 * Datatype package: theorems generated for executable equality (class
  2014 "eq") carry proper names and are treated as default code equations.
  2015 
  2016 * Inductive package: now offers command 'inductive_simps' to
  2017 automatically derive instantiated and simplified equations for
  2018 inductive predicates, similar to 'inductive_cases'.
  2019 
  2020 * Command 'enriched_type' allows to register properties of the
  2021 functorial structure of types.
  2022 
  2023 * Improved infrastructure for term evaluation using code generator
  2024 techniques, in particular static evaluation conversions.
  2025 
  2026 * Code generator: Scala (2.8 or higher) has been added to the target
  2027 languages.
  2028 
  2029 * Code generator: globbing constant expressions "*" and "Theory.*"
  2030 have been replaced by the more idiomatic "_" and "Theory._".
  2031 INCOMPATIBILITY.
  2032 
  2033 * Code generator: export_code without explicit file declaration prints
  2034 to standard output.  INCOMPATIBILITY.
  2035 
  2036 * Code generator: do not print function definitions for case
  2037 combinators any longer.
  2038 
  2039 * Code generator: simplification with rules determined with
  2040 src/Tools/Code/code_simp.ML and method "code_simp".
  2041 
  2042 * Code generator for records: more idiomatic representation of record
  2043 types.  Warning: records are not covered by ancient SML code
  2044 generation any longer.  INCOMPATIBILITY.  In cases of need, a suitable
  2045 rep_datatype declaration helps to succeed then:
  2046 
  2047   record 'a foo = ...
  2048   ...
  2049   rep_datatype foo_ext ...
  2050 
  2051 * Records: logical foundation type for records does not carry a
  2052 '_type' suffix any longer (obsolete due to authentic syntax).
  2053 INCOMPATIBILITY.
  2054 
  2055 * Quickcheck now by default uses exhaustive testing instead of random
  2056 testing.  Random testing can be invoked by "quickcheck [random]",
  2057 exhaustive testing by "quickcheck [exhaustive]".
  2058 
  2059 * Quickcheck instantiates polymorphic types with small finite
  2060 datatypes by default. This enables a simple execution mechanism to
  2061 handle quantifiers and function equality over the finite datatypes.
  2062 
  2063 * Quickcheck random generator has been renamed from "code" to
  2064 "random".  INCOMPATIBILITY.
  2065 
  2066 * Quickcheck now has a configurable time limit which is set to 30
  2067 seconds by default. This can be changed by adding [timeout = n] to the
  2068 quickcheck command. The time limit for Auto Quickcheck is still set
  2069 independently.
  2070 
  2071 * Quickcheck in locales considers interpretations of that locale for
  2072 counter example search.
  2073 
  2074 * Sledgehammer:
  2075   - Added "smt" and "remote_smt" provers based on the "smt" proof
  2076     method. See the Sledgehammer manual for details ("isabelle doc
  2077     sledgehammer").
  2078   - Renamed commands:
  2079     sledgehammer atp_info ~> sledgehammer running_provers
  2080     sledgehammer atp_kill ~> sledgehammer kill_provers
  2081     sledgehammer available_atps ~> sledgehammer available_provers
  2082     INCOMPATIBILITY.
  2083   - Renamed options:
  2084     sledgehammer [atps = ...] ~> sledgehammer [provers = ...]
  2085     sledgehammer [atp = ...] ~> sledgehammer [prover = ...]
  2086     sledgehammer [timeout = 77 s] ~> sledgehammer [timeout = 77]
  2087     (and "ms" and "min" are no longer supported)
  2088     INCOMPATIBILITY.
  2089 
  2090 * Nitpick:
  2091   - Renamed options:
  2092     nitpick [timeout = 77 s] ~> nitpick [timeout = 77]
  2093     nitpick [tac_timeout = 777 ms] ~> nitpick [tac_timeout = 0.777]
  2094     INCOMPATIBILITY.
  2095   - Added support for partial quotient types.
  2096   - Added local versions of the "Nitpick.register_xxx" functions.
  2097   - Added "whack" option.
  2098   - Allow registration of quotient types as codatatypes.
  2099   - Improved "merge_type_vars" option to merge more types.
  2100   - Removed unsound "fast_descrs" option.
  2101   - Added custom symmetry breaking for datatypes, making it possible to reach
  2102     higher cardinalities.
  2103   - Prevent the expansion of too large definitions.
  2104 
  2105 * Proof methods "metis" and "meson" now have configuration options
  2106 "meson_trace", "metis_trace", and "metis_verbose" that can be enabled
  2107 to diagnose these tools. E.g.
  2108 
  2109     using [[metis_trace = true]]
  2110 
  2111 * Auto Solve: Renamed "Auto Solve Direct".  The tool is now available
  2112 manually as command 'solve_direct'.
  2113 
  2114 * The default SMT solver Z3 must be enabled explicitly (due to
  2115 licensing issues) by setting the environment variable
  2116 Z3_NON_COMMERCIAL in etc/settings of the component, for example.  For
  2117 commercial applications, the SMT solver CVC3 is provided as fall-back;
  2118 changing the SMT solver is done via the configuration option
  2119 "smt_solver".
  2120 
  2121 * Remote SMT solvers need to be referred to by the "remote_" prefix,
  2122 i.e. "remote_cvc3" and "remote_z3".
  2123 
  2124 * Added basic SMT support for datatypes, records, and typedefs using
  2125 the oracle mode (no proofs).  Direct support of pairs has been dropped
  2126 in exchange (pass theorems fst_conv snd_conv pair_collapse to the SMT
  2127 support for a similar behavior).  Minor INCOMPATIBILITY.
  2128 
  2129 * Changed SMT configuration options:
  2130   - Renamed:
  2131     z3_proofs ~> smt_oracle (with inverted meaning)
  2132     z3_trace_assms ~> smt_trace_used_facts
  2133     INCOMPATIBILITY.
  2134   - Added:
  2135     smt_verbose
  2136     smt_random_seed
  2137     smt_datatypes
  2138     smt_infer_triggers
  2139     smt_monomorph_limit
  2140     cvc3_options
  2141     remote_cvc3_options
  2142     remote_z3_options
  2143     yices_options
  2144 
  2145 * Boogie output files (.b2i files) need to be declared in the theory
  2146 header.
  2147 
  2148 * Simplification procedure "list_to_set_comprehension" rewrites list
  2149 comprehensions applied to List.set to set comprehensions.  Occasional
  2150 INCOMPATIBILITY, may be deactivated like this:
  2151 
  2152   declare [[simproc del: list_to_set_comprehension]]
  2153 
  2154 * Removed old version of primrec package.  INCOMPATIBILITY.
  2155 
  2156 * Removed simplifier congruence rule of "prod_case", as has for long
  2157 been the case with "split".  INCOMPATIBILITY.
  2158 
  2159 * String.literal is a type, but not a datatype.  INCOMPATIBILITY.
  2160 
  2161 * Removed [split_format ... and ... and ...] version of
  2162 [split_format].  Potential INCOMPATIBILITY.
  2163 
  2164 * Predicate "sorted" now defined inductively, with nice induction
  2165 rules.  INCOMPATIBILITY: former sorted.simps now named sorted_simps.
  2166 
  2167 * Constant "contents" renamed to "the_elem", to free the generic name
  2168 contents for other uses.  INCOMPATIBILITY.
  2169 
  2170 * Renamed class eq and constant eq (for code generation) to class
  2171 equal and constant equal, plus renaming of related facts and various
  2172 tuning.  INCOMPATIBILITY.
  2173 
  2174 * Dropped type classes mult_mono and mult_mono1.  INCOMPATIBILITY.
  2175 
  2176 * Removed output syntax "'a ~=> 'b" for "'a => 'b option".
  2177 INCOMPATIBILITY.
  2178 
  2179 * Renamed theory Fset to Cset, type Fset.fset to Cset.set, in order to
  2180 avoid confusion with finite sets.  INCOMPATIBILITY.
  2181 
  2182 * Abandoned locales equiv, congruent and congruent2 for equivalence
  2183 relations.  INCOMPATIBILITY: use equivI rather than equiv_intro (same
  2184 for congruent(2)).
  2185 
  2186 * Some previously unqualified names have been qualified:
  2187 
  2188   types
  2189     bool ~> HOL.bool
  2190     nat ~> Nat.nat
  2191 
  2192   constants
  2193     Trueprop ~> HOL.Trueprop
  2194     True ~> HOL.True
  2195     False ~> HOL.False
  2196     op & ~> HOL.conj
  2197     op | ~> HOL.disj
  2198     op --> ~> HOL.implies
  2199     op = ~> HOL.eq
  2200     Not ~> HOL.Not
  2201     The ~> HOL.The
  2202     All ~> HOL.All
  2203     Ex ~> HOL.Ex
  2204     Ex1 ~> HOL.Ex1
  2205     Let ~> HOL.Let
  2206     If ~> HOL.If
  2207     Ball ~> Set.Ball
  2208     Bex ~> Set.Bex
  2209     Suc ~> Nat.Suc
  2210     Pair ~> Product_Type.Pair
  2211     fst ~> Product_Type.fst
  2212     snd ~> Product_Type.snd
  2213     curry ~> Product_Type.curry
  2214     op : ~> Set.member
  2215     Collect ~> Set.Collect
  2216 
  2217 INCOMPATIBILITY.
  2218 
  2219 * More canonical naming convention for some fundamental definitions:
  2220 
  2221     bot_bool_eq ~> bot_bool_def
  2222     top_bool_eq ~> top_bool_def
  2223     inf_bool_eq ~> inf_bool_def
  2224     sup_bool_eq ~> sup_bool_def
  2225     bot_fun_eq  ~> bot_fun_def
  2226     top_fun_eq  ~> top_fun_def
  2227     inf_fun_eq  ~> inf_fun_def
  2228     sup_fun_eq  ~> sup_fun_def
  2229 
  2230 INCOMPATIBILITY.
  2231 
  2232 * More stylized fact names:
  2233 
  2234   expand_fun_eq ~> fun_eq_iff
  2235   expand_set_eq ~> set_eq_iff
  2236   set_ext       ~> set_eqI
  2237   nat_number    ~> eval_nat_numeral
  2238 
  2239 INCOMPATIBILITY.
  2240 
  2241 * Refactoring of code-generation specific operations in theory List:
  2242 
  2243   constants
  2244     null ~> List.null
  2245 
  2246   facts
  2247     mem_iff ~> member_def
  2248     null_empty ~> null_def
  2249 
  2250 INCOMPATIBILITY.  Note that these were not supposed to be used
  2251 regularly unless for striking reasons; their main purpose was code
  2252 generation.
  2253 
  2254 Various operations from the Haskell prelude are used for generating
  2255 Haskell code.
  2256 
  2257 * Term "bij f" is now an abbreviation of "bij_betw f UNIV UNIV".  Term
  2258 "surj f" is now an abbreviation of "range f = UNIV".  The theorems
  2259 bij_def and surj_def are unchanged.  INCOMPATIBILITY.
  2260 
  2261 * Abolished some non-alphabetic type names: "prod" and "sum" replace
  2262 "*" and "+" respectively.  INCOMPATIBILITY.
  2263 
  2264 * Name "Plus" of disjoint sum operator "<+>" is now hidden.  Write
  2265 "Sum_Type.Plus" instead.
  2266 
  2267 * Constant "split" has been merged with constant "prod_case"; names of
  2268 ML functions, facts etc. involving split have been retained so far,
  2269 though.  INCOMPATIBILITY.
  2270 
  2271 * Dropped old infix syntax "_ mem _" for List.member; use "_ : set _"
  2272 instead.  INCOMPATIBILITY.
  2273 
  2274 * Removed lemma "Option.is_none_none" which duplicates "is_none_def".
  2275 INCOMPATIBILITY.
  2276 
  2277 * Former theory Library/Enum is now part of the HOL-Main image.
  2278 INCOMPATIBILITY: all constants of the Enum theory now have to be
  2279 referred to by its qualified name.
  2280 
  2281   enum    ~>  Enum.enum
  2282   nlists  ~>  Enum.nlists
  2283   product ~>  Enum.product
  2284 
  2285 * Theory Library/Monad_Syntax provides do-syntax for monad types.
  2286 Syntax in Library/State_Monad has been changed to avoid ambiguities.
  2287 INCOMPATIBILITY.
  2288 
  2289 * Theory Library/SetsAndFunctions has been split into
  2290 Library/Function_Algebras and Library/Set_Algebras; canonical names
  2291 for instance definitions for functions; various improvements.
  2292 INCOMPATIBILITY.
  2293 
  2294 * Theory Library/Multiset provides stable quicksort implementation of
  2295 sort_key.
  2296 
  2297 * Theory Library/Multiset: renamed empty_idemp ~> empty_neutral.
  2298 INCOMPATIBILITY.
  2299 
  2300 * Session Multivariate_Analysis: introduced a type class for euclidean
  2301 space.  Most theorems are now stated in terms of euclidean spaces
  2302 instead of finite cartesian products.
  2303 
  2304   types
  2305     real ^ 'n ~>  'a::real_vector
  2306               ~>  'a::euclidean_space
  2307               ~>  'a::ordered_euclidean_space
  2308         (depends on your needs)
  2309 
  2310   constants
  2311      _ $ _        ~> _ $$ _
  2312      \<chi> x. _  ~> \<chi>\<chi> x. _
  2313      CARD('n)     ~> DIM('a)
  2314 
  2315 Also note that the indices are now natural numbers and not from some
  2316 finite type. Finite cartesian products of euclidean spaces, products
  2317 of euclidean spaces the real and complex numbers are instantiated to
  2318 be euclidean_spaces.  INCOMPATIBILITY.
  2319 
  2320 * Session Probability: introduced pextreal as positive extended real
  2321 numbers.  Use pextreal as value for measures.  Introduce the
  2322 Radon-Nikodym derivative, product spaces and Fubini's theorem for
  2323 arbitrary sigma finite measures.  Introduces Lebesgue measure based on
  2324 the integral in Multivariate Analysis.  INCOMPATIBILITY.
  2325 
  2326 * Session Imperative_HOL: revamped, corrected dozens of inadequacies.
  2327 INCOMPATIBILITY.
  2328 
  2329 * Session SPARK (with image HOL-SPARK) provides commands to load and
  2330 prove verification conditions generated by the SPARK Ada program
  2331 verifier.  See also src/HOL/SPARK and src/HOL/SPARK/Examples.
  2332 
  2333 
  2334 *** HOL-Algebra ***
  2335 
  2336 * Theorems for additive ring operations (locale abelian_monoid and
  2337 descendants) are generated by interpretation from their multiplicative
  2338 counterparts.  Names (in particular theorem names) have the mandatory
  2339 qualifier 'add'.  Previous theorem names are redeclared for
  2340 compatibility.
  2341 
  2342 * Structure "int_ring" is now an abbreviation (previously a
  2343 definition).  This fits more natural with advanced interpretations.
  2344 
  2345 
  2346 *** HOLCF ***
  2347 
  2348 * The domain package now runs in definitional mode by default: The
  2349 former command 'new_domain' is now called 'domain'.  To use the domain
  2350 package in its original axiomatic mode, use 'domain (unsafe)'.
  2351 INCOMPATIBILITY.
  2352 
  2353 * The new class "domain" is now the default sort.  Class "predomain"
  2354 is an unpointed version of "domain". Theories can be updated by
  2355 replacing sort annotations as shown below.  INCOMPATIBILITY.
  2356 
  2357   'a::type ~> 'a::countable
  2358   'a::cpo  ~> 'a::predomain
  2359   'a::pcpo ~> 'a::domain
  2360 
  2361 * The old type class "rep" has been superseded by class "domain".
  2362 Accordingly, users of the definitional package must remove any
  2363 "default_sort rep" declarations.  INCOMPATIBILITY.
  2364 
  2365 * The domain package (definitional mode) now supports unpointed
  2366 predomain argument types, as long as they are marked 'lazy'. (Strict
  2367 arguments must be in class "domain".) For example, the following
  2368 domain definition now works:
  2369 
  2370   domain natlist = nil | cons (lazy "nat discr") (lazy "natlist")
  2371 
  2372 * Theory HOLCF/Library/HOL_Cpo provides cpo and predomain class
  2373 instances for types from main HOL: bool, nat, int, char, 'a + 'b,
  2374 'a option, and 'a list.  Additionally, it configures fixrec and the
  2375 domain package to work with these types.  For example:
  2376 
  2377   fixrec isInl :: "('a + 'b) u -> tr"
  2378     where "isInl$(up$(Inl x)) = TT" | "isInl$(up$(Inr y)) = FF"
  2379 
  2380   domain V = VFun (lazy "V -> V") | VCon (lazy "nat") (lazy "V list")
  2381 
  2382 * The "(permissive)" option of fixrec has been replaced with a
  2383 per-equation "(unchecked)" option. See
  2384 src/HOL/HOLCF/Tutorial/Fixrec_ex.thy for examples. INCOMPATIBILITY.
  2385 
  2386 * The "bifinite" class no longer fixes a constant "approx"; the class
  2387 now just asserts that such a function exists.  INCOMPATIBILITY.
  2388 
  2389 * Former type "alg_defl" has been renamed to "defl".  HOLCF no longer
  2390 defines an embedding of type 'a defl into udom by default; instances
  2391 of "bifinite" and "domain" classes are available in
  2392 src/HOL/HOLCF/Library/Defl_Bifinite.thy.
  2393 
  2394 * The syntax "REP('a)" has been replaced with "DEFL('a)".
  2395 
  2396 * The predicate "directed" has been removed.  INCOMPATIBILITY.
  2397 
  2398 * The type class "finite_po" has been removed.  INCOMPATIBILITY.
  2399 
  2400 * The function "cprod_map" has been renamed to "prod_map".
  2401 INCOMPATIBILITY.
  2402 
  2403 * The monadic bind operator on each powerdomain has new binder syntax
  2404 similar to sets, e.g. "\<Union>\<sharp>x\<in>xs. t" represents
  2405 "upper_bind\<cdot>xs\<cdot>(\<Lambda> x. t)".
  2406 
  2407 * The infix syntax for binary union on each powerdomain has changed
  2408 from e.g. "+\<sharp>" to "\<union>\<sharp>", for consistency with set
  2409 syntax.  INCOMPATIBILITY.
  2410 
  2411 * The constant "UU" has been renamed to "bottom".  The syntax "UU" is
  2412 still supported as an input translation.
  2413 
  2414 * Renamed some theorems (the original names are also still available).
  2415 
  2416   expand_fun_below   ~> fun_below_iff
  2417   below_fun_ext      ~> fun_belowI
  2418   expand_cfun_eq     ~> cfun_eq_iff
  2419   ext_cfun           ~> cfun_eqI
  2420   expand_cfun_below  ~> cfun_below_iff
  2421   below_cfun_ext     ~> cfun_belowI
  2422   cont2cont_Rep_CFun ~> cont2cont_APP
  2423 
  2424 * The Abs and Rep functions for various types have changed names.
  2425 Related theorem names have also changed to match. INCOMPATIBILITY.
  2426 
  2427   Rep_CFun  ~> Rep_cfun
  2428   Abs_CFun  ~> Abs_cfun
  2429   Rep_Sprod ~> Rep_sprod
  2430   Abs_Sprod ~> Abs_sprod
  2431   Rep_Ssum  ~> Rep_ssum
  2432   Abs_Ssum  ~> Abs_ssum
  2433 
  2434 * Lemmas with names of the form *_defined_iff or *_strict_iff have
  2435 been renamed to *_bottom_iff.  INCOMPATIBILITY.
  2436 
  2437 * Various changes to bisimulation/coinduction with domain package:
  2438 
  2439   - Definitions of "bisim" constants no longer mention definedness.
  2440   - With mutual recursion, "bisim" predicate is now curried.
  2441   - With mutual recursion, each type gets a separate coind theorem.
  2442   - Variable names in bisim_def and coinduct rules have changed.
  2443 
  2444 INCOMPATIBILITY.
  2445 
  2446 * Case combinators generated by the domain package for type "foo" are
  2447 now named "foo_case" instead of "foo_when".  INCOMPATIBILITY.
  2448 
  2449 * Several theorems have been renamed to more accurately reflect the
  2450 names of constants and types involved.  INCOMPATIBILITY.
  2451 
  2452   thelub_const    ~> lub_const
  2453   lub_const       ~> is_lub_const
  2454   thelubI         ~> lub_eqI
  2455   is_lub_lub      ~> is_lubD2
  2456   lubI            ~> is_lub_lub
  2457   unique_lub      ~> is_lub_unique
  2458   is_ub_lub       ~> is_lub_rangeD1
  2459   lub_bin_chain   ~> is_lub_bin_chain
  2460   lub_fun         ~> is_lub_fun
  2461   thelub_fun      ~> lub_fun
  2462   thelub_cfun     ~> lub_cfun
  2463   thelub_Pair     ~> lub_Pair
  2464   lub_cprod       ~> is_lub_prod
  2465   thelub_cprod    ~> lub_prod
  2466   minimal_cprod   ~> minimal_prod
  2467   inst_cprod_pcpo ~> inst_prod_pcpo
  2468   UU_I            ~> bottomI
  2469   compact_UU      ~> compact_bottom
  2470   deflation_UU    ~> deflation_bottom
  2471   finite_deflation_UU ~> finite_deflation_bottom
  2472 
  2473 * Many legacy theorem names have been discontinued.  INCOMPATIBILITY.
  2474 
  2475   sq_ord_less_eq_trans ~> below_eq_trans
  2476   sq_ord_eq_less_trans ~> eq_below_trans
  2477   refl_less            ~> below_refl
  2478   trans_less           ~> below_trans
  2479   antisym_less         ~> below_antisym
  2480   antisym_less_inverse ~> po_eq_conv [THEN iffD1]
  2481   box_less             ~> box_below
  2482   rev_trans_less       ~> rev_below_trans
  2483   not_less2not_eq      ~> not_below2not_eq
  2484   less_UU_iff          ~> below_UU_iff
  2485   flat_less_iff        ~> flat_below_iff
  2486   adm_less             ~> adm_below
  2487   adm_not_less         ~> adm_not_below
  2488   adm_compact_not_less ~> adm_compact_not_below
  2489   less_fun_def         ~> below_fun_def
  2490   expand_fun_less      ~> fun_below_iff
  2491   less_fun_ext         ~> fun_belowI
  2492   less_discr_def       ~> below_discr_def
  2493   discr_less_eq        ~> discr_below_eq
  2494   less_unit_def        ~> below_unit_def
  2495   less_cprod_def       ~> below_prod_def
  2496   prod_lessI           ~> prod_belowI
  2497   Pair_less_iff        ~> Pair_below_iff
  2498   fst_less_iff         ~> fst_below_iff
  2499   snd_less_iff         ~> snd_below_iff
  2500   expand_cfun_less     ~> cfun_below_iff
  2501   less_cfun_ext        ~> cfun_belowI
  2502   injection_less       ~> injection_below
  2503   less_up_def          ~> below_up_def
  2504   not_Iup_less         ~> not_Iup_below
  2505   Iup_less             ~> Iup_below
  2506   up_less              ~> up_below
  2507   Def_inject_less_eq   ~> Def_below_Def
  2508   Def_less_is_eq       ~> Def_below_iff
  2509   spair_less_iff       ~> spair_below_iff
  2510   less_sprod           ~> below_sprod
  2511   spair_less           ~> spair_below
  2512   sfst_less_iff        ~> sfst_below_iff
  2513   ssnd_less_iff        ~> ssnd_below_iff
  2514   fix_least_less       ~> fix_least_below
  2515   dist_less_one        ~> dist_below_one
  2516   less_ONE             ~> below_ONE
  2517   ONE_less_iff         ~> ONE_below_iff
  2518   less_sinlD           ~> below_sinlD
  2519   less_sinrD           ~> below_sinrD
  2520 
  2521 
  2522 *** FOL and ZF ***
  2523 
  2524 * All constant names are now qualified internally and use proper
  2525 identifiers, e.g. "IFOL.eq" instead of "op =".  INCOMPATIBILITY.
  2526 
  2527 
  2528 *** ML ***
  2529 
  2530 * Antiquotation @{assert} inlines a function bool -> unit that raises
  2531 Fail if the argument is false.  Due to inlining the source position of
  2532 failed assertions is included in the error output.
  2533 
  2534 * Discontinued antiquotation @{theory_ref}, which is obsolete since ML
  2535 text is in practice always evaluated with a stable theory checkpoint.
  2536 Minor INCOMPATIBILITY, use (Theory.check_thy @{theory}) instead.
  2537 
  2538 * Antiquotation @{theory A} refers to theory A from the ancestry of
  2539 the current context, not any accidental theory loader state as before.
  2540 Potential INCOMPATIBILITY, subtle change in semantics.
  2541 
  2542 * Syntax.pretty_priority (default 0) configures the required priority
  2543 of pretty-printed output and thus affects insertion of parentheses.
  2544 
  2545 * Syntax.default_root (default "any") configures the inner syntax
  2546 category (nonterminal symbol) for parsing of terms.
  2547 
  2548 * Former exception Library.UnequalLengths now coincides with
  2549 ListPair.UnequalLengths.
  2550 
  2551 * Renamed structure MetaSimplifier to Raw_Simplifier.  Note that the
  2552 main functionality is provided by structure Simplifier.
  2553 
  2554 * Renamed raw "explode" function to "raw_explode" to emphasize its
  2555 meaning.  Note that internally to Isabelle, Symbol.explode is used in
  2556 almost all situations.
  2557 
  2558 * Discontinued obsolete function sys_error and exception SYS_ERROR.
  2559 See implementation manual for further details on exceptions in
  2560 Isabelle/ML.
  2561 
  2562 * Renamed setmp_noncritical to Unsynchronized.setmp to emphasize its
  2563 meaning.
  2564 
  2565 * Renamed structure PureThy to Pure_Thy and moved most of its
  2566 operations to structure Global_Theory, to emphasize that this is
  2567 rarely-used global-only stuff.
  2568 
  2569 * Discontinued Output.debug.  Minor INCOMPATIBILITY, use plain writeln
  2570 instead (or tracing for high-volume output).
  2571 
  2572 * Configuration option show_question_marks only affects regular pretty
  2573 printing of types and terms, not raw Term.string_of_vname.
  2574 
  2575 * ML_Context.thm and ML_Context.thms are no longer pervasive.  Rare
  2576 INCOMPATIBILITY, superseded by static antiquotations @{thm} and
  2577 @{thms} for most purposes.
  2578 
  2579 * ML structure Unsynchronized is never opened, not even in Isar
  2580 interaction mode as before.  Old Unsynchronized.set etc. have been
  2581 discontinued -- use plain := instead.  This should be *rare* anyway,
  2582 since modern tools always work via official context data, notably
  2583 configuration options.
  2584 
  2585 * Parallel and asynchronous execution requires special care concerning
  2586 interrupts.  Structure Exn provides some convenience functions that
  2587 avoid working directly with raw Interrupt.  User code must not absorb
  2588 interrupts -- intermediate handling (for cleanup etc.) needs to be
  2589 followed by re-raising of the original exception.  Another common
  2590 source of mistakes are "handle _" patterns, which make the meaning of
  2591 the program subject to physical effects of the environment.
  2592 
  2593 
  2594 
  2595 New in Isabelle2009-2 (June 2010)
  2596 ---------------------------------
  2597 
  2598 *** General ***
  2599 
  2600 * Authentic syntax for *all* logical entities (type classes, type
  2601 constructors, term constants): provides simple and robust
  2602 correspondence between formal entities and concrete syntax.  Within
  2603 the parse tree / AST representations, "constants" are decorated by
  2604 their category (class, type, const) and spelled out explicitly with
  2605 their full internal name.
  2606 
  2607 Substantial INCOMPATIBILITY concerning low-level syntax declarations
  2608 and translations (translation rules and translation functions in ML).
  2609 Some hints on upgrading:
  2610 
  2611   - Many existing uses of 'syntax' and 'translations' can be replaced
  2612     by more modern 'type_notation', 'notation' and 'abbreviation',
  2613     which are independent of this issue.
  2614 
  2615   - 'translations' require markup within the AST; the term syntax
  2616     provides the following special forms:
  2617 
  2618       CONST c   -- produces syntax version of constant c from context
  2619       XCONST c  -- literally c, checked as constant from context
  2620       c         -- literally c, if declared by 'syntax'
  2621 
  2622     Plain identifiers are treated as AST variables -- occasionally the
  2623     system indicates accidental variables via the error "rhs contains
  2624     extra variables".
  2625 
  2626     Type classes and type constructors are marked according to their
  2627     concrete syntax.  Some old translations rules need to be written
  2628     for the "type" category, using type constructor application
  2629     instead of pseudo-term application of the default category
  2630     "logic".
  2631 
  2632   - 'parse_translation' etc. in ML may use the following
  2633     antiquotations:
  2634 
  2635       @{class_syntax c}   -- type class c within parse tree / AST
  2636       @{term_syntax c}    -- type constructor c within parse tree / AST
  2637       @{const_syntax c}   -- ML version of "CONST c" above
  2638       @{syntax_const c}   -- literally c (checked wrt. 'syntax' declarations)
  2639 
  2640   - Literal types within 'typed_print_translations', i.e. those *not*
  2641     represented as pseudo-terms are represented verbatim.  Use @{class
  2642     c} or @{type_name c} here instead of the above syntax
  2643     antiquotations.
  2644 
  2645 Note that old non-authentic syntax was based on unqualified base
  2646 names, so all of the above "constant" names would coincide.  Recall
  2647 that 'print_syntax' and ML_command "set Syntax.trace_ast" help to
  2648 diagnose syntax problems.
  2649 
  2650 * Type constructors admit general mixfix syntax, not just infix.
  2651 
  2652 * Concrete syntax may be attached to local entities without a proof
  2653 body, too.  This works via regular mixfix annotations for 'fix',
  2654 'def', 'obtain' etc. or via the explicit 'write' command, which is
  2655 similar to the 'notation' command in theory specifications.
  2656 
  2657 * Discontinued unnamed infix syntax (legacy feature for many years) --
  2658 need to specify constant name and syntax separately.  Internal ML
  2659 datatype constructors have been renamed from InfixName to Infix etc.
  2660 Minor INCOMPATIBILITY.
  2661 
  2662 * Schematic theorem statements need to be explicitly markup as such,
  2663 via commands 'schematic_lemma', 'schematic_theorem',
  2664 'schematic_corollary'.  Thus the relevance of the proof is made
  2665 syntactically clear, which impacts performance in a parallel or
  2666 asynchronous interactive environment.  Minor INCOMPATIBILITY.
  2667 
  2668 * Use of cumulative prems via "!" in some proof methods has been
  2669 discontinued (old legacy feature).
  2670 
  2671 * References 'trace_simp' and 'debug_simp' have been replaced by
  2672 configuration options stored in the context. Enabling tracing (the
  2673 case of debugging is similar) in proofs works via
  2674 
  2675   using [[trace_simp = true]]
  2676 
  2677 Tracing is then active for all invocations of the simplifier in
  2678 subsequent goal refinement steps. Tracing may also still be enabled or
  2679 disabled via the ProofGeneral settings menu.
  2680 
  2681 * Separate commands 'hide_class', 'hide_type', 'hide_const',
  2682 'hide_fact' replace the former 'hide' KIND command.  Minor
  2683 INCOMPATIBILITY.
  2684 
  2685 * Improved parallelism of proof term normalization: usedir -p2 -q0 is
  2686 more efficient than combinations with -q1 or -q2.
  2687 
  2688 
  2689 *** Pure ***
  2690 
  2691 * Proofterms record type-class reasoning explicitly, using the
  2692 "unconstrain" operation internally.  This eliminates all sort
  2693 constraints from a theorem and proof, introducing explicit
  2694 OFCLASS-premises.  On the proof term level, this operation is
  2695 automatically applied at theorem boundaries, such that closed proofs
  2696 are always free of sort constraints.  INCOMPATIBILITY for tools that
  2697 inspect proof terms.
  2698 
  2699 * Local theory specifications may depend on extra type variables that
  2700 are not present in the result type -- arguments TYPE('a) :: 'a itself
  2701 are added internally.  For example:
  2702 
  2703   definition unitary :: bool where "unitary = (ALL (x::'a) y. x = y)"
  2704 
  2705 * Predicates of locales introduced by classes carry a mandatory
  2706 "class" prefix.  INCOMPATIBILITY.
  2707 
  2708 * Vacuous class specifications observe default sort.  INCOMPATIBILITY.
  2709 
  2710 * Old 'axclass' command has been discontinued.  INCOMPATIBILITY, use
  2711 'class' instead.
  2712 
  2713 * Command 'code_reflect' allows to incorporate generated ML code into
  2714 runtime environment; replaces immature code_datatype antiquotation.
  2715 INCOMPATIBILITY.
  2716 
  2717 * Code generator: simple concept for abstract datatypes obeying
  2718 invariants.
  2719 
  2720 * Code generator: details of internal data cache have no impact on the
  2721 user space functionality any longer.
  2722 
  2723 * Methods "unfold_locales" and "intro_locales" ignore non-locale
  2724 subgoals.  This is more appropriate for interpretations with 'where'.
  2725 INCOMPATIBILITY.
  2726 
  2727 * Command 'example_proof' opens an empty proof body.  This allows to
  2728 experiment with Isar, without producing any persistent result.
  2729 
  2730 * Commands 'type_notation' and 'no_type_notation' declare type syntax
  2731 within a local theory context, with explicit checking of the
  2732 constructors involved (in contrast to the raw 'syntax' versions).
  2733 
  2734 * Commands 'types' and 'typedecl' now work within a local theory
  2735 context -- without introducing dependencies on parameters or
  2736 assumptions, which is not possible in Isabelle/Pure.
  2737 
  2738 * Command 'defaultsort' has been renamed to 'default_sort', it works
  2739 within a local theory context.  Minor INCOMPATIBILITY.
  2740 
  2741 
  2742 *** HOL ***
  2743 
  2744 * Command 'typedef' now works within a local theory context -- without
  2745 introducing dependencies on parameters or assumptions, which is not
  2746 possible in Isabelle/Pure/HOL.  Note that the logical environment may
  2747 contain multiple interpretations of local typedefs (with different
  2748 non-emptiness proofs), even in a global theory context.
  2749 
  2750 * New package for quotient types.  Commands 'quotient_type' and
  2751 'quotient_definition' may be used for defining types and constants by
  2752 quotient constructions.  An example is the type of integers created by
  2753 quotienting pairs of natural numbers:
  2754 
  2755   fun
  2756     intrel :: "(nat * nat) => (nat * nat) => bool"
  2757   where
  2758     "intrel (x, y) (u, v) = (x + v = u + y)"
  2759 
  2760   quotient_type int = "nat * nat" / intrel
  2761     by (auto simp add: equivp_def expand_fun_eq)
  2762 
  2763   quotient_definition
  2764     "0::int" is "(0::nat, 0::nat)"
  2765 
  2766 The method "lifting" can be used to lift of theorems from the
  2767 underlying "raw" type to the quotient type.  The example
  2768 src/HOL/Quotient_Examples/FSet.thy includes such a quotient
  2769 construction and provides a reasoning infrastructure for finite sets.
  2770 
  2771 * Renamed Library/Quotient.thy to Library/Quotient_Type.thy to avoid
  2772 clash with new theory Quotient in Main HOL.
  2773 
  2774 * Moved the SMT binding into the main HOL session, eliminating
  2775 separate HOL-SMT session.
  2776 
  2777 * List membership infix mem operation is only an input abbreviation.
  2778 INCOMPATIBILITY.
  2779 
  2780 * Theory Library/Word.thy has been removed.  Use library Word/Word.thy
  2781 for future developements; former Library/Word.thy is still present in
  2782 the AFP entry RSAPPS.
  2783 
  2784 * Theorem Int.int_induct renamed to Int.int_of_nat_induct and is no
  2785 longer shadowed.  INCOMPATIBILITY.
  2786 
  2787 * Dropped theorem duplicate comp_arith; use semiring_norm instead.
  2788 INCOMPATIBILITY.
  2789 
  2790 * Dropped theorem RealPow.real_sq_order; use power2_le_imp_le instead.
  2791 INCOMPATIBILITY.
  2792 
  2793 * Dropped normalizing_semiring etc; use the facts in semiring classes
  2794 instead.  INCOMPATIBILITY.
  2795 
  2796 * Dropped several real-specific versions of lemmas about floor and
  2797 ceiling; use the generic lemmas from theory "Archimedean_Field"
  2798 instead.  INCOMPATIBILITY.
  2799 
  2800   floor_number_of_eq         ~> floor_number_of
  2801   le_floor_eq_number_of      ~> number_of_le_floor
  2802   le_floor_eq_zero           ~> zero_le_floor
  2803   le_floor_eq_one            ~> one_le_floor
  2804   floor_less_eq_number_of    ~> floor_less_number_of
  2805   floor_less_eq_zero         ~> floor_less_zero
  2806   floor_less_eq_one          ~> floor_less_one
  2807   less_floor_eq_number_of    ~> number_of_less_floor
  2808   less_floor_eq_zero         ~> zero_less_floor
  2809   less_floor_eq_one          ~> one_less_floor
  2810   floor_le_eq_number_of      ~> floor_le_number_of
  2811   floor_le_eq_zero           ~> floor_le_zero
  2812   floor_le_eq_one            ~> floor_le_one
  2813   floor_subtract_number_of   ~> floor_diff_number_of
  2814   floor_subtract_one         ~> floor_diff_one
  2815   ceiling_number_of_eq       ~> ceiling_number_of
  2816   ceiling_le_eq_number_of    ~> ceiling_le_number_of
  2817   ceiling_le_zero_eq         ~> ceiling_le_zero
  2818   ceiling_le_eq_one          ~> ceiling_le_one
  2819   less_ceiling_eq_number_of  ~> number_of_less_ceiling
  2820   less_ceiling_eq_zero       ~> zero_less_ceiling
  2821   less_ceiling_eq_one        ~> one_less_ceiling
  2822   ceiling_less_eq_number_of  ~> ceiling_less_number_of
  2823   ceiling_less_eq_zero       ~> ceiling_less_zero
  2824   ceiling_less_eq_one        ~> ceiling_less_one
  2825   le_ceiling_eq_number_of    ~> number_of_le_ceiling
  2826   le_ceiling_eq_zero         ~> zero_le_ceiling
  2827   le_ceiling_eq_one          ~> one_le_ceiling
  2828   ceiling_subtract_number_of ~> ceiling_diff_number_of
  2829   ceiling_subtract_one       ~> ceiling_diff_one
  2830 
  2831 * Theory "Finite_Set": various folding_XXX locales facilitate the
  2832 application of the various fold combinators on finite sets.
  2833 
  2834 * Library theory "RBT" renamed to "RBT_Impl"; new library theory "RBT"
  2835 provides abstract red-black tree type which is backed by "RBT_Impl" as
  2836 implementation.  INCOMPATIBILTY.
  2837 
  2838 * Theory Library/Coinductive_List has been removed -- superseded by
  2839 AFP/thys/Coinductive.
  2840 
  2841 * Theory PReal, including the type "preal" and related operations, has
  2842 been removed.  INCOMPATIBILITY.
  2843 
  2844 * Real: new development using Cauchy Sequences.
  2845 
  2846 * Split off theory "Big_Operators" containing setsum, setprod,
  2847 Inf_fin, Sup_fin, Min, Max from theory Finite_Set.  INCOMPATIBILITY.
  2848 
  2849 * Theory "Rational" renamed to "Rat", for consistency with "Nat",
  2850 "Int" etc.  INCOMPATIBILITY.
  2851 
  2852 * Constant Rat.normalize needs to be qualified.  INCOMPATIBILITY.
  2853 
  2854 * New set of rules "ac_simps" provides combined assoc / commute
  2855 rewrites for all interpretations of the appropriate generic locales.
  2856 
  2857 * Renamed theory "OrderedGroup" to "Groups" and split theory
  2858 "Ring_and_Field" into theories "Rings" and "Fields"; for more
  2859 appropriate and more consistent names suitable for name prefixes
  2860 within the HOL theories.  INCOMPATIBILITY.
  2861 
  2862 * Some generic constants have been put to appropriate theories:
  2863   - less_eq, less: Orderings
  2864   - zero, one, plus, minus, uminus, times, abs, sgn: Groups
  2865   - inverse, divide: Rings
  2866 INCOMPATIBILITY.
  2867 
  2868 * More consistent naming of type classes involving orderings (and
  2869 lattices):
  2870 
  2871     lower_semilattice                   ~> semilattice_inf
  2872     upper_semilattice                   ~> semilattice_sup
  2873 
  2874     dense_linear_order                  ~> dense_linorder
  2875 
  2876     pordered_ab_group_add               ~> ordered_ab_group_add
  2877     pordered_ab_group_add_abs           ~> ordered_ab_group_add_abs
  2878     pordered_ab_semigroup_add           ~> ordered_ab_semigroup_add
  2879     pordered_ab_semigroup_add_imp_le    ~> ordered_ab_semigroup_add_imp_le
  2880     pordered_cancel_ab_semigroup_add    ~> ordered_cancel_ab_semigroup_add
  2881     pordered_cancel_comm_semiring       ~> ordered_cancel_comm_semiring
  2882     pordered_cancel_semiring            ~> ordered_cancel_semiring
  2883     pordered_comm_monoid_add            ~> ordered_comm_monoid_add
  2884     pordered_comm_ring                  ~> ordered_comm_ring
  2885     pordered_comm_semiring              ~> ordered_comm_semiring
  2886     pordered_ring                       ~> ordered_ring
  2887     pordered_ring_abs                   ~> ordered_ring_abs
  2888     pordered_semiring                   ~> ordered_semiring
  2889 
  2890     ordered_ab_group_add                ~> linordered_ab_group_add
  2891     ordered_ab_semigroup_add            ~> linordered_ab_semigroup_add
  2892     ordered_cancel_ab_semigroup_add     ~> linordered_cancel_ab_semigroup_add
  2893     ordered_comm_semiring_strict        ~> linordered_comm_semiring_strict
  2894     ordered_field                       ~> linordered_field
  2895     ordered_field_no_lb                 ~> linordered_field_no_lb
  2896     ordered_field_no_ub                 ~> linordered_field_no_ub
  2897     ordered_field_dense_linear_order    ~> dense_linordered_field
  2898     ordered_idom                        ~> linordered_idom
  2899     ordered_ring                        ~> linordered_ring
  2900     ordered_ring_le_cancel_factor       ~> linordered_ring_le_cancel_factor
  2901     ordered_ring_less_cancel_factor     ~> linordered_ring_less_cancel_factor
  2902     ordered_ring_strict                 ~> linordered_ring_strict
  2903     ordered_semidom                     ~> linordered_semidom
  2904     ordered_semiring                    ~> linordered_semiring
  2905     ordered_semiring_1                  ~> linordered_semiring_1
  2906     ordered_semiring_1_strict           ~> linordered_semiring_1_strict
  2907     ordered_semiring_strict             ~> linordered_semiring_strict
  2908 
  2909   The following slightly odd type classes have been moved to a
  2910   separate theory Library/Lattice_Algebras:
  2911 
  2912     lordered_ab_group_add               ~> lattice_ab_group_add
  2913     lordered_ab_group_add_abs           ~> lattice_ab_group_add_abs
  2914     lordered_ab_group_add_meet          ~> semilattice_inf_ab_group_add
  2915     lordered_ab_group_add_join          ~> semilattice_sup_ab_group_add
  2916     lordered_ring                       ~> lattice_ring
  2917 
  2918 INCOMPATIBILITY.
  2919 
  2920 * Refined field classes:
  2921   - classes division_ring_inverse_zero, field_inverse_zero,
  2922     linordered_field_inverse_zero include rule inverse 0 = 0 --
  2923     subsumes former division_by_zero class;
  2924   - numerous lemmas have been ported from field to division_ring.
  2925 INCOMPATIBILITY.
  2926 
  2927 * Refined algebra theorem collections:
  2928   - dropped theorem group group_simps, use algebra_simps instead;
  2929   - dropped theorem group ring_simps, use field_simps instead;
  2930   - proper theorem collection field_simps subsumes former theorem
  2931     groups field_eq_simps and field_simps;
  2932   - dropped lemma eq_minus_self_iff which is a duplicate for
  2933     equal_neg_zero.
  2934 INCOMPATIBILITY.
  2935 
  2936 * Theory Finite_Set and List: some lemmas have been generalized from
  2937 sets to lattices:
  2938 
  2939   fun_left_comm_idem_inter      ~> fun_left_comm_idem_inf
  2940   fun_left_comm_idem_union      ~> fun_left_comm_idem_sup
  2941   inter_Inter_fold_inter        ~> inf_Inf_fold_inf
  2942   union_Union_fold_union        ~> sup_Sup_fold_sup
  2943   Inter_fold_inter              ~> Inf_fold_inf
  2944   Union_fold_union              ~> Sup_fold_sup
  2945   inter_INTER_fold_inter        ~> inf_INFI_fold_inf
  2946   union_UNION_fold_union        ~> sup_SUPR_fold_sup
  2947   INTER_fold_inter              ~> INFI_fold_inf
  2948   UNION_fold_union              ~> SUPR_fold_sup
  2949 
  2950 * Theory "Complete_Lattice": lemmas top_def and bot_def have been
  2951 replaced by the more convenient lemmas Inf_empty and Sup_empty.
  2952 Dropped lemmas Inf_insert_simp and Sup_insert_simp, which are subsumed
  2953 by Inf_insert and Sup_insert.  Lemmas Inf_UNIV and Sup_UNIV replace
  2954 former Inf_Univ and Sup_Univ.  Lemmas inf_top_right and sup_bot_right
  2955 subsume inf_top and sup_bot respectively.  INCOMPATIBILITY.
  2956 
  2957 * Reorganized theory Multiset: swapped notation of pointwise and
  2958 multiset order:
  2959 
  2960   - pointwise ordering is instance of class order with standard syntax
  2961     <= and <;
  2962   - multiset ordering has syntax <=# and <#; partial order properties
  2963     are provided by means of interpretation with prefix
  2964     multiset_order;
  2965   - less duplication, less historical organization of sections,
  2966     conversion from associations lists to multisets, rudimentary code
  2967     generation;
  2968   - use insert_DiffM2 [symmetric] instead of elem_imp_eq_diff_union,
  2969     if needed.
  2970 
  2971 Renamed:
  2972 
  2973   multiset_eq_conv_count_eq  ~>  multiset_ext_iff
  2974   multi_count_ext  ~>  multiset_ext
  2975   diff_union_inverse2  ~>  diff_union_cancelR
  2976 
  2977 INCOMPATIBILITY.
  2978 
  2979 * Theory Permutation: replaced local "remove" by List.remove1.
  2980 
  2981 * Code generation: ML and OCaml code is decorated with signatures.
  2982 
  2983 * Theory List: added transpose.
  2984 
  2985 * Library/Nat_Bijection.thy is a collection of bijective functions
  2986 between nat and other types, which supersedes the older libraries
  2987 Library/Nat_Int_Bij.thy and HOLCF/NatIso.thy.  INCOMPATIBILITY.
  2988 
  2989   Constants:
  2990   Nat_Int_Bij.nat2_to_nat         ~> prod_encode
  2991   Nat_Int_Bij.nat_to_nat2         ~> prod_decode
  2992   Nat_Int_Bij.int_to_nat_bij      ~> int_encode
  2993   Nat_Int_Bij.nat_to_int_bij      ~> int_decode
  2994   Countable.pair_encode           ~> prod_encode
  2995   NatIso.prod2nat                 ~> prod_encode
  2996   NatIso.nat2prod                 ~> prod_decode
  2997   NatIso.sum2nat                  ~> sum_encode
  2998   NatIso.nat2sum                  ~> sum_decode
  2999   NatIso.list2nat                 ~> list_encode
  3000   NatIso.nat2list                 ~> list_decode
  3001   NatIso.set2nat                  ~> set_encode
  3002   NatIso.nat2set                  ~> set_decode
  3003 
  3004   Lemmas:
  3005   Nat_Int_Bij.bij_nat_to_int_bij  ~> bij_int_decode
  3006   Nat_Int_Bij.nat2_to_nat_inj     ~> inj_prod_encode
  3007   Nat_Int_Bij.nat2_to_nat_surj    ~> surj_prod_encode
  3008   Nat_Int_Bij.nat_to_nat2_inj     ~> inj_prod_decode
  3009   Nat_Int_Bij.nat_to_nat2_surj    ~> surj_prod_decode
  3010   Nat_Int_Bij.i2n_n2i_id          ~> int_encode_inverse
  3011   Nat_Int_Bij.n2i_i2n_id          ~> int_decode_inverse
  3012   Nat_Int_Bij.surj_nat_to_int_bij ~> surj_int_encode
  3013   Nat_Int_Bij.surj_int_to_nat_bij ~> surj_int_decode
  3014   Nat_Int_Bij.inj_nat_to_int_bij  ~> inj_int_encode
  3015   Nat_Int_Bij.inj_int_to_nat_bij  ~> inj_int_decode
  3016   Nat_Int_Bij.bij_nat_to_int_bij  ~> bij_int_encode
  3017   Nat_Int_Bij.bij_int_to_nat_bij  ~> bij_int_decode
  3018 
  3019 * Sledgehammer:
  3020   - Renamed ATP commands:
  3021     atp_info     ~> sledgehammer running_atps
  3022     atp_kill     ~> sledgehammer kill_atps
  3023     atp_messages ~> sledgehammer messages
  3024     atp_minimize ~> sledgehammer minimize
  3025     print_atps   ~> sledgehammer available_atps
  3026     INCOMPATIBILITY.
  3027   - Added user's manual ("isabelle doc sledgehammer").
  3028   - Added option syntax and "sledgehammer_params" to customize
  3029     Sledgehammer's behavior.  See the manual for details.
  3030   - Modified the Isar proof reconstruction code so that it produces
  3031     direct proofs rather than proofs by contradiction.  (This feature
  3032     is still experimental.)
  3033   - Made Isar proof reconstruction work for SPASS, remote ATPs, and in
  3034     full-typed mode.
  3035   - Added support for TPTP syntax for SPASS via the "spass_tptp" ATP.
  3036 
  3037 * Nitpick:
  3038   - Added and implemented "binary_ints" and "bits" options.
  3039   - Added "std" option and implemented support for nonstandard models.
  3040   - Added and implemented "finitize" option to improve the precision
  3041     of infinite datatypes based on a monotonicity analysis.
  3042   - Added support for quotient types.
  3043   - Added support for "specification" and "ax_specification"
  3044     constructs.
  3045   - Added support for local definitions (for "function" and
  3046     "termination" proofs).
  3047   - Added support for term postprocessors.
  3048   - Optimized "Multiset.multiset" and "FinFun.finfun".
  3049   - Improved efficiency of "destroy_constrs" optimization.
  3050   - Fixed soundness bugs related to "destroy_constrs" optimization and
  3051     record getters.
  3052   - Fixed soundness bug related to higher-order constructors.
  3053   - Fixed soundness bug when "full_descrs" is enabled.
  3054   - Improved precision of set constructs.
  3055   - Added "atoms" option.
  3056   - Added cache to speed up repeated Kodkod invocations on the same
  3057     problems.
  3058   - Renamed "MiniSatJNI", "zChaffJNI", "BerkMinAlloy", and
  3059     "SAT4JLight" to "MiniSat_JNI", "zChaff_JNI", "BerkMin_Alloy", and
  3060     "SAT4J_Light".  INCOMPATIBILITY.
  3061   - Removed "skolemize", "uncurry", "sym_break", "flatten_prop",
  3062     "sharing_depth", and "show_skolems" options.  INCOMPATIBILITY.
  3063   - Removed "nitpick_intro" attribute.  INCOMPATIBILITY.
  3064 
  3065 * Method "induct" now takes instantiations of the form t, where t is not
  3066   a variable, as a shorthand for "x == t", where x is a fresh variable.
  3067   If this is not intended, t has to be enclosed in parentheses.
  3068   By default, the equalities generated by definitional instantiations
  3069   are pre-simplified, which may cause parameters of inductive cases
  3070   to disappear, or may even delete some of the inductive cases.
  3071   Use "induct (no_simp)" instead of "induct" to restore the old
  3072   behaviour. The (no_simp) option is also understood by the "cases"
  3073   and "nominal_induct" methods, which now perform pre-simplification, too.
  3074   INCOMPATIBILITY.
  3075 
  3076 
  3077 *** HOLCF ***
  3078 
  3079 * Variable names in lemmas generated by the domain package have
  3080 changed; the naming scheme is now consistent with the HOL datatype
  3081 package.  Some proof scripts may be affected, INCOMPATIBILITY.
  3082 
  3083 * The domain package no longer defines the function "foo_copy" for
  3084 recursive domain "foo".  The reach lemma is now stated directly in
  3085 terms of "foo_take".  Lemmas and proofs that mention "foo_copy" must
  3086 be reformulated in terms of "foo_take", INCOMPATIBILITY.
  3087 
  3088 * Most definedness lemmas generated by the domain package (previously
  3089 of the form "x ~= UU ==> foo$x ~= UU") now have an if-and-only-if form
  3090 like "foo$x = UU <-> x = UU", which works better as a simp rule.
  3091 Proofs that used definedness lemmas as intro rules may break,
  3092 potential INCOMPATIBILITY.
  3093 
  3094 * Induction and casedist rules generated by the domain package now
  3095 declare proper case_names (one called "bottom", and one named for each
  3096 constructor).  INCOMPATIBILITY.
  3097 
  3098 * For mutually-recursive domains, separate "reach" and "take_lemma"
  3099 rules are generated for each domain, INCOMPATIBILITY.
  3100 
  3101   foo_bar.reach       ~> foo.reach  bar.reach
  3102   foo_bar.take_lemmas ~> foo.take_lemma  bar.take_lemma
  3103 
  3104 * Some lemmas generated by the domain package have been renamed for
  3105 consistency with the datatype package, INCOMPATIBILITY.
  3106 
  3107   foo.ind        ~> foo.induct
  3108   foo.finite_ind ~> foo.finite_induct
  3109   foo.coind      ~> foo.coinduct
  3110   foo.casedist   ~> foo.exhaust
  3111   foo.exhaust    ~> foo.nchotomy
  3112 
  3113 * For consistency with other definition packages, the fixrec package
  3114 now generates qualified theorem names, INCOMPATIBILITY.
  3115 
  3116   foo_simps  ~> foo.simps
  3117   foo_unfold ~> foo.unfold
  3118   foo_induct ~> foo.induct
  3119 
  3120 * The "fixrec_simp" attribute has been removed.  The "fixrec_simp"
  3121 method and internal fixrec proofs now use the default simpset instead.
  3122 INCOMPATIBILITY.
  3123 
  3124 * The "contlub" predicate has been removed.  Proof scripts should use
  3125 lemma contI2 in place of monocontlub2cont, INCOMPATIBILITY.
  3126 
  3127 * The "admw" predicate has been removed, INCOMPATIBILITY.
  3128 
  3129 * The constants cpair, cfst, and csnd have been removed in favor of
  3130 Pair, fst, and snd from Isabelle/HOL, INCOMPATIBILITY.
  3131 
  3132 
  3133 *** ML ***
  3134 
  3135 * Antiquotations for basic formal entities:
  3136 
  3137     @{class NAME}         -- type class
  3138     @{class_syntax NAME}  -- syntax representation of the above
  3139 
  3140     @{type_name NAME}     -- logical type
  3141     @{type_abbrev NAME}   -- type abbreviation
  3142     @{nonterminal NAME}   -- type of concrete syntactic category
  3143     @{type_syntax NAME}   -- syntax representation of any of the above
  3144 
  3145     @{const_name NAME}    -- logical constant (INCOMPATIBILITY)
  3146     @{const_abbrev NAME}  -- abbreviated constant
  3147     @{const_syntax NAME}  -- syntax representation of any of the above
  3148 
  3149 * Antiquotation @{syntax_const NAME} ensures that NAME refers to a raw
  3150 syntax constant (cf. 'syntax' command).
  3151 
  3152 * Antiquotation @{make_string} inlines a function to print arbitrary
  3153 values similar to the ML toplevel.  The result is compiler dependent
  3154 and may fall back on "?" in certain situations.
  3155 
  3156 * Diagnostic commands 'ML_val' and 'ML_command' may refer to
  3157 antiquotations @{Isar.state} and @{Isar.goal}.  This replaces impure
  3158 Isar.state() and Isar.goal(), which belong to the old TTY loop and do
  3159 not work with the asynchronous Isar document model.
  3160 
  3161 * Configuration options now admit dynamic default values, depending on
  3162 the context or even global references.
  3163 
  3164 * SHA1.digest digests strings according to SHA-1 (see RFC 3174).  It
  3165 uses an efficient external library if available (for Poly/ML).
  3166 
  3167 * Renamed some important ML structures, while keeping the old names
  3168 for some time as aliases within the structure Legacy:
  3169 
  3170   OuterKeyword  ~>  Keyword
  3171   OuterLex      ~>  Token
  3172   OuterParse    ~>  Parse
  3173   OuterSyntax   ~>  Outer_Syntax
  3174   PrintMode     ~>  Print_Mode
  3175   SpecParse     ~>  Parse_Spec
  3176   ThyInfo       ~>  Thy_Info
  3177   ThyLoad       ~>  Thy_Load
  3178   ThyOutput     ~>  Thy_Output
  3179   TypeInfer     ~>  Type_Infer
  3180 
  3181 Note that "open Legacy" simplifies porting of sources, but forgetting
  3182 to remove it again will complicate porting again in the future.
  3183 
  3184 * Most operations that refer to a global context are named
  3185 accordingly, e.g. Simplifier.global_context or
  3186 ProofContext.init_global.  There are some situations where a global
  3187 context actually works, but under normal circumstances one needs to
  3188 pass the proper local context through the code!
  3189 
  3190 * Discontinued old TheoryDataFun with its copy/init operation -- data
  3191 needs to be pure.  Functor Theory_Data_PP retains the traditional
  3192 Pretty.pp argument to merge, which is absent in the standard
  3193 Theory_Data version.
  3194 
  3195 * Sorts.certify_sort and derived "cert" operations for types and terms
  3196 no longer minimize sorts.  Thus certification at the boundary of the
  3197 inference kernel becomes invariant under addition of class relations,
  3198 which is an important monotonicity principle.  Sorts are now minimized
  3199 in the syntax layer only, at the boundary between the end-user and the
  3200 system.  Subtle INCOMPATIBILITY, may have to use Sign.minimize_sort
  3201 explicitly in rare situations.
  3202 
  3203 * Renamed old-style Drule.standard to Drule.export_without_context, to
  3204 emphasize that this is in no way a standard operation.
  3205 INCOMPATIBILITY.
  3206 
  3207 * Subgoal.FOCUS (and variants): resulting goal state is normalized as
  3208 usual for resolution.  Rare INCOMPATIBILITY.
  3209 
  3210 * Renamed varify/unvarify operations to varify_global/unvarify_global
  3211 to emphasize that these only work in a global situation (which is
  3212 quite rare).
  3213 
  3214 * Curried take and drop in library.ML; negative length is interpreted
  3215 as infinity (as in chop).  Subtle INCOMPATIBILITY.
  3216 
  3217 * Proof terms: type substitutions on proof constants now use canonical
  3218 order of type variables.  INCOMPATIBILITY for tools working with proof
  3219 terms.
  3220 
  3221 * Raw axioms/defs may no longer carry sort constraints, and raw defs
  3222 may no longer carry premises.  User-level specifications are
  3223 transformed accordingly by Thm.add_axiom/add_def.
  3224 
  3225 
  3226 *** System ***
  3227 
  3228 * Discontinued special HOL_USEDIR_OPTIONS for the main HOL image;
  3229 ISABELLE_USEDIR_OPTIONS applies uniformly to all sessions.  Note that
  3230 proof terms are enabled unconditionally in the new HOL-Proofs image.
  3231 
  3232 * Discontinued old ISABELLE and ISATOOL environment settings (legacy
  3233 feature since Isabelle2009).  Use ISABELLE_PROCESS and ISABELLE_TOOL,
  3234 respectively.
  3235 
  3236 * Old lib/scripts/polyml-platform is superseded by the
  3237 ISABELLE_PLATFORM setting variable, which defaults to the 32 bit
  3238 variant, even on a 64 bit machine.  The following example setting
  3239 prefers 64 bit if available:
  3240 
  3241   ML_PLATFORM="${ISABELLE_PLATFORM64:-$ISABELLE_PLATFORM}"
  3242 
  3243 * The preliminary Isabelle/jEdit application demonstrates the emerging
  3244 Isabelle/Scala layer for advanced prover interaction and integration.
  3245 See src/Tools/jEdit or "isabelle jedit" provided by the properly built
  3246 component.
  3247 
  3248 * "IsabelleText" is a Unicode font derived from Bitstream Vera Mono
  3249 and Bluesky TeX fonts.  It provides the usual Isabelle symbols,
  3250 similar to the default assignment of the document preparation system
  3251 (cf. isabellesym.sty).  The Isabelle/Scala class Isabelle_System
  3252 provides some operations for direct access to the font without asking
  3253 the user for manual installation.
  3254 
  3255 
  3256 
  3257 New in Isabelle2009-1 (December 2009)
  3258 -------------------------------------
  3259 
  3260 *** General ***
  3261 
  3262 * Discontinued old form of "escaped symbols" such as \\<forall>.  Only
  3263 one backslash should be used, even in ML sources.
  3264 
  3265 
  3266 *** Pure ***
  3267 
  3268 * Locale interpretation propagates mixins along the locale hierarchy.
  3269 The currently only available mixins are the equations used to map
  3270 local definitions to terms of the target domain of an interpretation.
  3271 
  3272 * Reactivated diagnostic command 'print_interps'.  Use "print_interps
  3273 loc" to print all interpretations of locale "loc" in the theory.
  3274 Interpretations in proofs are not shown.
  3275 
  3276 * Thoroughly revised locales tutorial.  New section on conditional
  3277 interpretation.
  3278 
  3279 * On instantiation of classes, remaining undefined class parameters
  3280 are formally declared.  INCOMPATIBILITY.
  3281 
  3282 
  3283 *** Document preparation ***
  3284 
  3285 * New generalized style concept for printing terms: @{foo (style) ...}
  3286 instead of @{foo_style style ...}  (old form is still retained for
  3287 backward compatibility).  Styles can be also applied for
  3288 antiquotations prop, term_type and typeof.
  3289 
  3290 
  3291 *** HOL ***
  3292 
  3293 * New proof method "smt" for a combination of first-order logic with
  3294 equality, linear and nonlinear (natural/integer/real) arithmetic, and
  3295 fixed-size bitvectors; there is also basic support for higher-order
  3296 features (esp. lambda abstractions).  It is an incomplete decision
  3297 procedure based on external SMT solvers using the oracle mechanism;
  3298 for the SMT solver Z3, this method is proof-producing.  Certificates
  3299 are provided to avoid calling the external solvers solely for
  3300 re-checking proofs.  Due to a remote SMT service there is no need for
  3301 installing SMT solvers locally.  See src/HOL/SMT.
  3302 
  3303 * New commands to load and prove verification conditions generated by
  3304 the Boogie program verifier or derived systems (e.g. the Verifying C
  3305 Compiler (VCC) or Spec#).  See src/HOL/Boogie.
  3306 
  3307 * New counterexample generator tool 'nitpick' based on the Kodkod
  3308 relational model finder.  See src/HOL/Tools/Nitpick and
  3309 src/HOL/Nitpick_Examples.
  3310 
  3311 * New commands 'code_pred' and 'values' to invoke the predicate
  3312 compiler and to enumerate values of inductive predicates.
  3313 
  3314 * A tabled implementation of the reflexive transitive closure.
  3315 
  3316 * New implementation of quickcheck uses generic code generator;
  3317 default generators are provided for all suitable HOL types, records
  3318 and datatypes.  Old quickcheck can be re-activated importing theory
  3319 Library/SML_Quickcheck.
  3320 
  3321 * New testing tool Mirabelle for automated proof tools.  Applies
  3322 several tools and tactics like sledgehammer, metis, or quickcheck, to
  3323 every proof step in a theory.  To be used in batch mode via the
  3324 "mirabelle" utility.
  3325 
  3326 * New proof method "sos" (sum of squares) for nonlinear real
  3327 arithmetic (originally due to John Harison). It requires theory
  3328 Library/Sum_Of_Squares.  It is not a complete decision procedure but
  3329 works well in practice on quantifier-free real arithmetic with +, -,
  3330 *, ^, =, <= and <, i.e. boolean combinations of equalities and
  3331 inequalities between polynomials.  It makes use of external
  3332 semidefinite programming solvers.  Method "sos" generates a
  3333 certificate that can be pasted into the proof thus avoiding the need
  3334 to call an external tool every time the proof is checked.  See
  3335 src/HOL/Library/Sum_Of_Squares.
  3336 
  3337 * New method "linarith" invokes existing linear arithmetic decision
  3338 procedure only.
  3339 
  3340 * New command 'atp_minimal' reduces result produced by Sledgehammer.
  3341 
  3342 * New Sledgehammer option "Full Types" in Proof General settings menu.
  3343 Causes full type information to be output to the ATPs.  This slows
  3344 ATPs down considerably but eliminates a source of unsound "proofs"
  3345 that fail later.
  3346 
  3347 * New method "metisFT": A version of metis that uses full type
  3348 information in order to avoid failures of proof reconstruction.
  3349 
  3350 * New evaluator "approximate" approximates an real valued term using
  3351 the same method as the approximation method.
  3352 
  3353 * Method "approximate" now supports arithmetic expressions as
  3354 boundaries of intervals and implements interval splitting and Taylor
  3355 series expansion.
  3356 
  3357 * ML antiquotation @{code_datatype} inserts definition of a datatype
  3358 generated by the code generator; e.g. see src/HOL/Predicate.thy.
  3359 
  3360 * New theory SupInf of the supremum and infimum operators for sets of
  3361 reals.
  3362 
  3363 * New theory Probability, which contains a development of measure
  3364 theory, eventually leading to Lebesgue integration and probability.
  3365 
  3366 * Extended Multivariate Analysis to include derivation and Brouwer's
  3367 fixpoint theorem.
  3368 
  3369 * Reorganization of number theory, INCOMPATIBILITY:
  3370   - new number theory development for nat and int, in theories Divides
  3371     and GCD as well as in new session Number_Theory
  3372   - some constants and facts now suffixed with _nat and _int
  3373     accordingly
  3374   - former session NumberTheory now named Old_Number_Theory, including
  3375     theories Legacy_GCD and Primes (prefer Number_Theory if possible)
  3376   - moved theory Pocklington from src/HOL/Library to
  3377     src/HOL/Old_Number_Theory
  3378 
  3379 * Theory GCD includes functions Gcd/GCD and Lcm/LCM for the gcd and
  3380 lcm of finite and infinite sets. It is shown that they form a complete
  3381 lattice.
  3382 
  3383 * Class semiring_div requires superclass no_zero_divisors and proof of
  3384 div_mult_mult1; theorems div_mult_mult1, div_mult_mult2,
  3385 div_mult_mult1_if, div_mult_mult1 and div_mult_mult2 have been
  3386 generalized to class semiring_div, subsuming former theorems
  3387 zdiv_zmult_zmult1, zdiv_zmult_zmult1_if, zdiv_zmult_zmult1 and
  3388 zdiv_zmult_zmult2.  div_mult_mult1 is now [simp] by default.
  3389 INCOMPATIBILITY.
  3390 
  3391 * Refinements to lattice classes and sets:
  3392   - less default intro/elim rules in locale variant, more default
  3393     intro/elim rules in class variant: more uniformity
  3394   - lemma ge_sup_conv renamed to le_sup_iff, in accordance with
  3395     le_inf_iff
  3396   - dropped lemma alias inf_ACI for inf_aci (same for sup_ACI and
  3397     sup_aci)
  3398   - renamed ACI to inf_sup_aci
  3399   - new class "boolean_algebra"
  3400   - class "complete_lattice" moved to separate theory
  3401     "Complete_Lattice"; corresponding constants (and abbreviations)
  3402     renamed and with authentic syntax:
  3403     Set.Inf ~>    Complete_Lattice.Inf
  3404     Set.Sup ~>    Complete_Lattice.Sup
  3405     Set.INFI ~>   Complete_Lattice.INFI
  3406     Set.SUPR ~>   Complete_Lattice.SUPR
  3407     Set.Inter ~>  Complete_Lattice.Inter
  3408     Set.Union ~>  Complete_Lattice.Union
  3409     Set.INTER ~>  Complete_Lattice.INTER
  3410     Set.UNION ~>  Complete_Lattice.UNION
  3411   - authentic syntax for
  3412     Set.Pow
  3413     Set.image
  3414   - mere abbreviations:
  3415     Set.empty               (for bot)
  3416     Set.UNIV                (for top)
  3417     Set.inter               (for inf, formerly Set.Int)
  3418     Set.union               (for sup, formerly Set.Un)
  3419     Complete_Lattice.Inter  (for Inf)
  3420     Complete_Lattice.Union  (for Sup)
  3421     Complete_Lattice.INTER  (for INFI)
  3422     Complete_Lattice.UNION  (for SUPR)
  3423   - object-logic definitions as far as appropriate
  3424 
  3425 INCOMPATIBILITY.  Care is required when theorems Int_subset_iff or
  3426 Un_subset_iff are explicitly deleted as default simp rules; then also
  3427 their lattice counterparts le_inf_iff and le_sup_iff have to be
  3428 deleted to achieve the desired effect.
  3429 
  3430 * Rules inf_absorb1, inf_absorb2, sup_absorb1, sup_absorb2 are no simp
  3431 rules by default any longer; the same applies to min_max.inf_absorb1
  3432 etc.  INCOMPATIBILITY.
  3433 
  3434 * Rules sup_Int_eq and sup_Un_eq are no longer declared as
  3435 pred_set_conv by default.  INCOMPATIBILITY.
  3436 
  3437 * Power operations on relations and functions are now one dedicated
  3438 constant "compow" with infix syntax "^^".  Power operation on
  3439 multiplicative monoids retains syntax "^" and is now defined generic
  3440 in class power.  INCOMPATIBILITY.
  3441 
  3442 * Relation composition "R O S" now has a more standard argument order:
  3443 "R O S = {(x, z). EX y. (x, y) : R & (y, z) : S}".  INCOMPATIBILITY,
  3444 rewrite propositions with "S O R" --> "R O S". Proofs may occasionally
  3445 break, since the O_assoc rule was not rewritten like this.  Fix using
  3446 O_assoc[symmetric].  The same applies to the curried version "R OO S".
  3447 
  3448 * Function "Inv" is renamed to "inv_into" and function "inv" is now an
  3449 abbreviation for "inv_into UNIV".  Lemmas are renamed accordingly.
  3450 INCOMPATIBILITY.
  3451 
  3452 * Most rules produced by inductive and datatype package have mandatory
  3453 prefixes.  INCOMPATIBILITY.
  3454 
  3455 * Changed "DERIV_intros" to a dynamic fact, which can be augmented by
  3456 the attribute of the same name.  Each of the theorems in the list
  3457 DERIV_intros assumes composition with an additional function and
  3458 matches a variable to the derivative, which has to be solved by the
  3459 Simplifier.  Hence (auto intro!: DERIV_intros) computes the derivative
  3460 of most elementary terms.  Former Maclauren.DERIV_tac and
  3461 Maclauren.deriv_tac should be replaced by (auto intro!: DERIV_intros).
  3462 INCOMPATIBILITY.
  3463 
  3464 * Code generator attributes follow the usual underscore convention:
  3465     code_unfold     replaces    code unfold
  3466     code_post       replaces    code post
  3467     etc.
  3468   INCOMPATIBILITY.
  3469 
  3470 * Renamed methods:
  3471     sizechange -> size_change
  3472     induct_scheme -> induction_schema
  3473   INCOMPATIBILITY.
  3474 
  3475 * Discontinued abbreviation "arbitrary" of constant "undefined".
  3476 INCOMPATIBILITY, use "undefined" directly.
  3477 
  3478 * Renamed theorems:
  3479     Suc_eq_add_numeral_1 -> Suc_eq_plus1
  3480     Suc_eq_add_numeral_1_left -> Suc_eq_plus1_left
  3481     Suc_plus1 -> Suc_eq_plus1
  3482     *anti_sym -> *antisym*
  3483     vector_less_eq_def -> vector_le_def
  3484   INCOMPATIBILITY.
  3485 
  3486 * Added theorem List.map_map as [simp].  Removed List.map_compose.
  3487 INCOMPATIBILITY.
  3488 
  3489 * Removed predicate "M hassize n" (<--> card M = n & finite M).
  3490 INCOMPATIBILITY.
  3491 
  3492 
  3493 *** HOLCF ***
  3494 
  3495 * Theory Representable defines a class "rep" of domains that are
  3496 representable (via an ep-pair) in the universal domain type "udom".
  3497 Instances are provided for all type constructors defined in HOLCF.
  3498 
  3499 * The 'new_domain' command is a purely definitional version of the
  3500 domain package, for representable domains.  Syntax is identical to the
  3501 old domain package.  The 'new_domain' package also supports indirect
  3502 recursion using previously-defined type constructors.  See
  3503 src/HOLCF/ex/New_Domain.thy for examples.
  3504 
  3505 * Method "fixrec_simp" unfolds one step of a fixrec-defined constant
  3506 on the left-hand side of an equation, and then performs
  3507 simplification.  Rewriting is done using rules declared with the
  3508 "fixrec_simp" attribute.  The "fixrec_simp" method is intended as a
  3509 replacement for "fixpat"; see src/HOLCF/ex/Fixrec_ex.thy for examples.
  3510 
  3511 * The pattern-match compiler in 'fixrec' can now handle constructors
  3512 with HOL function types.  Pattern-match combinators for the Pair
  3513 constructor are pre-configured.
  3514 
  3515 * The 'fixrec' package now produces better fixed-point induction rules
  3516 for mutually-recursive definitions:  Induction rules have conclusions
  3517 of the form "P foo bar" instead of "P <foo, bar>".
  3518 
  3519 * The constant "sq_le" (with infix syntax "<<" or "\<sqsubseteq>") has
  3520 been renamed to "below".  The name "below" now replaces "less" in many
  3521 theorem names.  (Legacy theorem names using "less" are still supported
  3522 as well.)
  3523 
  3524 * The 'fixrec' package now supports "bottom patterns".  Bottom
  3525 patterns can be used to generate strictness rules, or to make
  3526 functions more strict (much like the bang-patterns supported by the
  3527 Glasgow Haskell Compiler).  See src/HOLCF/ex/Fixrec_ex.thy for
  3528 examples.
  3529 
  3530 
  3531 *** ML ***
  3532 
  3533 * Support for Poly/ML 5.3.0, with improved reporting of compiler
  3534 errors and run-time exceptions, including detailed source positions.
  3535 
  3536 * Structure Name_Space (formerly NameSpace) now manages uniquely
  3537 identified entries, with some additional information such as source
  3538 position, logical grouping etc.
  3539 
  3540 * Theory and context data is now introduced by the simplified and
  3541 modernized functors Theory_Data, Proof_Data, Generic_Data.  Data needs
  3542 to be pure, but the old TheoryDataFun for mutable data (with explicit
  3543 copy operation) is still available for some time.
  3544 
  3545 * Structure Synchronized (cf. src/Pure/Concurrent/synchronized.ML)
  3546 provides a high-level programming interface to synchronized state
  3547 variables with atomic update.  This works via pure function
  3548 application within a critical section -- its runtime should be as
  3549 short as possible; beware of deadlocks if critical code is nested,
  3550 either directly or indirectly via other synchronized variables!
  3551 
  3552 * Structure Unsynchronized (cf. src/Pure/ML-Systems/unsynchronized.ML)
  3553 wraps raw ML references, explicitly indicating their non-thread-safe
  3554 behaviour.  The Isar toplevel keeps this structure open, to
  3555 accommodate Proof General as well as quick and dirty interactive
  3556 experiments with references.
  3557 
  3558 * PARALLEL_CHOICE and PARALLEL_GOALS provide basic support for
  3559 parallel tactical reasoning.
  3560 
  3561 * Tacticals Subgoal.FOCUS, Subgoal.FOCUS_PREMS, Subgoal.FOCUS_PARAMS
  3562 are similar to SUBPROOF, but are slightly more flexible: only the
  3563 specified parts of the subgoal are imported into the context, and the
  3564 body tactic may introduce new subgoals and schematic variables.
  3565 
  3566 * Old tactical METAHYPS, which does not observe the proof context, has
  3567 been renamed to Old_Goals.METAHYPS and awaits deletion.  Use SUBPROOF
  3568 or Subgoal.FOCUS etc.
  3569 
  3570 * Renamed functor TableFun to Table, and GraphFun to Graph.  (Since
  3571 functors have their own ML name space there is no point to mark them
  3572 separately.)  Minor INCOMPATIBILITY.
  3573 
  3574 * Renamed NamedThmsFun to Named_Thms.  INCOMPATIBILITY.
  3575 
  3576 * Renamed several structures FooBar to Foo_Bar.  Occasional,
  3577 INCOMPATIBILITY.
  3578 
  3579 * Operations of structure Skip_Proof no longer require quick_and_dirty
  3580 mode, which avoids critical setmp.
  3581 
  3582 * Eliminated old Attrib.add_attributes, Method.add_methods and related
  3583 combinators for "args".  INCOMPATIBILITY, need to use simplified
  3584 Attrib/Method.setup introduced in Isabelle2009.
  3585 
  3586 * Proper context for simpset_of, claset_of, clasimpset_of.  May fall
  3587 back on global_simpset_of, global_claset_of, global_clasimpset_of as
  3588 last resort.  INCOMPATIBILITY.
  3589 
  3590 * Display.pretty_thm now requires a proper context (cf. former
  3591 ProofContext.pretty_thm).  May fall back on Display.pretty_thm_global
  3592 or even Display.pretty_thm_without_context as last resort.
  3593 INCOMPATIBILITY.
  3594 
  3595 * Discontinued Display.pretty_ctyp/cterm etc.  INCOMPATIBILITY, use
  3596 Syntax.pretty_typ/term directly, preferably with proper context
  3597 instead of global theory.
  3598 
  3599 
  3600 *** System ***
  3601 
  3602 * Further fine tuning of parallel proof checking, scales up to 8 cores
  3603 (max. speedup factor 5.0).  See also Goal.parallel_proofs in ML and
  3604 usedir option -q.
  3605 
  3606 * Support for additional "Isabelle components" via etc/components, see
  3607 also the system manual.
  3608 
  3609 * The isabelle makeall tool now operates on all components with
  3610 IsaMakefile, not just hardwired "logics".
  3611 
  3612 * Removed "compress" option from isabelle-process and isabelle usedir;
  3613 this is always enabled.
  3614 
  3615 * Discontinued support for Poly/ML 4.x versions.
  3616 
  3617 * Isabelle tool "wwwfind" provides web interface for 'find_theorems'
  3618 on a given logic image.  This requires the lighttpd webserver and is
  3619 currently supported on Linux only.
  3620 
  3621 
  3622 
  3623 New in Isabelle2009 (April 2009)
  3624 --------------------------------
  3625 
  3626 *** General ***
  3627 
  3628 * Simplified main Isabelle executables, with less surprises on
  3629 case-insensitive file-systems (such as Mac OS).
  3630 
  3631   - The main Isabelle tool wrapper is now called "isabelle" instead of
  3632     "isatool."
  3633 
  3634   - The former "isabelle" alias for "isabelle-process" has been
  3635     removed (should rarely occur to regular users).
  3636 
  3637   - The former "isabelle-interface" and its alias "Isabelle" have been
  3638     removed (interfaces are now regular Isabelle tools).
  3639 
  3640 Within scripts and make files, the Isabelle environment variables
  3641 ISABELLE_TOOL and ISABELLE_PROCESS replace old ISATOOL and ISABELLE,
  3642 respectively.  (The latter are still available as legacy feature.)
  3643 
  3644 The old isabelle-interface wrapper could react in confusing ways if
  3645 the interface was uninstalled or changed otherwise.  Individual
  3646 interface tool configuration is now more explicit, see also the
  3647 Isabelle system manual.  In particular, Proof General is now available
  3648 via "isabelle emacs".
  3649 
  3650 INCOMPATIBILITY, need to adapt derivative scripts.  Users may need to
  3651 purge installed copies of Isabelle executables and re-run "isabelle
  3652 install -p ...", or use symlinks.
  3653 
  3654 * The default for ISABELLE_HOME_USER is now ~/.isabelle instead of the
  3655 old ~/isabelle, which was slightly non-standard and apt to cause
  3656 surprises on case-insensitive file-systems (such as Mac OS).
  3657 
  3658 INCOMPATIBILITY, need to move existing ~/isabelle/etc,
  3659 ~/isabelle/heaps, ~/isabelle/browser_info to the new place.  Special
  3660 care is required when using older releases of Isabelle.  Note that
  3661 ISABELLE_HOME_USER can be changed in Isabelle/etc/settings of any
  3662 Isabelle distribution, in order to use the new ~/.isabelle uniformly.
  3663 
  3664 * Proofs of fully specified statements are run in parallel on
  3665 multi-core systems.  A speedup factor of 2.5 to 3.2 can be expected on
  3666 a regular 4-core machine, if the initial heap space is made reasonably
  3667 large (cf. Poly/ML option -H).  (Requires Poly/ML 5.2.1 or later.)
  3668 
  3669 * The main reference manuals ("isar-ref", "implementation", and
  3670 "system") have been updated and extended.  Formally checked references
  3671 as hyperlinks are now available uniformly.
  3672 
  3673 
  3674 *** Pure ***
  3675 
  3676 * Complete re-implementation of locales.  INCOMPATIBILITY in several
  3677 respects.  The most important changes are listed below.  See the
  3678 Tutorial on Locales ("locales" manual) for details.
  3679 
  3680 - In locale expressions, instantiation replaces renaming.  Parameters
  3681 must be declared in a for clause.  To aid compatibility with previous
  3682 parameter inheritance, in locale declarations, parameters that are not
  3683 'touched' (instantiation position "_" or omitted) are implicitly added
  3684 with their syntax at the beginning of the for clause.
  3685 
  3686 - Syntax from abbreviations and definitions in locales is available in
  3687 locale expressions and context elements.  The latter is particularly
  3688 useful in locale declarations.
  3689 
  3690 - More flexible mechanisms to qualify names generated by locale
  3691 expressions.  Qualifiers (prefixes) may be specified in locale
  3692 expressions, and can be marked as mandatory (syntax: "name!:") or
  3693 optional (syntax "name?:").  The default depends for plain "name:"
  3694 depends on the situation where a locale expression is used: in
  3695 commands 'locale' and 'sublocale' prefixes are optional, in
  3696 'interpretation' and 'interpret' prefixes are mandatory.  The old
  3697 implicit qualifiers derived from the parameter names of a locale are
  3698 no longer generated.
  3699 
  3700 - Command "sublocale l < e" replaces "interpretation l < e".  The
  3701 instantiation clause in "interpretation" and "interpret" (square
  3702 brackets) is no longer available.  Use locale expressions.
  3703 
  3704 - When converting proof scripts, mandatory qualifiers in
  3705 'interpretation' and 'interpret' should be retained by default, even
  3706 if this is an INCOMPATIBILITY compared to former behavior.  In the
  3707 worst case, use the "name?:" form for non-mandatory ones.  Qualifiers
  3708 in locale expressions range over a single locale instance only.
  3709 
  3710 - Dropped locale element "includes".  This is a major INCOMPATIBILITY.
  3711 In existing theorem specifications replace the includes element by the
  3712 respective context elements of the included locale, omitting those
  3713 that are already present in the theorem specification.  Multiple
  3714 assume elements of a locale should be replaced by a single one
  3715 involving the locale predicate.  In the proof body, declarations (most
  3716 notably theorems) may be regained by interpreting the respective
  3717 locales in the proof context as required (command "interpret").
  3718 
  3719 If using "includes" in replacement of a target solely because the
  3720 parameter types in the theorem are not as general as in the target,
  3721 consider declaring a new locale with additional type constraints on
  3722 the parameters (context element "constrains").
  3723 
  3724 - Discontinued "locale (open)".  INCOMPATIBILITY.
  3725 
  3726 - Locale interpretation commands no longer attempt to simplify goal.
  3727 INCOMPATIBILITY: in rare situations the generated goal differs.  Use
  3728 methods intro_locales and unfold_locales to clarify.
  3729 
  3730 - Locale interpretation commands no longer accept interpretation
  3731 attributes.  INCOMPATIBILITY.
  3732 
  3733 * Class declaration: so-called "base sort" must not be given in import
  3734 list any longer, but is inferred from the specification.  Particularly
  3735 in HOL, write
  3736 
  3737     class foo = ...
  3738 
  3739 instead of
  3740 
  3741     class foo = type + ...
  3742 
  3743 * Class target: global versions of theorems stemming do not carry a
  3744 parameter prefix any longer.  INCOMPATIBILITY.
  3745 
  3746 * Class 'instance' command no longer accepts attached definitions.
  3747 INCOMPATIBILITY, use proper 'instantiation' target instead.
  3748 
  3749 * Recovered hiding of consts, which was accidentally broken in
  3750 Isabelle2007.  Potential INCOMPATIBILITY, ``hide const c'' really
  3751 makes c inaccessible; consider using ``hide (open) const c'' instead.
  3752 
  3753 * Slightly more coherent Pure syntax, with updated documentation in
  3754 isar-ref manual.  Removed locales meta_term_syntax and
  3755 meta_conjunction_syntax: TERM and &&& (formerly &&) are now permanent,
  3756 INCOMPATIBILITY in rare situations.  Note that &&& should not be used
  3757 directly in regular applications.
  3758 
  3759 * There is a new syntactic category "float_const" for signed decimal
  3760 fractions (e.g. 123.45 or -123.45).
  3761 
  3762 * Removed exotic 'token_translation' command.  INCOMPATIBILITY, use ML
  3763 interface with 'setup' command instead.
  3764 
  3765 * Command 'local_setup' is similar to 'setup', but operates on a local
  3766 theory context.
  3767 
  3768 * The 'axiomatization' command now only works within a global theory
  3769 context.  INCOMPATIBILITY.
  3770 
  3771 * Goal-directed proof now enforces strict proof irrelevance wrt. sort
  3772 hypotheses.  Sorts required in the course of reasoning need to be
  3773 covered by the constraints in the initial statement, completed by the
  3774 type instance information of the background theory.  Non-trivial sort
  3775 hypotheses, which rarely occur in practice, may be specified via
  3776 vacuous propositions of the form SORT_CONSTRAINT('a::c).  For example:
  3777 
  3778   lemma assumes "SORT_CONSTRAINT('a::empty)" shows False ...
  3779 
  3780 The result contains an implicit sort hypotheses as before --
  3781 SORT_CONSTRAINT premises are eliminated as part of the canonical rule
  3782 normalization.
  3783 
  3784 * Generalized Isar history, with support for linear undo, direct state
  3785 addressing etc.
  3786 
  3787 * Changed defaults for unify configuration options:
  3788 
  3789   unify_trace_bound = 50 (formerly 25)
  3790   unify_search_bound = 60 (formerly 30)
  3791 
  3792 * Different bookkeeping for code equations (INCOMPATIBILITY):
  3793 
  3794   a) On theory merge, the last set of code equations for a particular
  3795      constant is taken (in accordance with the policy applied by other
  3796      parts of the code generator framework).
  3797 
  3798   b) Code equations stemming from explicit declarations (e.g. code
  3799      attribute) gain priority over default code equations stemming
  3800      from definition, primrec, fun etc.
  3801 
  3802 * Keyword 'code_exception' now named 'code_abort'.  INCOMPATIBILITY.
  3803 
  3804 * Unified theorem tables for both code generators.  Thus [code
  3805 func] has disappeared and only [code] remains.  INCOMPATIBILITY.
  3806 
  3807 * Command 'find_consts' searches for constants based on type and name
  3808 patterns, e.g.
  3809 
  3810     find_consts "_ => bool"
  3811 
  3812 By default, matching is against subtypes, but it may be restricted to
  3813 the whole type.  Searching by name is possible.  Multiple queries are
  3814 conjunctive and queries may be negated by prefixing them with a
  3815 hyphen:
  3816 
  3817     find_consts strict: "_ => bool" name: "Int" -"int => int"
  3818 
  3819 * New 'find_theorems' criterion "solves" matches theorems that
  3820 directly solve the current goal (modulo higher-order unification).
  3821 
  3822 * Auto solve feature for main theorem statements: whenever a new goal
  3823 is stated, "find_theorems solves" is called; any theorems that could
  3824 solve the lemma directly are listed as part of the goal state.
  3825 Cf. associated options in Proof General Isabelle settings menu,
  3826 enabled by default, with reasonable timeout for pathological cases of
  3827 higher-order unification.
  3828 
  3829 
  3830 *** Document preparation ***
  3831 
  3832 * Antiquotation @{lemma} now imitates a regular terminal proof,
  3833 demanding keyword 'by' and supporting the full method expression
  3834 syntax just like the Isar command 'by'.
  3835 
  3836 
  3837 *** HOL ***
  3838 
  3839 * Integrated main parts of former image HOL-Complex with HOL.  Entry
  3840 points Main and Complex_Main remain as before.
  3841 
  3842 * Logic image HOL-Plain provides a minimal HOL with the most important
  3843 tools available (inductive, datatype, primrec, ...).  This facilitates
  3844 experimentation and tool development.  Note that user applications
  3845 (and library theories) should never refer to anything below theory
  3846 Main, as before.
  3847 
  3848 * Logic image HOL-Main stops at theory Main, and thus facilitates
  3849 experimentation due to shorter build times.
  3850 
  3851 * Logic image HOL-NSA contains theories of nonstandard analysis which
  3852 were previously part of former HOL-Complex.  Entry point Hyperreal
  3853 remains valid, but theories formerly using Complex_Main should now use
  3854 new entry point Hypercomplex.
  3855 
  3856 * Generic ATP manager for Sledgehammer, based on ML threads instead of
  3857 Posix processes.  Avoids potentially expensive forking of the ML
  3858 process.  New thread-based implementation also works on non-Unix
  3859 platforms (Cygwin).  Provers are no longer hardwired, but defined
  3860 within the theory via plain ML wrapper functions.  Basic Sledgehammer
  3861 commands are covered in the isar-ref manual.
  3862 
  3863 * Wrapper scripts for remote SystemOnTPTP service allows to use
  3864 sledgehammer without local ATP installation (Vampire etc.). Other
  3865 provers may be included via suitable ML wrappers, see also
  3866 src/HOL/ATP_Linkup.thy.
  3867 
  3868 * ATP selection (E/Vampire/Spass) is now via Proof General's settings
  3869 menu.
  3870 
  3871 * The metis method no longer fails because the theorem is too trivial
  3872 (contains the empty clause).
  3873 
  3874 * The metis method now fails in the usual manner, rather than raising
  3875 an exception, if it determines that it cannot prove the theorem.
  3876 
  3877 * Method "coherent" implements a prover for coherent logic (see also
  3878 src/Tools/coherent.ML).
  3879 
  3880 * Constants "undefined" and "default" replace "arbitrary".  Usually
  3881 "undefined" is the right choice to replace "arbitrary", though
  3882 logically there is no difference.  INCOMPATIBILITY.
  3883 
  3884 * Command "value" now integrates different evaluation mechanisms.  The
  3885 result of the first successful evaluation mechanism is printed.  In
  3886 square brackets a particular named evaluation mechanisms may be
  3887 specified (currently, [SML], [code] or [nbe]).  See further
  3888 src/HOL/ex/Eval_Examples.thy.
  3889 
  3890 * Normalization by evaluation now allows non-leftlinear equations.
  3891 Declare with attribute [code nbe].
  3892 
  3893 * Methods "case_tac" and "induct_tac" now refer to the very same rules
  3894 as the structured Isar versions "cases" and "induct", cf. the
  3895 corresponding "cases" and "induct" attributes.  Mutual induction rules
  3896 are now presented as a list of individual projections
  3897 (e.g. foo_bar.inducts for types foo and bar); the old format with
  3898 explicit HOL conjunction is no longer supported.  INCOMPATIBILITY, in
  3899 rare situations a different rule is selected --- notably nested tuple
  3900 elimination instead of former prod.exhaust: use explicit (case_tac t
  3901 rule: prod.exhaust) here.
  3902 
  3903 * Attributes "cases", "induct", "coinduct" support "del" option.
  3904 
  3905 * Removed fact "case_split_thm", which duplicates "case_split".
  3906 
  3907 * The option datatype has been moved to a new theory Option.  Renamed
  3908 option_map to Option.map, and o2s to Option.set, INCOMPATIBILITY.
  3909 
  3910 * New predicate "strict_mono" classifies strict functions on partial
  3911 orders.  With strict functions on linear orders, reasoning about
  3912 (in)equalities is facilitated by theorems "strict_mono_eq",
  3913 "strict_mono_less_eq" and "strict_mono_less".
  3914 
  3915 * Some set operations are now proper qualified constants with
  3916 authentic syntax.  INCOMPATIBILITY:
  3917 
  3918     op Int ~>   Set.Int
  3919     op Un ~>    Set.Un
  3920     INTER ~>    Set.INTER
  3921     UNION ~>    Set.UNION
  3922     Inter ~>    Set.Inter
  3923     Union ~>    Set.Union
  3924     {} ~>       Set.empty
  3925     UNIV ~>     Set.UNIV
  3926 
  3927 * Class complete_lattice with operations Inf, Sup, INFI, SUPR now in
  3928 theory Set.
  3929 
  3930 * Auxiliary class "itself" has disappeared -- classes without any
  3931 parameter are treated as expected by the 'class' command.
  3932 
  3933 * Leibnitz's Series for Pi and the arcus tangens and logarithm series.
  3934 
  3935 * Common decision procedures (Cooper, MIR, Ferrack, Approximation,
  3936 Dense_Linear_Order) are now in directory HOL/Decision_Procs.
  3937 
  3938 * Theory src/HOL/Decision_Procs/Approximation provides the new proof
  3939 method "approximation".  It proves formulas on real values by using
  3940 interval arithmetic.  In the formulas are also the transcendental
  3941 functions sin, cos, tan, atan, ln, exp and the constant pi are
  3942 allowed. For examples see
  3943 src/HOL/Descision_Procs/ex/Approximation_Ex.thy.
  3944 
  3945 * Theory "Reflection" now resides in HOL/Library.
  3946 
  3947 * Entry point to Word library now simply named "Word".
  3948 INCOMPATIBILITY.
  3949 
  3950 * Made source layout more coherent with logical distribution
  3951 structure:
  3952 
  3953     src/HOL/Library/RType.thy ~> src/HOL/Typerep.thy
  3954     src/HOL/Library/Code_Message.thy ~> src/HOL/
  3955     src/HOL/Library/GCD.thy ~> src/HOL/
  3956     src/HOL/Library/Order_Relation.thy ~> src/HOL/
  3957     src/HOL/Library/Parity.thy ~> src/HOL/
  3958     src/HOL/Library/Univ_Poly.thy ~> src/HOL/
  3959     src/HOL/Real/ContNotDenum.thy ~> src/HOL/Library/
  3960     src/HOL/Real/Lubs.thy ~> src/HOL/
  3961     src/HOL/Real/PReal.thy ~> src/HOL/
  3962     src/HOL/Real/Rational.thy ~> src/HOL/
  3963     src/HOL/Real/RComplete.thy ~> src/HOL/
  3964     src/HOL/Real/RealDef.thy ~> src/HOL/
  3965     src/HOL/Real/RealPow.thy ~> src/HOL/
  3966     src/HOL/Real/Real.thy ~> src/HOL/
  3967     src/HOL/Complex/Complex_Main.thy ~> src/HOL/
  3968     src/HOL/Complex/Complex.thy ~> src/HOL/
  3969     src/HOL/Complex/FrechetDeriv.thy ~> src/HOL/Library/
  3970     src/HOL/Complex/Fundamental_Theorem_Algebra.thy ~> src/HOL/Library/
  3971     src/HOL/Hyperreal/Deriv.thy ~> src/HOL/
  3972     src/HOL/Hyperreal/Fact.thy ~> src/HOL/
  3973     src/HOL/Hyperreal/Integration.thy ~> src/HOL/
  3974     src/HOL/Hyperreal/Lim.thy ~> src/HOL/
  3975     src/HOL/Hyperreal/Ln.thy ~> src/HOL/
  3976     src/HOL/Hyperreal/Log.thy ~> src/HOL/
  3977     src/HOL/Hyperreal/MacLaurin.thy ~> src/HOL/
  3978     src/HOL/Hyperreal/NthRoot.thy ~> src/HOL/
  3979     src/HOL/Hyperreal/Series.thy ~> src/HOL/
  3980     src/HOL/Hyperreal/SEQ.thy ~> src/HOL/
  3981     src/HOL/Hyperreal/Taylor.thy ~> src/HOL/
  3982     src/HOL/Hyperreal/Transcendental.thy ~> src/HOL/
  3983     src/HOL/Real/Float ~> src/HOL/Library/
  3984     src/HOL/Real/HahnBanach ~> src/HOL/HahnBanach
  3985     src/HOL/Real/RealVector.thy ~> src/HOL/
  3986 
  3987     src/HOL/arith_data.ML ~> src/HOL/Tools
  3988     src/HOL/hologic.ML ~> src/HOL/Tools
  3989     src/HOL/simpdata.ML ~> src/HOL/Tools
  3990     src/HOL/int_arith1.ML ~> src/HOL/Tools/int_arith.ML
  3991     src/HOL/int_factor_simprocs.ML ~> src/HOL/Tools
  3992     src/HOL/nat_simprocs.ML ~> src/HOL/Tools
  3993     src/HOL/Real/float_arith.ML ~> src/HOL/Tools
  3994     src/HOL/Real/float_syntax.ML ~> src/HOL/Tools
  3995     src/HOL/Real/rat_arith.ML ~> src/HOL/Tools
  3996     src/HOL/Real/real_arith.ML ~> src/HOL/Tools
  3997 
  3998     src/HOL/Library/Array.thy ~> src/HOL/Imperative_HOL
  3999     src/HOL/Library/Heap_Monad.thy ~> src/HOL/Imperative_HOL
  4000     src/HOL/Library/Heap.thy ~> src/HOL/Imperative_HOL
  4001     src/HOL/Library/Imperative_HOL.thy ~> src/HOL/Imperative_HOL
  4002     src/HOL/Library/Ref.thy ~> src/HOL/Imperative_HOL
  4003     src/HOL/Library/Relational.thy ~> src/HOL/Imperative_HOL
  4004 
  4005 * If methods "eval" and "evaluation" encounter a structured proof
  4006 state with !!/==>, only the conclusion is evaluated to True (if
  4007 possible), avoiding strange error messages.
  4008 
  4009 * Method "sizechange" automates termination proofs using (a
  4010 modification of) the size-change principle.  Requires SAT solver.  See
  4011 src/HOL/ex/Termination.thy for examples.
  4012 
  4013 * Simplifier: simproc for let expressions now unfolds if bound
  4014 variable occurs at most once in let expression body.  INCOMPATIBILITY.
  4015 
  4016 * Method "arith": Linear arithmetic now ignores all inequalities when
  4017 fast_arith_neq_limit is exceeded, instead of giving up entirely.
  4018 
  4019 * New attribute "arith" for facts that should always be used
  4020 automatically by arithmetic. It is intended to be used locally in
  4021 proofs, e.g.
  4022 
  4023   assumes [arith]: "x > 0"
  4024 
  4025 Global usage is discouraged because of possible performance impact.
  4026 
  4027 * New classes "top" and "bot" with corresponding operations "top" and
  4028 "bot" in theory Orderings; instantiation of class "complete_lattice"
  4029 requires instantiation of classes "top" and "bot".  INCOMPATIBILITY.
  4030 
  4031 * Changed definition lemma "less_fun_def" in order to provide an
  4032 instance for preorders on functions; use lemma "less_le" instead.
  4033 INCOMPATIBILITY.
  4034 
  4035 * Theory Orderings: class "wellorder" moved here, with explicit
  4036 induction rule "less_induct" as assumption.  For instantiation of
  4037 "wellorder" by means of predicate "wf", use rule wf_wellorderI.
  4038 INCOMPATIBILITY.
  4039 
  4040 * Theory Orderings: added class "preorder" as superclass of "order".
  4041 INCOMPATIBILITY: Instantiation proofs for order, linorder
  4042 etc. slightly changed.  Some theorems named order_class.* now named
  4043 preorder_class.*.
  4044 
  4045 * Theory Relation: renamed "refl" to "refl_on", "reflexive" to "refl,
  4046 "diag" to "Id_on".
  4047 
  4048 * Theory Finite_Set: added a new fold combinator of type
  4049 
  4050   ('a => 'b => 'b) => 'b => 'a set => 'b
  4051 
  4052 Occasionally this is more convenient than the old fold combinator
  4053 which is now defined in terms of the new one and renamed to
  4054 fold_image.
  4055 
  4056 * Theories Ring_and_Field and OrderedGroup: The lemmas "group_simps"
  4057 and "ring_simps" have been replaced by "algebra_simps" (which can be
  4058 extended with further lemmas!).  At the moment both still exist but
  4059 the former will disappear at some point.
  4060 
  4061 * Theory Power: Lemma power_Suc is now declared as a simp rule in
  4062 class recpower.  Type-specific simp rules for various recpower types
  4063 have been removed.  INCOMPATIBILITY, rename old lemmas as follows:
  4064 
  4065 rat_power_0    -> power_0
  4066 rat_power_Suc  -> power_Suc
  4067 realpow_0      -> power_0
  4068 realpow_Suc    -> power_Suc
  4069 complexpow_0   -> power_0
  4070 complexpow_Suc -> power_Suc
  4071 power_poly_0   -> power_0
  4072 power_poly_Suc -> power_Suc
  4073 
  4074 * Theories Ring_and_Field and Divides: Definition of "op dvd" has been
  4075 moved to separate class dvd in Ring_and_Field; a couple of lemmas on
  4076 dvd has been generalized to class comm_semiring_1.  Likewise a bunch
  4077 of lemmas from Divides has been generalized from nat to class
  4078 semiring_div.  INCOMPATIBILITY.  This involves the following theorem
  4079 renames resulting from duplicate elimination:
  4080 
  4081     dvd_def_mod ~>          dvd_eq_mod_eq_0
  4082     zero_dvd_iff ~>         dvd_0_left_iff
  4083     dvd_0 ~>                dvd_0_right
  4084     DIVISION_BY_ZERO_DIV ~> div_by_0
  4085     DIVISION_BY_ZERO_MOD ~> mod_by_0
  4086     mult_div ~>             div_mult_self2_is_id
  4087     mult_mod ~>             mod_mult_self2_is_0
  4088 
  4089 * Theory IntDiv: removed many lemmas that are instances of class-based
  4090 generalizations (from Divides and Ring_and_Field).  INCOMPATIBILITY,
  4091 rename old lemmas as follows:
  4092 
  4093 dvd_diff               -> nat_dvd_diff
  4094 dvd_zminus_iff         -> dvd_minus_iff
  4095 mod_add1_eq            -> mod_add_eq
  4096 mod_mult1_eq           -> mod_mult_right_eq
  4097 mod_mult1_eq'          -> mod_mult_left_eq
  4098 mod_mult_distrib_mod   -> mod_mult_eq
  4099 nat_mod_add_left_eq    -> mod_add_left_eq
  4100 nat_mod_add_right_eq   -> mod_add_right_eq
  4101 nat_mod_div_trivial    -> mod_div_trivial
  4102 nat_mod_mod_trivial    -> mod_mod_trivial
  4103 zdiv_zadd_self1        -> div_add_self1
  4104 zdiv_zadd_self2        -> div_add_self2
  4105 zdiv_zmult_self1       -> div_mult_self2_is_id
  4106 zdiv_zmult_self2       -> div_mult_self1_is_id
  4107 zdvd_triv_left         -> dvd_triv_left
  4108 zdvd_triv_right        -> dvd_triv_right
  4109 zdvd_zmult_cancel_disj -> dvd_mult_cancel_left
  4110 zmod_eq0_zdvd_iff      -> dvd_eq_mod_eq_0[symmetric]
  4111 zmod_zadd_left_eq      -> mod_add_left_eq
  4112 zmod_zadd_right_eq     -> mod_add_right_eq
  4113 zmod_zadd_self1        -> mod_add_self1
  4114 zmod_zadd_self2        -> mod_add_self2
  4115 zmod_zadd1_eq          -> mod_add_eq
  4116 zmod_zdiff1_eq         -> mod_diff_eq
  4117 zmod_zdvd_zmod         -> mod_mod_cancel
  4118 zmod_zmod_cancel       -> mod_mod_cancel
  4119 zmod_zmult_self1       -> mod_mult_self2_is_0
  4120 zmod_zmult_self2       -> mod_mult_self1_is_0
  4121 zmod_1                 -> mod_by_1
  4122 zdiv_1                 -> div_by_1
  4123 zdvd_abs1              -> abs_dvd_iff
  4124 zdvd_abs2              -> dvd_abs_iff
  4125 zdvd_refl              -> dvd_refl
  4126 zdvd_trans             -> dvd_trans
  4127 zdvd_zadd              -> dvd_add
  4128 zdvd_zdiff             -> dvd_diff
  4129 zdvd_zminus_iff        -> dvd_minus_iff
  4130 zdvd_zminus2_iff       -> minus_dvd_iff
  4131 zdvd_zmultD            -> dvd_mult_right
  4132 zdvd_zmultD2           -> dvd_mult_left
  4133 zdvd_zmult_mono        -> mult_dvd_mono
  4134 zdvd_0_right           -> dvd_0_right
  4135 zdvd_0_left            -> dvd_0_left_iff
  4136 zdvd_1_left            -> one_dvd
  4137 zminus_dvd_iff         -> minus_dvd_iff
  4138 
  4139 * Theory Rational: 'Fract k 0' now equals '0'.  INCOMPATIBILITY.
  4140 
  4141 * The real numbers offer decimal input syntax: 12.34 is translated
  4142 into 1234/10^2. This translation is not reversed upon output.
  4143 
  4144 * Theory Library/Polynomial defines an abstract type 'a poly of
  4145 univariate polynomials with coefficients of type 'a.  In addition to
  4146 the standard ring operations, it also supports div and mod.  Code
  4147 generation is also supported, using list-style constructors.
  4148 
  4149 * Theory Library/Inner_Product defines a class of real_inner for real
  4150 inner product spaces, with an overloaded operation inner :: 'a => 'a
  4151 => real.  Class real_inner is a subclass of real_normed_vector from
  4152 theory RealVector.
  4153 
  4154 * Theory Library/Product_Vector provides instances for the product
  4155 type 'a * 'b of several classes from RealVector and Inner_Product.
  4156 Definitions of addition, subtraction, scalar multiplication, norms,
  4157 and inner products are included.
  4158 
  4159 * Theory Library/Bit defines the field "bit" of integers modulo 2.  In
  4160 addition to the field operations, numerals and case syntax are also
  4161 supported.
  4162 
  4163 * Theory Library/Diagonalize provides constructive version of Cantor's
  4164 first diagonalization argument.
  4165 
  4166 * Theory Library/GCD: Curried operations gcd, lcm (for nat) and zgcd,
  4167 zlcm (for int); carried together from various gcd/lcm developements in
  4168 the HOL Distribution.  Constants zgcd and zlcm replace former igcd and
  4169 ilcm; corresponding theorems renamed accordingly.  INCOMPATIBILITY,
  4170 may recover tupled syntax as follows:
  4171 
  4172     hide (open) const gcd
  4173     abbreviation gcd where
  4174       "gcd == (%(a, b). GCD.gcd a b)"
  4175     notation (output)
  4176       GCD.gcd ("gcd '(_, _')")
  4177 
  4178 The same works for lcm, zgcd, zlcm.
  4179 
  4180 * Theory Library/Nat_Infinity: added addition, numeral syntax and more
  4181 instantiations for algebraic structures.  Removed some duplicate
  4182 theorems.  Changes in simp rules.  INCOMPATIBILITY.
  4183 
  4184 * ML antiquotation @{code} takes a constant as argument and generates
  4185 corresponding code in background and inserts name of the corresponding
  4186 resulting ML value/function/datatype constructor binding in place.
  4187 All occurrences of @{code} with a single ML block are generated
  4188 simultaneously.  Provides a generic and safe interface for
  4189 instrumentalizing code generation.  See
  4190 src/HOL/Decision_Procs/Ferrack.thy for a more ambitious application.
  4191 In future you ought to refrain from ad-hoc compiling generated SML
  4192 code on the ML toplevel.  Note that (for technical reasons) @{code}
  4193 cannot refer to constants for which user-defined serializations are
  4194 set.  Refer to the corresponding ML counterpart directly in that
  4195 cases.
  4196 
  4197 * Command 'rep_datatype': instead of theorem names the command now
  4198 takes a list of terms denoting the constructors of the type to be
  4199 represented as datatype.  The characteristic theorems have to be
  4200 proven.  INCOMPATIBILITY.  Also observe that the following theorems
  4201 have disappeared in favour of existing ones:
  4202 
  4203     unit_induct                 ~> unit.induct
  4204     prod_induct                 ~> prod.induct
  4205     sum_induct                  ~> sum.induct
  4206     Suc_Suc_eq                  ~> nat.inject
  4207     Suc_not_Zero Zero_not_Suc   ~> nat.distinct
  4208 
  4209 
  4210 *** HOL-Algebra ***
  4211 
  4212 * New locales for orders and lattices where the equivalence relation
  4213 is not restricted to equality.  INCOMPATIBILITY: all order and lattice
  4214 locales use a record structure with field eq for the equivalence.
  4215 
  4216 * New theory of factorial domains.
  4217 
  4218 * Units_l_inv and Units_r_inv are now simp rules by default.
  4219 INCOMPATIBILITY.  Simplifier proof that require deletion of l_inv
  4220 and/or r_inv will now also require deletion of these lemmas.
  4221 
  4222 * Renamed the following theorems, INCOMPATIBILITY:
  4223 
  4224 UpperD ~> Upper_memD
  4225 LowerD ~> Lower_memD
  4226 least_carrier ~> least_closed
  4227 greatest_carrier ~> greatest_closed
  4228 greatest_Lower_above ~> greatest_Lower_below
  4229 one_zero ~> carrier_one_zero
  4230 one_not_zero ~> carrier_one_not_zero  (collision with assumption)
  4231 
  4232 
  4233 *** HOL-Nominal ***
  4234 
  4235 * Nominal datatypes can now contain type-variables.
  4236 
  4237 * Commands 'nominal_inductive' and 'equivariance' work with local
  4238 theory targets.
  4239 
  4240 * Nominal primrec can now works with local theory targets and its
  4241 specification syntax now conforms to the general format as seen in
  4242 'inductive' etc.
  4243 
  4244 * Method "perm_simp" honours the standard simplifier attributes
  4245 (no_asm), (no_asm_use) etc.
  4246 
  4247 * The new predicate #* is defined like freshness, except that on the
  4248 left hand side can be a set or list of atoms.
  4249 
  4250 * Experimental command 'nominal_inductive2' derives strong induction
  4251 principles for inductive definitions.  In contrast to
  4252 'nominal_inductive', which can only deal with a fixed number of
  4253 binders, it can deal with arbitrary expressions standing for sets of
  4254 atoms to be avoided.  The only inductive definition we have at the
  4255 moment that needs this generalisation is the typing rule for Lets in
  4256 the algorithm W:
  4257 
  4258  Gamma |- t1 : T1   (x,close Gamma T1)::Gamma |- t2 : T2   x#Gamma
  4259  -----------------------------------------------------------------
  4260          Gamma |- Let x be t1 in t2 : T2
  4261 
  4262 In this rule one wants to avoid all the binders that are introduced by
  4263 "close Gamma T1".  We are looking for other examples where this
  4264 feature might be useful.  Please let us know.
  4265 
  4266 
  4267 *** HOLCF ***
  4268 
  4269 * Reimplemented the simplification procedure for proving continuity
  4270 subgoals.  The new simproc is extensible; users can declare additional
  4271 continuity introduction rules with the attribute [cont2cont].
  4272 
  4273 * The continuity simproc now uses a different introduction rule for
  4274 solving continuity subgoals on terms with lambda abstractions.  In
  4275 some rare cases the new simproc may fail to solve subgoals that the
  4276 old one could solve, and "simp add: cont2cont_LAM" may be necessary.
  4277 Potential INCOMPATIBILITY.
  4278 
  4279 * Command 'fixrec': specification syntax now conforms to the general
  4280 format as seen in 'inductive' etc.  See src/HOLCF/ex/Fixrec_ex.thy for
  4281 examples.  INCOMPATIBILITY.
  4282 
  4283 
  4284 *** ZF ***
  4285 
  4286 * Proof of Zorn's Lemma for partial orders.
  4287 
  4288 
  4289 *** ML ***
  4290 
  4291 * Multithreading for Poly/ML 5.1/5.2 is no longer supported, only for
  4292 Poly/ML 5.2.1 or later.  Important note: the TimeLimit facility
  4293 depends on multithreading, so timouts will not work before Poly/ML
  4294 5.2.1!
  4295 
  4296 * High-level support for concurrent ML programming, see
  4297 src/Pure/Cuncurrent.  The data-oriented model of "future values" is
  4298 particularly convenient to organize independent functional
  4299 computations.  The concept of "synchronized variables" provides a
  4300 higher-order interface for components with shared state, avoiding the
  4301 delicate details of mutexes and condition variables.  (Requires
  4302 Poly/ML 5.2.1 or later.)
  4303 
  4304 * ML bindings produced via Isar commands are stored within the Isar
  4305 context (theory or proof).  Consequently, commands like 'use' and 'ML'
  4306 become thread-safe and work with undo as expected (concerning
  4307 top-level bindings, not side-effects on global references).
  4308 INCOMPATIBILITY, need to provide proper Isar context when invoking the
  4309 compiler at runtime; really global bindings need to be given outside a
  4310 theory.  (Requires Poly/ML 5.2 or later.)
  4311 
  4312 * Command 'ML_prf' is analogous to 'ML' but works within a proof
  4313 context.  Top-level ML bindings are stored within the proof context in
  4314 a purely sequential fashion, disregarding the nested proof structure.
  4315 ML bindings introduced by 'ML_prf' are discarded at the end of the
  4316 proof.  (Requires Poly/ML 5.2 or later.)
  4317 
  4318 * Simplified ML attribute and method setup, cf. functions Attrib.setup
  4319 and Method.setup, as well as Isar commands 'attribute_setup' and
  4320 'method_setup'.  INCOMPATIBILITY for 'method_setup', need to simplify
  4321 existing code accordingly, or use plain 'setup' together with old
  4322 Method.add_method.
  4323 
  4324 * Simplified ML oracle interface Thm.add_oracle promotes 'a -> cterm
  4325 to 'a -> thm, while results are always tagged with an authentic oracle
  4326 name.  The Isar command 'oracle' is now polymorphic, no argument type
  4327 is specified.  INCOMPATIBILITY, need to simplify existing oracle code
  4328 accordingly.  Note that extra performance may be gained by producing
  4329 the cterm carefully, avoiding slow Thm.cterm_of.
  4330 
  4331 * Simplified interface for defining document antiquotations via
  4332 ThyOutput.antiquotation, ThyOutput.output, and optionally
  4333 ThyOutput.maybe_pretty_source.  INCOMPATIBILITY, need to simplify user
  4334 antiquotations accordingly, see src/Pure/Thy/thy_output.ML for common
  4335 examples.
  4336 
  4337 * More systematic treatment of long names, abstract name bindings, and
  4338 name space operations.  Basic operations on qualified names have been
  4339 move from structure NameSpace to Long_Name, e.g. Long_Name.base_name,
  4340 Long_Name.append.  Old type bstring has been mostly replaced by
  4341 abstract type binding (see structure Binding), which supports precise
  4342 qualification by packages and local theory targets, as well as proper
  4343 tracking of source positions.  INCOMPATIBILITY, need to wrap old
  4344 bstring values into Binding.name, or better pass through abstract
  4345 bindings everywhere.  See further src/Pure/General/long_name.ML,
  4346 src/Pure/General/binding.ML and src/Pure/General/name_space.ML
  4347 
  4348 * Result facts (from PureThy.note_thms, ProofContext.note_thms,
  4349 LocalTheory.note etc.) now refer to the *full* internal name, not the
  4350 bstring as before.  INCOMPATIBILITY, not detected by ML type-checking!
  4351 
  4352 * Disposed old type and term read functions (Sign.read_def_typ,
  4353 Sign.read_typ, Sign.read_def_terms, Sign.read_term,
  4354 Thm.read_def_cterms, Thm.read_cterm etc.).  INCOMPATIBILITY, should
  4355 use regular Syntax.read_typ, Syntax.read_term, Syntax.read_typ_global,
  4356 Syntax.read_term_global etc.; see also OldGoals.read_term as last
  4357 resort for legacy applications.
  4358 
  4359 * Disposed old declarations, tactics, tactic combinators that refer to
  4360 the simpset or claset of an implicit theory (such as Addsimps,
  4361 Simp_tac, SIMPSET).  INCOMPATIBILITY, should use @{simpset} etc. in
  4362 embedded ML text, or local_simpset_of with a proper context passed as
  4363 explicit runtime argument.
  4364 
  4365 * Rules and tactics that read instantiations (read_instantiate,
  4366 res_inst_tac, thin_tac, subgoal_tac etc.) now demand a proper proof
  4367 context, which is required for parsing and type-checking.  Moreover,
  4368 the variables are specified as plain indexnames, not string encodings
  4369 thereof.  INCOMPATIBILITY.
  4370 
  4371 * Generic Toplevel.add_hook interface allows to analyze the result of
  4372 transactions.  E.g. see src/Pure/ProofGeneral/proof_general_pgip.ML
  4373 for theorem dependency output of transactions resulting in a new
  4374 theory state.
  4375 
  4376 * ML antiquotations: block-structured compilation context indicated by
  4377 \<lbrace> ... \<rbrace>; additional antiquotation forms:
  4378 
  4379   @{binding name}                         - basic name binding
  4380   @{let ?pat = term}                      - term abbreviation (HO matching)
  4381   @{note name = fact}                     - fact abbreviation
  4382   @{thm fact}                             - singleton fact (with attributes)
  4383   @{thms fact}                            - general fact (with attributes)
  4384   @{lemma prop by method}                 - singleton goal
  4385   @{lemma prop by meth1 meth2}            - singleton goal
  4386   @{lemma prop1 ... propN by method}      - general goal
  4387   @{lemma prop1 ... propN by meth1 meth2} - general goal
  4388   @{lemma (open) ...}                     - open derivation
  4389 
  4390 
  4391 *** System ***
  4392 
  4393 * The Isabelle "emacs" tool provides a specific interface to invoke
  4394 Proof General / Emacs, with more explicit failure if that is not
  4395 installed (the old isabelle-interface script silently falls back on
  4396 isabelle-process).  The PROOFGENERAL_HOME setting determines the
  4397 installation location of the Proof General distribution.
  4398 
  4399 * Isabelle/lib/classes/Pure.jar provides basic support to integrate
  4400 the Isabelle process into a JVM/Scala application.  See
  4401 Isabelle/lib/jedit/plugin for a minimal example.  (The obsolete Java
  4402 process wrapper has been discontinued.)
  4403 
  4404 * Added homegrown Isabelle font with unicode layout, see lib/fonts.
  4405 
  4406 * Various status messages (with exact source position information) are
  4407 emitted, if proper markup print mode is enabled.  This allows
  4408 user-interface components to provide detailed feedback on internal
  4409 prover operations.
  4410 
  4411 
  4412 
  4413 New in Isabelle2008 (June 2008)
  4414 -------------------------------
  4415 
  4416 *** General ***
  4417 
  4418 * The Isabelle/Isar Reference Manual (isar-ref) has been reorganized
  4419 and updated, with formally checked references as hyperlinks.
  4420 
  4421 * Theory loader: use_thy (and similar operations) no longer set the
  4422 implicit ML context, which was occasionally hard to predict and in
  4423 conflict with concurrency.  INCOMPATIBILITY, use ML within Isar which
  4424 provides a proper context already.
  4425 
  4426 * Theory loader: old-style ML proof scripts being *attached* to a thy
  4427 file are no longer supported.  INCOMPATIBILITY, regular 'uses' and
  4428 'use' within a theory file will do the job.
  4429 
  4430 * Name space merge now observes canonical order, i.e. the second space
  4431 is inserted into the first one, while existing entries in the first
  4432 space take precedence.  INCOMPATIBILITY in rare situations, may try to
  4433 swap theory imports.
  4434 
  4435 * Syntax: symbol \<chi> is now considered a letter.  Potential
  4436 INCOMPATIBILITY in identifier syntax etc.
  4437 
  4438 * Outer syntax: string tokens no longer admit escaped white space,
  4439 which was an accidental (undocumented) feature.  INCOMPATIBILITY, use
  4440 white space without escapes.
  4441 
  4442 * Outer syntax: string tokens may contain arbitrary character codes
  4443 specified via 3 decimal digits (as in SML).  E.g. "foo\095bar" for
  4444 "foo_bar".
  4445 
  4446 
  4447 *** Pure ***
  4448 
  4449 * Context-dependent token translations.  Default setup reverts locally
  4450 fixed variables, and adds hilite markup for undeclared frees.
  4451 
  4452 * Unused theorems can be found using the new command 'unused_thms'.
  4453 There are three ways of invoking it:
  4454 
  4455 (1) unused_thms
  4456      Only finds unused theorems in the current theory.
  4457 
  4458 (2) unused_thms thy_1 ... thy_n -
  4459      Finds unused theorems in the current theory and all of its ancestors,
  4460      excluding the theories thy_1 ... thy_n and all of their ancestors.
  4461 
  4462 (3) unused_thms thy_1 ... thy_n - thy'_1 ... thy'_m
  4463      Finds unused theorems in the theories thy'_1 ... thy'_m and all of
  4464      their ancestors, excluding the theories thy_1 ... thy_n and all of
  4465      their ancestors.
  4466 
  4467 In order to increase the readability of the list produced by
  4468 unused_thms, theorems that have been created by a particular instance
  4469 of a theory command such as 'inductive' or 'function' are considered
  4470 to belong to the same "group", meaning that if at least one theorem in
  4471 this group is used, the other theorems in the same group are no longer
  4472 reported as unused.  Moreover, if all theorems in the group are
  4473 unused, only one theorem in the group is displayed.
  4474 
  4475 Note that proof objects have to be switched on in order for
  4476 unused_thms to work properly (i.e. !proofs must be >= 1, which is
  4477 usually the case when using Proof General with the default settings).
  4478 
  4479 * Authentic naming of facts disallows ad-hoc overwriting of previous
  4480 theorems within the same name space.  INCOMPATIBILITY, need to remove
  4481 duplicate fact bindings, or even accidental fact duplications.  Note
  4482 that tools may maintain dynamically scoped facts systematically, using
  4483 PureThy.add_thms_dynamic.
  4484 
  4485 * Command 'hide' now allows to hide from "fact" name space as well.
  4486 
  4487 * Eliminated destructive theorem database, simpset, claset, and
  4488 clasimpset.  Potential INCOMPATIBILITY, really need to observe linear
  4489 update of theories within ML code.
  4490 
  4491 * Eliminated theory ProtoPure and CPure, leaving just one Pure theory.
  4492 INCOMPATIBILITY, object-logics depending on former Pure require
  4493 additional setup PureThy.old_appl_syntax_setup; object-logics
  4494 depending on former CPure need to refer to Pure.
  4495 
  4496 * Commands 'use' and 'ML' are now purely functional, operating on
  4497 theory/local_theory.  Removed former 'ML_setup' (on theory), use 'ML'
  4498 instead.  Added 'ML_val' as mere diagnostic replacement for 'ML'.
  4499 INCOMPATIBILITY.
  4500 
  4501 * Command 'setup': discontinued implicit version with ML reference.
  4502 
  4503 * Instantiation target allows for simultaneous specification of class
  4504 instance operations together with an instantiation proof.
  4505 Type-checking phase allows to refer to class operations uniformly.
  4506 See src/HOL/Complex/Complex.thy for an Isar example and
  4507 src/HOL/Library/Eval.thy for an ML example.
  4508 
  4509 * Indexing of literal facts: be more serious about including only
  4510 facts from the visible specification/proof context, but not the
  4511 background context (locale etc.).  Affects `prop` notation and method
  4512 "fact".  INCOMPATIBILITY: need to name facts explicitly in rare
  4513 situations.
  4514 
  4515 * Method "cases", "induct", "coinduct": removed obsolete/undocumented
  4516 "(open)" option, which used to expose internal bound variables to the
  4517 proof text.
  4518 
  4519 * Isar statements: removed obsolete case "rule_context".
  4520 INCOMPATIBILITY, better use explicit fixes/assumes.
  4521 
  4522 * Locale proofs: default proof step now includes 'unfold_locales';
  4523 hence 'proof' without argument may be used to unfold locale
  4524 predicates.
  4525 
  4526 
  4527 *** Document preparation ***
  4528 
  4529 * Simplified pdfsetup.sty: color/hyperref is used unconditionally for
  4530 both pdf and dvi (hyperlinks usually work in xdvi as well); removed
  4531 obsolete thumbpdf setup (contemporary PDF viewers do this on the
  4532 spot); renamed link color from "darkblue" to "linkcolor" (default
  4533 value unchanged, can be redefined via \definecolor); no longer sets
  4534 "a4paper" option (unnecessary or even intrusive).
  4535 
  4536 * Antiquotation @{lemma A method} proves proposition A by the given
  4537 method (either a method name or a method name plus (optional) method
  4538 arguments in parentheses) and prints A just like @{prop A}.
  4539 
  4540 
  4541 *** HOL ***
  4542 
  4543 * New primrec package.  Specification syntax conforms in style to
  4544 definition/function/....  No separate induction rule is provided.  The
  4545 "primrec" command distinguishes old-style and new-style specifications
  4546 by syntax.  The former primrec package is now named OldPrimrecPackage.
  4547 When adjusting theories, beware: constants stemming from new-style
  4548 primrec specifications have authentic syntax.
  4549 
  4550 * Metis prover is now an order of magnitude faster, and also works
  4551 with multithreading.
  4552 
  4553 * Metis: the maximum number of clauses that can be produced from a
  4554 theorem is now given by the attribute max_clauses.  Theorems that
  4555 exceed this number are ignored, with a warning printed.
  4556 
  4557 * Sledgehammer no longer produces structured proofs by default. To
  4558 enable, declare [[sledgehammer_full = true]].  Attributes
  4559 reconstruction_modulus, reconstruction_sorts renamed
  4560 sledgehammer_modulus, sledgehammer_sorts.  INCOMPATIBILITY.
  4561 
  4562 * Method "induct_scheme" derives user-specified induction rules
  4563 from well-founded induction and completeness of patterns. This factors
  4564 out some operations that are done internally by the function package
  4565 and makes them available separately.  See
  4566 src/HOL/ex/Induction_Scheme.thy for examples.
  4567 
  4568 * More flexible generation of measure functions for termination
  4569 proofs: Measure functions can be declared by proving a rule of the
  4570 form "is_measure f" and giving it the [measure_function] attribute.
  4571 The "is_measure" predicate is logically meaningless (always true), and
  4572 just guides the heuristic.  To find suitable measure functions, the
  4573 termination prover sets up the goal "is_measure ?f" of the appropriate
  4574 type and generates all solutions by prolog-style backwards proof using
  4575 the declared rules.
  4576 
  4577 This setup also deals with rules like
  4578 
  4579   "is_measure f ==> is_measure (list_size f)"
  4580 
  4581 which accommodates nested datatypes that recurse through lists.
  4582 Similar rules are predeclared for products and option types.
  4583 
  4584 * Turned the type of sets "'a set" into an abbreviation for "'a => bool"
  4585 
  4586   INCOMPATIBILITIES:
  4587 
  4588   - Definitions of overloaded constants on sets have to be replaced by
  4589     definitions on => and bool.
  4590 
  4591   - Some definitions of overloaded operators on sets can now be proved
  4592     using the definitions of the operators on => and bool.  Therefore,
  4593     the following theorems have been renamed:
  4594 
  4595       subset_def   -> subset_eq
  4596       psubset_def  -> psubset_eq
  4597       set_diff_def -> set_diff_eq
  4598       Compl_def    -> Compl_eq
  4599       Sup_set_def  -> Sup_set_eq
  4600       Inf_set_def  -> Inf_set_eq
  4601       sup_set_def  -> sup_set_eq
  4602       inf_set_def  -> inf_set_eq
  4603 
  4604   - Due to the incompleteness of the HO unification algorithm, some
  4605     rules such as subst may require manual instantiation, if some of
  4606     the unknowns in the rule is a set.
  4607 
  4608   - Higher order unification and forward proofs:
  4609     The proof pattern
  4610 
  4611       have "P (S::'a set)" <...>
  4612       then have "EX S. P S" ..
  4613 
  4614     no longer works (due to the incompleteness of the HO unification
  4615     algorithm) and must be replaced by the pattern
  4616 
  4617       have "EX S. P S"
  4618       proof
  4619         show "P S" <...>
  4620       qed
  4621 
  4622   - Calculational reasoning with subst (or similar rules):
  4623     The proof pattern
  4624 
  4625       have "P (S::'a set)" <...>
  4626       also have "S = T" <...>
  4627       finally have "P T" .
  4628 
  4629     no longer works (for similar reasons as the previous example) and
  4630     must be replaced by something like
  4631 
  4632       have "P (S::'a set)" <...>
  4633       moreover have "S = T" <...>
  4634       ultimately have "P T" by simp
  4635 
  4636   - Tactics or packages written in ML code:
  4637     Code performing pattern matching on types via
  4638 
  4639       Type ("set", [T]) => ...
  4640 
  4641     must be rewritten. Moreover, functions like strip_type or
  4642     binder_types no longer return the right value when applied to a
  4643     type of the form
  4644 
  4645       T1 => ... => Tn => U => bool
  4646 
  4647     rather than
  4648 
  4649       T1 => ... => Tn => U set
  4650 
  4651 * Merged theories Wellfounded_Recursion, Accessible_Part and
  4652 Wellfounded_Relations to theory Wellfounded.
  4653 
  4654 * Explicit class "eq" for executable equality.  INCOMPATIBILITY.
  4655 
  4656 * Class finite no longer treats UNIV as class parameter.  Use class
  4657 enum from theory Library/Enum instead to achieve a similar effect.
  4658 INCOMPATIBILITY.
  4659 
  4660 * Theory List: rule list_induct2 now has explicitly named cases "Nil"
  4661 and "Cons".  INCOMPATIBILITY.
  4662 
  4663 * HOL (and FOL): renamed variables in rules imp_elim and swap.
  4664 Potential INCOMPATIBILITY.
  4665 
  4666 * Theory Product_Type: duplicated lemmas split_Pair_apply and
  4667 injective_fst_snd removed, use split_eta and prod_eqI instead.
  4668 Renamed upd_fst to apfst and upd_snd to apsnd.  INCOMPATIBILITY.
  4669 
  4670 * Theory Nat: removed redundant lemmas that merely duplicate lemmas of
  4671 the same name in theory Orderings:
  4672 
  4673   less_trans
  4674   less_linear
  4675   le_imp_less_or_eq
  4676   le_less_trans
  4677   less_le_trans
  4678   less_not_sym
  4679   less_asym
  4680 
  4681 Renamed less_imp_le to less_imp_le_nat, and less_irrefl to
  4682 less_irrefl_nat.  Potential INCOMPATIBILITY due to more general types
  4683 and different variable names.
  4684 
  4685 * Library/Option_ord.thy: Canonical order on option type.
  4686 
  4687 * Library/RBT.thy: Red-black trees, an efficient implementation of
  4688 finite maps.
  4689 
  4690 * Library/Countable.thy: Type class for countable types.
  4691 
  4692 * Theory Int: The representation of numerals has changed.  The infix
  4693 operator BIT and the bit datatype with constructors B0 and B1 have
  4694 disappeared.  INCOMPATIBILITY, use "Int.Bit0 x" and "Int.Bit1 y" in
  4695 place of "x BIT bit.B0" and "y BIT bit.B1", respectively.  Theorems
  4696 involving BIT, B0, or B1 have been renamed with "Bit0" or "Bit1"
  4697 accordingly.
  4698 
  4699 * Theory Nat: definition of <= and < on natural numbers no longer
  4700 depend on well-founded relations.  INCOMPATIBILITY.  Definitions
  4701 le_def and less_def have disappeared.  Consider lemmas not_less
  4702 [symmetric, where ?'a = nat] and less_eq [symmetric] instead.
  4703 
  4704 * Theory Finite_Set: locales ACf, ACe, ACIf, ACIfSL and ACIfSLlin
  4705 (whose purpose mainly is for various fold_set functionals) have been
  4706 abandoned in favor of the existing algebraic classes
  4707 ab_semigroup_mult, comm_monoid_mult, ab_semigroup_idem_mult,
  4708 lower_semilattice (resp. upper_semilattice) and linorder.
  4709 INCOMPATIBILITY.
  4710 
  4711 * Theory Transitive_Closure: induct and cases rules now declare proper
  4712 case_names ("base" and "step").  INCOMPATIBILITY.
  4713 
  4714 * Theorem Inductive.lfp_ordinal_induct generalized to complete
  4715 lattices.  The form set-specific version is available as
  4716 Inductive.lfp_ordinal_induct_set.
  4717 
  4718 * Renamed theorems "power.simps" to "power_int.simps".
  4719 INCOMPATIBILITY.
  4720 
  4721 * Class semiring_div provides basic abstract properties of semirings
  4722 with division and modulo operations.  Subsumes former class dvd_mod.
  4723 
  4724 * Merged theories IntDef, Numeral and IntArith into unified theory
  4725 Int.  INCOMPATIBILITY.
  4726 
  4727 * Theory Library/Code_Index: type "index" now represents natural
  4728 numbers rather than integers.  INCOMPATIBILITY.
  4729 
  4730 * New class "uminus" with operation "uminus" (split of from class
  4731 "minus" which now only has operation "minus", binary).
  4732 INCOMPATIBILITY.
  4733 
  4734 * Constants "card", "internal_split", "option_map" now with authentic
  4735 syntax.  INCOMPATIBILITY.
  4736 
  4737 * Definitions subset_def, psubset_def, set_diff_def, Compl_def,
  4738 le_bool_def, less_bool_def, le_fun_def, less_fun_def, inf_bool_def,
  4739 sup_bool_def, Inf_bool_def, Sup_bool_def, inf_fun_def, sup_fun_def,
  4740 Inf_fun_def, Sup_fun_def, inf_set_def, sup_set_def, Inf_set_def,
  4741 Sup_set_def, le_def, less_def, option_map_def now with object
  4742 equality.  INCOMPATIBILITY.
  4743 
  4744 * Records. Removed K_record, and replaced it by pure lambda term
  4745 %x. c. The simplifier setup is now more robust against eta expansion.
  4746 INCOMPATIBILITY: in cases explicitly referring to K_record.
  4747 
  4748 * Library/Multiset: {#a, b, c#} abbreviates {#a#} + {#b#} + {#c#}.
  4749 
  4750 * Library/ListVector: new theory of arithmetic vector operations.
  4751 
  4752 * Library/Order_Relation: new theory of various orderings as sets of
  4753 pairs.  Defines preorders, partial orders, linear orders and
  4754 well-orders on sets and on types.
  4755 
  4756 
  4757 *** ZF ***
  4758 
  4759 * Renamed some theories to allow to loading both ZF and HOL in the
  4760 same session:
  4761 
  4762   Datatype  -> Datatype_ZF
  4763   Inductive -> Inductive_ZF
  4764   Int       -> Int_ZF
  4765   IntDiv    -> IntDiv_ZF
  4766   Nat       -> Nat_ZF
  4767   List      -> List_ZF
  4768   Main      -> Main_ZF
  4769 
  4770 INCOMPATIBILITY: ZF theories that import individual theories below
  4771 Main might need to be adapted.  Regular theory Main is still
  4772 available, as trivial extension of Main_ZF.
  4773 
  4774 
  4775 *** ML ***
  4776 
  4777 * ML within Isar: antiquotation @{const name} or @{const
  4778 name(typargs)} produces statically-checked Const term.
  4779 
  4780 * Functor NamedThmsFun: data is available to the user as dynamic fact
  4781 (of the same name).  Removed obsolete print command.
  4782 
  4783 * Removed obsolete "use_legacy_bindings" function.
  4784 
  4785 * The ``print mode'' is now a thread-local value derived from a global
  4786 template (the former print_mode reference), thus access becomes
  4787 non-critical.  The global print_mode reference is for session
  4788 management only; user-code should use print_mode_value,
  4789 print_mode_active, PrintMode.setmp etc.  INCOMPATIBILITY.
  4790 
  4791 * Functions system/system_out provide a robust way to invoke external
  4792 shell commands, with propagation of interrupts (requires Poly/ML
  4793 5.2.1).  Do not use OS.Process.system etc. from the basis library!
  4794 
  4795 
  4796 *** System ***
  4797 
  4798 * Default settings: PROOFGENERAL_OPTIONS no longer impose xemacs ---
  4799 in accordance with Proof General 3.7, which prefers GNU emacs.
  4800 
  4801 * isatool tty runs Isabelle process with plain tty interaction;
  4802 optional line editor may be specified via ISABELLE_LINE_EDITOR
  4803 setting, the default settings attempt to locate "ledit" and "rlwrap".
  4804 
  4805 * isatool browser now works with Cygwin as well, using general
  4806 "javapath" function defined in Isabelle process environment.
  4807 
  4808 * YXML notation provides a simple and efficient alternative to
  4809 standard XML transfer syntax.  See src/Pure/General/yxml.ML and
  4810 isatool yxml as described in the Isabelle system manual.
  4811 
  4812 * JVM class isabelle.IsabelleProcess (located in Isabelle/lib/classes)
  4813 provides general wrapper for managing an Isabelle process in a robust
  4814 fashion, with ``cooked'' output from stdin/stderr.
  4815 
  4816 * Rudimentary Isabelle plugin for jEdit (see Isabelle/lib/jedit),
  4817 based on Isabelle/JVM process wrapper (see Isabelle/lib/classes).
  4818 
  4819 * Removed obsolete THIS_IS_ISABELLE_BUILD feature.  NB: the documented
  4820 way of changing the user's settings is via
  4821 ISABELLE_HOME_USER/etc/settings, which is a fully featured bash
  4822 script.
  4823 
  4824 * Multithreading.max_threads := 0 refers to the number of actual CPU
  4825 cores of the underlying machine, which is a good starting point for
  4826 optimal performance tuning.  The corresponding usedir option -M allows
  4827 "max" as an alias for "0".  WARNING: does not work on certain versions
  4828 of Mac OS (with Poly/ML 5.1).
  4829 
  4830 * isabelle-process: non-ML sessions are run with "nice", to reduce the
  4831 adverse effect of Isabelle flooding interactive front-ends (notably
  4832 ProofGeneral / XEmacs).
  4833 
  4834 
  4835 
  4836 New in Isabelle2007 (November 2007)
  4837 -----------------------------------
  4838 
  4839 *** General ***
  4840 
  4841 * More uniform information about legacy features, notably a
  4842 warning/error of "Legacy feature: ...", depending on the state of the
  4843 tolerate_legacy_features flag (default true). FUTURE INCOMPATIBILITY:
  4844 legacy features will disappear eventually.
  4845 
  4846 * Theory syntax: the header format ``theory A = B + C:'' has been
  4847 discontinued in favour of ``theory A imports B C begin''.  Use isatool
  4848 fixheaders to convert existing theory files.  INCOMPATIBILITY.
  4849 
  4850 * Theory syntax: the old non-Isar theory file format has been
  4851 discontinued altogether.  Note that ML proof scripts may still be used
  4852 with Isar theories; migration is usually quite simple with the ML
  4853 function use_legacy_bindings.  INCOMPATIBILITY.
  4854 
  4855 * Theory syntax: some popular names (e.g. 'class', 'declaration',
  4856 'fun', 'help', 'if') are now keywords.  INCOMPATIBILITY, use double
  4857 quotes.
  4858 
  4859 * Theory loader: be more serious about observing the static theory
  4860 header specifications (including optional directories), but not the
  4861 accidental file locations of previously successful loads.  The strict
  4862 update policy of former update_thy is now already performed by
  4863 use_thy, so the former has been removed; use_thys updates several
  4864 theories simultaneously, just as 'imports' within a theory header
  4865 specification, but without merging the results.  Potential
  4866 INCOMPATIBILITY: may need to refine theory headers and commands
  4867 ROOT.ML which depend on load order.
  4868 
  4869 * Theory loader: optional support for content-based file
  4870 identification, instead of the traditional scheme of full physical
  4871 path plus date stamp; configured by the ISABELLE_FILE_IDENT setting
  4872 (cf. the system manual).  The new scheme allows to work with
  4873 non-finished theories in persistent session images, such that source
  4874 files may be moved later on without requiring reloads.
  4875 
  4876 * Theory loader: old-style ML proof scripts being *attached* to a thy
  4877 file (with the same base name as the theory) are considered a legacy
  4878 feature, which will disappear eventually. Even now, the theory loader
  4879 no longer maintains dependencies on such files.
  4880 
  4881 * Syntax: the scope for resolving ambiguities via type-inference is
  4882 now limited to individual terms, instead of whole simultaneous
  4883 specifications as before. This greatly reduces the complexity of the
  4884 syntax module and improves flexibility by separating parsing and
  4885 type-checking. INCOMPATIBILITY: additional type-constraints (explicit
  4886 'fixes' etc.) are required in rare situations.
  4887 
  4888 * Syntax: constants introduced by new-style packages ('definition',
  4889 'abbreviation' etc.) are passed through the syntax module in
  4890 ``authentic mode''. This means that associated mixfix annotations
  4891 really stick to such constants, independently of potential name space
  4892 ambiguities introduced later on. INCOMPATIBILITY: constants in parse
  4893 trees are represented slightly differently, may need to adapt syntax
  4894 translations accordingly. Use CONST marker in 'translations' and
  4895 @{const_syntax} antiquotation in 'parse_translation' etc.
  4896 
  4897 * Legacy goal package: reduced interface to the bare minimum required
  4898 to keep existing proof scripts running.  Most other user-level
  4899 functions are now part of the OldGoals structure, which is *not* open
  4900 by default (consider isatool expandshort before open OldGoals).
  4901 Removed top_sg, prin, printyp, pprint_term/typ altogether, because
  4902 these tend to cause confusion about the actual goal (!) context being
  4903 used here, which is not necessarily the same as the_context().
  4904 
  4905 * Command 'find_theorems': supports "*" wild-card in "name:"
  4906 criterion; "with_dups" option.  Certain ProofGeneral versions might
  4907 support a specific search form (see ProofGeneral/CHANGES).
  4908 
  4909 * The ``prems limit'' option (cf. ProofContext.prems_limit) is now -1
  4910 by default, which means that "prems" (and also "fixed variables") are
  4911 suppressed from proof state output.  Note that the ProofGeneral
  4912 settings mechanism allows to change and save options persistently, but
  4913 older versions of Isabelle will fail to start up if a negative prems
  4914 limit is imposed.
  4915 
  4916 * Local theory targets may be specified by non-nested blocks of
  4917 ``context/locale/class ... begin'' followed by ``end''.  The body may
  4918 contain definitions, theorems etc., including any derived mechanism
  4919 that has been implemented on top of these primitives.  This concept
  4920 generalizes the existing ``theorem (in ...)'' towards more versatility
  4921 and scalability.
  4922 
  4923 * Proof General interface: proper undo of final 'end' command;
  4924 discontinued Isabelle/classic mode (ML proof scripts).
  4925 
  4926 
  4927 *** Document preparation ***
  4928 
  4929 * Added antiquotation @{theory name} which prints the given name,
  4930 after checking that it refers to a valid ancestor theory in the
  4931 current context.
  4932 
  4933 * Added antiquotations @{ML_type text} and @{ML_struct text} which
  4934 check the given source text as ML type/structure, printing verbatim.
  4935 
  4936 * Added antiquotation @{abbrev "c args"} which prints the abbreviation
  4937 "c args == rhs" given in the current context.  (Any number of
  4938 arguments may be given on the LHS.)
  4939 
  4940 
  4941 *** Pure ***
  4942 
  4943 * The 'class' package offers a combination of axclass and locale to
  4944 achieve Haskell-like type classes in Isabelle.  Definitions and
  4945 theorems within a class context produce both relative results (with
  4946 implicit parameters according to the locale context), and polymorphic
  4947 constants with qualified polymorphism (according to the class
  4948 context).  Within the body context of a 'class' target, a separate
  4949 syntax layer ("user space type system") takes care of converting
  4950 between global polymorphic consts and internal locale representation.
  4951 See src/HOL/ex/Classpackage.thy for examples (as well as main HOL).
  4952 "isatool doc classes" provides a tutorial.
  4953 
  4954 * Generic code generator framework allows to generate executable
  4955 code for ML and Haskell (including Isabelle classes).  A short usage
  4956 sketch:
  4957 
  4958     internal compilation:
  4959         export_code <list of constants (term syntax)> in SML
  4960     writing SML code to a file:
  4961         export_code <list of constants (term syntax)> in SML <filename>
  4962     writing OCaml code to a file:
  4963         export_code <list of constants (term syntax)> in OCaml <filename>
  4964     writing Haskell code to a bunch of files:
  4965         export_code <list of constants (term syntax)> in Haskell <filename>
  4966 
  4967     evaluating closed propositions to True/False using code generation:
  4968         method ``eval''
  4969 
  4970 Reasonable default setup of framework in HOL.
  4971 
  4972 Theorem attributs for selecting and transforming function equations theorems:
  4973 
  4974     [code fun]:        select a theorem as function equation for a specific constant
  4975     [code fun del]:    deselect a theorem as function equation for a specific constant
  4976     [code inline]:     select an equation theorem for unfolding (inlining) in place
  4977     [code inline del]: deselect an equation theorem for unfolding (inlining) in place
  4978 
  4979 User-defined serializations (target in {SML, OCaml, Haskell}):
  4980 
  4981     code_const <and-list of constants (term syntax)>
  4982       {(target) <and-list of const target syntax>}+
  4983 
  4984     code_type <and-list of type constructors>
  4985       {(target) <and-list of type target syntax>}+
  4986 
  4987     code_instance <and-list of instances>
  4988       {(target)}+
  4989         where instance ::= <type constructor> :: <class>
  4990 
  4991     code_class <and_list of classes>
  4992       {(target) <and-list of class target syntax>}+
  4993         where class target syntax ::= <class name> {where {<classop> == <target syntax>}+}?
  4994 
  4995 code_instance and code_class only are effective to target Haskell.
  4996 
  4997 For example usage see src/HOL/ex/Codegenerator.thy and
  4998 src/HOL/ex/Codegenerator_Pretty.thy.  A separate tutorial on code
  4999 generation from Isabelle/HOL theories is available via "isatool doc
  5000 codegen".
  5001 
  5002 * Code generator: consts in 'consts_code' Isar commands are now
  5003 referred to by usual term syntax (including optional type
  5004 annotations).
  5005 
  5006 * Command 'no_translations' removes translation rules from theory
  5007 syntax.
  5008 
  5009 * Overloaded definitions are now actually checked for acyclic
  5010 dependencies.  The overloading scheme is slightly more general than
  5011 that of Haskell98, although Isabelle does not demand an exact
  5012 correspondence to type class and instance declarations.
  5013 INCOMPATIBILITY, use ``defs (unchecked overloaded)'' to admit more
  5014 exotic versions of overloading -- at the discretion of the user!
  5015 
  5016 Polymorphic constants are represented via type arguments, i.e. the
  5017 instantiation that matches an instance against the most general
  5018 declaration given in the signature.  For example, with the declaration
  5019 c :: 'a => 'a => 'a, an instance c :: nat => nat => nat is represented
  5020 as c(nat).  Overloading is essentially simultaneous structural
  5021 recursion over such type arguments.  Incomplete specification patterns
  5022 impose global constraints on all occurrences, e.g. c('a * 'a) on the
  5023 LHS means that more general c('a * 'b) will be disallowed on any RHS.
  5024 Command 'print_theory' outputs the normalized system of recursive
  5025 equations, see section "definitions".
  5026 
  5027 * Configuration options are maintained within the theory or proof
  5028 context (with name and type bool/int/string), providing a very simple
  5029 interface to a poor-man's version of general context data.  Tools may
  5030 declare options in ML (e.g. using Attrib.config_int) and then refer to
  5031 these values using Config.get etc.  Users may change options via an
  5032 associated attribute of the same name.  This form of context
  5033 declaration works particularly well with commands 'declare' or
  5034 'using', for example ``declare [[foo = 42]]''.  Thus it has become
  5035 very easy to avoid global references, which would not observe Isar
  5036 toplevel undo/redo and fail to work with multithreading.
  5037 
  5038 Various global ML references of Pure and HOL have been turned into
  5039 configuration options:
  5040 
  5041   Unify.search_bound		unify_search_bound
  5042   Unify.trace_bound		unify_trace_bound
  5043   Unify.trace_simp		unify_trace_simp
  5044   Unify.trace_types		unify_trace_types
  5045   Simplifier.simp_depth_limit	simp_depth_limit
  5046   Blast.depth_limit		blast_depth_limit
  5047   DatatypeProp.dtK		datatype_distinctness_limit
  5048   fast_arith_neq_limit  	fast_arith_neq_limit
  5049   fast_arith_split_limit	fast_arith_split_limit
  5050 
  5051 * Named collections of theorems may be easily installed as context
  5052 data using the functor NamedThmsFun (see also
  5053 src/Pure/Tools/named_thms.ML).  The user may add or delete facts via
  5054 attributes; there is also a toplevel print command.  This facility is
  5055 just a common case of general context data, which is the preferred way
  5056 for anything more complex than just a list of facts in canonical
  5057 order.
  5058 
  5059 * Isar: command 'declaration' augments a local theory by generic
  5060 declaration functions written in ML.  This enables arbitrary content
  5061 being added to the context, depending on a morphism that tells the
  5062 difference of the original declaration context wrt. the application
  5063 context encountered later on.
  5064 
  5065 * Isar: proper interfaces for simplification procedures.  Command
  5066 'simproc_setup' declares named simprocs (with match patterns, and body
  5067 text in ML).  Attribute "simproc" adds/deletes simprocs in the current
  5068 context.  ML antiquotation @{simproc name} retrieves named simprocs.
  5069 
  5070 * Isar: an extra pair of brackets around attribute declarations
  5071 abbreviates a theorem reference involving an internal dummy fact,
  5072 which will be ignored later --- only the effect of the attribute on
  5073 the background context will persist.  This form of in-place
  5074 declarations is particularly useful with commands like 'declare' and
  5075 'using', for example ``have A using [[simproc a]] by simp''.
  5076 
  5077 * Isar: method "assumption" (and implicit closing of subproofs) now
  5078 takes simple non-atomic goal assumptions into account: after applying
  5079 an assumption as a rule the resulting subgoals are solved by atomic
  5080 assumption steps.  This is particularly useful to finish 'obtain'
  5081 goals, such as "!!x. (!!x. P x ==> thesis) ==> P x ==> thesis",
  5082 without referring to the original premise "!!x. P x ==> thesis" in the
  5083 Isar proof context.  POTENTIAL INCOMPATIBILITY: method "assumption" is
  5084 more permissive.
  5085 
  5086 * Isar: implicit use of prems from the Isar proof context is
  5087 considered a legacy feature.  Common applications like ``have A .''
  5088 may be replaced by ``have A by fact'' or ``note `A`''.  In general,
  5089 referencing facts explicitly here improves readability and
  5090 maintainability of proof texts.
  5091 
  5092 * Isar: improper proof element 'guess' is like 'obtain', but derives
  5093 the obtained context from the course of reasoning!  For example:
  5094 
  5095   assume "EX x y. A x & B y"   -- "any previous fact"
  5096   then guess x and y by clarify
  5097 
  5098 This technique is potentially adventurous, depending on the facts and
  5099 proof tools being involved here.
  5100 
  5101 * Isar: known facts from the proof context may be specified as literal
  5102 propositions, using ASCII back-quote syntax.  This works wherever
  5103 named facts used to be allowed so far, in proof commands, proof
  5104 methods, attributes etc.  Literal facts are retrieved from the context
  5105 according to unification of type and term parameters.  For example,
  5106 provided that "A" and "A ==> B" and "!!x. P x ==> Q x" are known
  5107 theorems in the current context, then these are valid literal facts:
  5108 `A` and `A ==> B` and `!!x. P x ==> Q x" as well as `P a ==> Q a` etc.
  5109 
  5110 There is also a proof method "fact" which does the same composition
  5111 for explicit goal states, e.g. the following proof texts coincide with
  5112 certain special cases of literal facts:
  5113 
  5114   have "A" by fact                 ==  note `A`
  5115   have "A ==> B" by fact           ==  note `A ==> B`
  5116   have "!!x. P x ==> Q x" by fact  ==  note `!!x. P x ==> Q x`
  5117   have "P a ==> Q a" by fact       ==  note `P a ==> Q a`
  5118 
  5119 * Isar: ":" (colon) is no longer a symbolic identifier character in
  5120 outer syntax.  Thus symbolic identifiers may be used without
  5121 additional white space in declarations like this: ``assume *: A''.
  5122 
  5123 * Isar: 'print_facts' prints all local facts of the current context,
  5124 both named and unnamed ones.
  5125 
  5126 * Isar: 'def' now admits simultaneous definitions, e.g.:
  5127 
  5128   def x == "t" and y == "u"
  5129 
  5130 * Isar: added command 'unfolding', which is structurally similar to
  5131 'using', but affects both the goal state and facts by unfolding given
  5132 rewrite rules.  Thus many occurrences of the 'unfold' method or
  5133 'unfolded' attribute may be replaced by first-class proof text.
  5134 
  5135 * Isar: methods 'unfold' / 'fold', attributes 'unfolded' / 'folded',
  5136 and command 'unfolding' now all support object-level equalities
  5137 (potentially conditional).  The underlying notion of rewrite rule is
  5138 analogous to the 'rule_format' attribute, but *not* that of the
  5139 Simplifier (which is usually more generous).
  5140 
  5141 * Isar: the new attribute [rotated n] (default n = 1) rotates the
  5142 premises of a theorem by n. Useful in conjunction with drule.
  5143 
  5144 * Isar: the goal restriction operator [N] (default N = 1) evaluates a
  5145 method expression within a sandbox consisting of the first N
  5146 sub-goals, which need to exist.  For example, ``simp_all [3]''
  5147 simplifies the first three sub-goals, while (rule foo, simp_all)[]
  5148 simplifies all new goals that emerge from applying rule foo to the
  5149 originally first one.
  5150 
  5151 * Isar: schematic goals are no longer restricted to higher-order
  5152 patterns; e.g. ``lemma "?P(?x)" by (rule TrueI)'' now works as
  5153 expected.
  5154 
  5155 * Isar: the conclusion of a long theorem statement is now either
  5156 'shows' (a simultaneous conjunction, as before), or 'obtains'
  5157 (essentially a disjunction of cases with local parameters and
  5158 assumptions).  The latter allows to express general elimination rules
  5159 adequately; in this notation common elimination rules look like this:
  5160 
  5161   lemma exE:    -- "EX x. P x ==> (!!x. P x ==> thesis) ==> thesis"
  5162     assumes "EX x. P x"
  5163     obtains x where "P x"
  5164 
  5165   lemma conjE:  -- "A & B ==> (A ==> B ==> thesis) ==> thesis"
  5166     assumes "A & B"
  5167     obtains A and B
  5168 
  5169   lemma disjE:  -- "A | B ==> (A ==> thesis) ==> (B ==> thesis) ==> thesis"
  5170     assumes "A | B"
  5171     obtains
  5172       A
  5173     | B
  5174 
  5175 The subsequent classical rules even refer to the formal "thesis"
  5176 explicitly:
  5177 
  5178   lemma classical:     -- "(~ thesis ==> thesis) ==> thesis"
  5179     obtains "~ thesis"
  5180 
  5181   lemma Peirce's_Law:  -- "((thesis ==> something) ==> thesis) ==> thesis"
  5182     obtains "thesis ==> something"
  5183 
  5184 The actual proof of an 'obtains' statement is analogous to that of the
  5185 Isar proof element 'obtain', only that there may be several cases.
  5186 Optional case names may be specified in parentheses; these will be
  5187 available both in the present proof and as annotations in the
  5188 resulting rule, for later use with the 'cases' method (cf. attribute
  5189 case_names).
  5190 
  5191 * Isar: the assumptions of a long theorem statement are available as
  5192 "assms" fact in the proof context.  This is more appropriate than the
  5193 (historical) "prems", which refers to all assumptions of the current
  5194 context, including those from the target locale, proof body etc.
  5195 
  5196 * Isar: 'print_statement' prints theorems from the current theory or
  5197 proof context in long statement form, according to the syntax of a
  5198 top-level lemma.
  5199 
  5200 * Isar: 'obtain' takes an optional case name for the local context
  5201 introduction rule (default "that").
  5202 
  5203 * Isar: removed obsolete 'concl is' patterns.  INCOMPATIBILITY, use
  5204 explicit (is "_ ==> ?foo") in the rare cases where this still happens
  5205 to occur.
  5206 
  5207 * Pure: syntax "CONST name" produces a fully internalized constant
  5208 according to the current context.  This is particularly useful for
  5209 syntax translations that should refer to internal constant
  5210 representations independently of name spaces.
  5211 
  5212 * Pure: syntax constant for foo (binder "FOO ") is called "foo_binder"
  5213 instead of "FOO ". This allows multiple binder declarations to coexist
  5214 in the same context.  INCOMPATIBILITY.
  5215 
  5216 * Isar/locales: 'notation' provides a robust interface to the 'syntax'
  5217 primitive that also works in a locale context (both for constants and
  5218 fixed variables). Type declaration and internal syntactic representation
  5219 of given constants retrieved from the context. Likewise, the
  5220 'no_notation' command allows to remove given syntax annotations from the
  5221 current context.
  5222 
  5223 * Isar/locales: new derived specification elements 'axiomatization',
  5224 'definition', 'abbreviation', which support type-inference, admit
  5225 object-level specifications (equality, equivalence).  See also the
  5226 isar-ref manual.  Examples:
  5227 
  5228   axiomatization
  5229     eq  (infix "===" 50) where
  5230     eq_refl: "x === x" and eq_subst: "x === y ==> P x ==> P y"
  5231 
  5232   definition "f x y = x + y + 1"
  5233   definition g where "g x = f x x"
  5234 
  5235   abbreviation
  5236     neq  (infix "=!=" 50) where
  5237     "x =!= y == ~ (x === y)"
  5238 
  5239 These specifications may be also used in a locale context.  Then the
  5240 constants being introduced depend on certain fixed parameters, and the
  5241 constant name is qualified by the locale base name.  An internal
  5242 abbreviation takes care for convenient input and output, making the
  5243 parameters implicit and using the original short name.  See also
  5244 src/HOL/ex/Abstract_NAT.thy for an example of deriving polymorphic
  5245 entities from a monomorphic theory.
  5246 
  5247 Presently, abbreviations are only available 'in' a target locale, but
  5248 not inherited by general import expressions.  Also note that
  5249 'abbreviation' may be used as a type-safe replacement for 'syntax' +
  5250 'translations' in common applications.  The "no_abbrevs" print mode
  5251 prevents folding of abbreviations in term output.
  5252 
  5253 Concrete syntax is attached to specified constants in internal form,
  5254 independently of name spaces.  The parse tree representation is
  5255 slightly different -- use 'notation' instead of raw 'syntax', and
  5256 'translations' with explicit "CONST" markup to accommodate this.
  5257 
  5258 * Pure/Isar: unified syntax for new-style specification mechanisms
  5259 (e.g.  'definition', 'abbreviation', or 'inductive' in HOL) admits
  5260 full type inference and dummy patterns ("_").  For example:
  5261 
  5262   definition "K x _ = x"
  5263 
  5264   inductive conj for A B
  5265   where "A ==> B ==> conj A B"
  5266 
  5267 * Pure: command 'print_abbrevs' prints all constant abbreviations of
  5268 the current context.  Print mode "no_abbrevs" prevents inversion of
  5269 abbreviations on output.
  5270 
  5271 * Isar/locales: improved parameter handling: use of locales "var" and
  5272 "struct" no longer necessary; - parameter renamings are no longer
  5273 required to be injective.  For example, this allows to define
  5274 endomorphisms as locale endom = homom mult mult h.
  5275 
  5276 * Isar/locales: changed the way locales with predicates are defined.
  5277 Instead of accumulating the specification, the imported expression is
  5278 now an interpretation.  INCOMPATIBILITY: different normal form of
  5279 locale expressions.  In particular, in interpretations of locales with
  5280 predicates, goals repesenting already interpreted fragments are not
  5281 removed automatically.  Use methods `intro_locales' and
  5282 `unfold_locales'; see below.
  5283 
  5284 * Isar/locales: new methods `intro_locales' and `unfold_locales'
  5285 provide backward reasoning on locales predicates.  The methods are
  5286 aware of interpretations and discharge corresponding goals.
  5287 `intro_locales' is less aggressive then `unfold_locales' and does not
  5288 unfold predicates to assumptions.
  5289 
  5290 * Isar/locales: the order in which locale fragments are accumulated
  5291 has changed.  This enables to override declarations from fragments due
  5292 to interpretations -- for example, unwanted simp rules.
  5293 
  5294 * Isar/locales: interpretation in theories and proof contexts has been
  5295 extended.  One may now specify (and prove) equations, which are
  5296 unfolded in interpreted theorems.  This is useful for replacing
  5297 defined concepts (constants depending on locale parameters) by
  5298 concepts already existing in the target context.  Example:
  5299 
  5300   interpretation partial_order ["op <= :: [int, int] => bool"]
  5301     where "partial_order.less (op <=) (x::int) y = (x < y)"
  5302 
  5303 Typically, the constant `partial_order.less' is created by a
  5304 definition specification element in the context of locale
  5305 partial_order.
  5306 
  5307 * Method "induct": improved internal context management to support
  5308 local fixes and defines on-the-fly. Thus explicit meta-level
  5309 connectives !!  and ==> are rarely required anymore in inductive goals
  5310 (using object-logic connectives for this purpose has been long
  5311 obsolete anyway). Common proof patterns are explained in
  5312 src/HOL/Induct/Common_Patterns.thy, see also
  5313 src/HOL/Isar_examples/Puzzle.thy and src/HOL/Lambda for realistic
  5314 examples.
  5315 
  5316 * Method "induct": improved handling of simultaneous goals. Instead of
  5317 introducing object-level conjunction, the statement is now split into
  5318 several conclusions, while the corresponding symbolic cases are nested
  5319 accordingly. INCOMPATIBILITY, proofs need to be structured explicitly,
  5320 see src/HOL/Induct/Common_Patterns.thy, for example.
  5321 
  5322 * Method "induct": mutual induction rules are now specified as a list
  5323 of rule sharing the same induction cases. HOL packages usually provide
  5324 foo_bar.inducts for mutually defined items foo and bar (e.g. inductive
  5325 predicates/sets or datatypes). INCOMPATIBILITY, users need to specify
  5326 mutual induction rules differently, i.e. like this:
  5327 
  5328   (induct rule: foo_bar.inducts)
  5329   (induct set: foo bar)
  5330   (induct pred: foo bar)
  5331   (induct type: foo bar)
  5332 
  5333 The ML function ProjectRule.projections turns old-style rules into the
  5334 new format.
  5335 
  5336 * Method "coinduct": dual of induction, see
  5337 src/HOL/Library/Coinductive_List.thy for various examples.
  5338 
  5339 * Method "cases", "induct", "coinduct": the ``(open)'' option is
  5340 considered a legacy feature.
  5341 
  5342 * Attribute "symmetric" produces result with standardized schematic
  5343 variables (index 0).  Potential INCOMPATIBILITY.
  5344 
  5345 * Simplifier: by default the simplifier trace only shows top level
  5346 rewrites now. That is, trace_simp_depth_limit is set to 1 by
  5347 default. Thus there is less danger of being flooded by the trace. The
  5348 trace indicates where parts have been suppressed.
  5349 
  5350 * Provers/classical: removed obsolete classical version of elim_format
  5351 attribute; classical elim/dest rules are now treated uniformly when
  5352 manipulating the claset.
  5353 
  5354 * Provers/classical: stricter checks to ensure that supplied intro,
  5355 dest and elim rules are well-formed; dest and elim rules must have at
  5356 least one premise.
  5357 
  5358 * Provers/classical: attributes dest/elim/intro take an optional
  5359 weight argument for the rule (just as the Pure versions).  Weights are
  5360 ignored by automated tools, but determine the search order of single
  5361 rule steps.
  5362 
  5363 * Syntax: input syntax now supports dummy variable binding "%_. b",
  5364 where the body does not mention the bound variable.  Note that dummy
  5365 patterns implicitly depend on their context of bounds, which makes
  5366 "{_. _}" match any set comprehension as expected.  Potential
  5367 INCOMPATIBILITY -- parse translations need to cope with syntactic
  5368 constant "_idtdummy" in the binding position.
  5369 
  5370 * Syntax: removed obsolete syntactic constant "_K" and its associated
  5371 parse translation.  INCOMPATIBILITY -- use dummy abstraction instead,
  5372 for example "A -> B" => "Pi A (%_. B)".
  5373 
  5374 * Pure: 'class_deps' command visualizes the subclass relation, using
  5375 the graph browser tool.
  5376 
  5377 * Pure: 'print_theory' now suppresses certain internal declarations by
  5378 default; use '!' option for full details.
  5379 
  5380 
  5381 *** HOL ***
  5382 
  5383 * Method "metis" proves goals by applying the Metis general-purpose
  5384 resolution prover (see also http://gilith.com/software/metis/).
  5385 Examples are in the directory MetisExamples.  WARNING: the
  5386 Isabelle/HOL-Metis integration does not yet work properly with
  5387 multi-threading.
  5388 
  5389 * Command 'sledgehammer' invokes external automatic theorem provers as
  5390 background processes.  It generates calls to the "metis" method if
  5391 successful. These can be pasted into the proof.  Users do not have to
  5392 wait for the automatic provers to return.  WARNING: does not really
  5393 work with multi-threading.
  5394 
  5395 * New "auto_quickcheck" feature tests outermost goal statements for
  5396 potential counter-examples.  Controlled by ML references
  5397 auto_quickcheck (default true) and auto_quickcheck_time_limit (default
  5398 5000 milliseconds).  Fails silently if statements is outside of
  5399 executable fragment, or any other codgenerator problem occurs.
  5400 
  5401 * New constant "undefined" with axiom "undefined x = undefined".
  5402 
  5403 * Added class "HOL.eq", allowing for code generation with polymorphic
  5404 equality.
  5405 
  5406 * Some renaming of class constants due to canonical name prefixing in
  5407 the new 'class' package:
  5408 
  5409     HOL.abs ~> HOL.abs_class.abs
  5410     HOL.divide ~> HOL.divide_class.divide
  5411     0 ~> HOL.zero_class.zero
  5412     1 ~> HOL.one_class.one
  5413     op + ~> HOL.plus_class.plus
  5414     op - ~> HOL.minus_class.minus
  5415     uminus ~> HOL.minus_class.uminus
  5416     op * ~> HOL.times_class.times
  5417     op < ~> HOL.ord_class.less
  5418     op <= > HOL.ord_class.less_eq
  5419     Nat.power ~> Power.power_class.power
  5420     Nat.size ~> Nat.size_class.size
  5421     Numeral.number_of ~> Numeral.number_class.number_of
  5422     FixedPoint.Inf ~> Lattices.complete_lattice_class.Inf
  5423     FixedPoint.Sup ~> Lattices.complete_lattice_class.Sup
  5424     Orderings.min ~> Orderings.ord_class.min
  5425     Orderings.max ~> Orderings.ord_class.max
  5426     Divides.op div ~> Divides.div_class.div
  5427     Divides.op mod ~> Divides.div_class.mod
  5428     Divides.op dvd ~> Divides.div_class.dvd
  5429 
  5430 INCOMPATIBILITY.  Adaptions may be required in the following cases:
  5431 
  5432 a) User-defined constants using any of the names "plus", "minus",
  5433 "times", "less" or "less_eq". The standard syntax translations for
  5434 "+", "-" and "*" may go wrong.  INCOMPATIBILITY: use more specific
  5435 names.
  5436 
  5437 b) Variables named "plus", "minus", "times", "less", "less_eq"
  5438 INCOMPATIBILITY: use more specific names.
  5439 
  5440 c) Permutative equations (e.g. "a + b = b + a")
  5441 Since the change of names also changes the order of terms, permutative
  5442 rewrite rules may get applied in a different order. Experience shows
  5443 that this is rarely the case (only two adaptions in the whole Isabelle
  5444 distribution).  INCOMPATIBILITY: rewrite proofs
  5445 
  5446 d) ML code directly refering to constant names
  5447 This in general only affects hand-written proof tactics, simprocs and
  5448 so on.  INCOMPATIBILITY: grep your sourcecode and replace names.
  5449 Consider using @{const_name} antiquotation.
  5450 
  5451 * New class "default" with associated constant "default".
  5452 
  5453 * Function "sgn" is now overloaded and available on int, real, complex
  5454 (and other numeric types), using class "sgn".  Two possible defs of
  5455 sgn are given as equational assumptions in the classes sgn_if and
  5456 sgn_div_norm; ordered_idom now also inherits from sgn_if.
  5457 INCOMPATIBILITY.
  5458 
  5459 * Locale "partial_order" now unified with class "order" (cf. theory
  5460 Orderings), added parameter "less".  INCOMPATIBILITY.
  5461 
  5462 * Renamings in classes "order" and "linorder": facts "refl", "trans" and
  5463 "cases" to "order_refl", "order_trans" and "linorder_cases", to avoid
  5464 clashes with HOL "refl" and "trans".  INCOMPATIBILITY.
  5465 
  5466 * Classes "order" and "linorder": potential INCOMPATIBILITY due to
  5467 changed order of proof goals in instance proofs.
  5468 
  5469 * The transitivity reasoner for partial and linear orders is set up
  5470 for classes "order" and "linorder".  Instances of the reasoner are available
  5471 in all contexts importing or interpreting the corresponding locales.
  5472 Method "order" invokes the reasoner separately; the reasoner
  5473 is also integrated with the Simplifier as a solver.  Diagnostic
  5474 command 'print_orders' shows the available instances of the reasoner
  5475 in the current context.
  5476 
  5477 * Localized monotonicity predicate in theory "Orderings"; integrated
  5478 lemmas max_of_mono and min_of_mono with this predicate.
  5479 INCOMPATIBILITY.
  5480 
  5481 * Formulation of theorem "dense" changed slightly due to integration
  5482 with new class dense_linear_order.
  5483 
  5484 * Uniform lattice theory development in HOL.
  5485 
  5486     constants "meet" and "join" now named "inf" and "sup"
  5487     constant "Meet" now named "Inf"
  5488 
  5489     classes "meet_semilorder" and "join_semilorder" now named
  5490       "lower_semilattice" and "upper_semilattice"
  5491     class "lorder" now named "lattice"
  5492     class "comp_lat" now named "complete_lattice"
  5493 
  5494     Instantiation of lattice classes allows explicit definitions
  5495     for "inf" and "sup" operations (or "Inf" and "Sup" for complete lattices).
  5496 
  5497   INCOMPATIBILITY.  Theorem renames:
  5498 
  5499     meet_left_le            ~> inf_le1
  5500     meet_right_le           ~> inf_le2
  5501     join_left_le            ~> sup_ge1
  5502     join_right_le           ~> sup_ge2
  5503     meet_join_le            ~> inf_sup_ord
  5504     le_meetI                ~> le_infI
  5505     join_leI                ~> le_supI
  5506     le_meet                 ~> le_inf_iff
  5507     le_join                 ~> ge_sup_conv
  5508     meet_idempotent         ~> inf_idem
  5509     join_idempotent         ~> sup_idem
  5510     meet_comm               ~> inf_commute
  5511     join_comm               ~> sup_commute
  5512     meet_leI1               ~> le_infI1
  5513     meet_leI2               ~> le_infI2
  5514     le_joinI1               ~> le_supI1
  5515     le_joinI2               ~> le_supI2
  5516     meet_assoc              ~> inf_assoc
  5517     join_assoc              ~> sup_assoc
  5518     meet_left_comm          ~> inf_left_commute
  5519     meet_left_idempotent    ~> inf_left_idem
  5520     join_left_comm          ~> sup_left_commute
  5521     join_left_idempotent    ~> sup_left_idem
  5522     meet_aci                ~> inf_aci
  5523     join_aci                ~> sup_aci
  5524     le_def_meet             ~> le_iff_inf
  5525     le_def_join             ~> le_iff_sup
  5526     join_absorp2            ~> sup_absorb2
  5527     join_absorp1            ~> sup_absorb1
  5528     meet_absorp1            ~> inf_absorb1
  5529     meet_absorp2            ~> inf_absorb2
  5530     meet_join_absorp        ~> inf_sup_absorb
  5531     join_meet_absorp        ~> sup_inf_absorb
  5532     distrib_join_le         ~> distrib_sup_le
  5533     distrib_meet_le         ~> distrib_inf_le
  5534 
  5535     add_meet_distrib_left   ~> add_inf_distrib_left
  5536     add_join_distrib_left   ~> add_sup_distrib_left
  5537     is_join_neg_meet        ~> is_join_neg_inf
  5538     is_meet_neg_join        ~> is_meet_neg_sup
  5539     add_meet_distrib_right  ~> add_inf_distrib_right
  5540     add_join_distrib_right  ~> add_sup_distrib_right
  5541     add_meet_join_distribs  ~> add_sup_inf_distribs
  5542     join_eq_neg_meet        ~> sup_eq_neg_inf
  5543     meet_eq_neg_join        ~> inf_eq_neg_sup
  5544     add_eq_meet_join        ~> add_eq_inf_sup
  5545     meet_0_imp_0            ~> inf_0_imp_0
  5546     join_0_imp_0            ~> sup_0_imp_0
  5547     meet_0_eq_0             ~> inf_0_eq_0
  5548     join_0_eq_0             ~> sup_0_eq_0
  5549     neg_meet_eq_join        ~> neg_inf_eq_sup
  5550     neg_join_eq_meet        ~> neg_sup_eq_inf
  5551     join_eq_if              ~> sup_eq_if
  5552 
  5553     mono_meet               ~> mono_inf
  5554     mono_join               ~> mono_sup
  5555     meet_bool_eq            ~> inf_bool_eq
  5556     join_bool_eq            ~> sup_bool_eq
  5557     meet_fun_eq             ~> inf_fun_eq
  5558     join_fun_eq             ~> sup_fun_eq
  5559     meet_set_eq             ~> inf_set_eq
  5560     join_set_eq             ~> sup_set_eq
  5561     meet1_iff               ~> inf1_iff
  5562     meet2_iff               ~> inf2_iff
  5563     meet1I                  ~> inf1I
  5564     meet2I                  ~> inf2I
  5565     meet1D1                 ~> inf1D1
  5566     meet2D1                 ~> inf2D1
  5567     meet1D2                 ~> inf1D2
  5568     meet2D2                 ~> inf2D2
  5569     meet1E                  ~> inf1E
  5570     meet2E                  ~> inf2E
  5571     join1_iff               ~> sup1_iff
  5572     join2_iff               ~> sup2_iff
  5573     join1I1                 ~> sup1I1
  5574     join2I1                 ~> sup2I1
  5575     join1I1                 ~> sup1I1
  5576     join2I2                 ~> sup1I2
  5577     join1CI                 ~> sup1CI
  5578     join2CI                 ~> sup2CI
  5579     join1E                  ~> sup1E
  5580     join2E                  ~> sup2E
  5581 
  5582     is_meet_Meet            ~> is_meet_Inf
  5583     Meet_bool_def           ~> Inf_bool_def
  5584     Meet_fun_def            ~> Inf_fun_def
  5585     Meet_greatest           ~> Inf_greatest
  5586     Meet_lower              ~> Inf_lower
  5587     Meet_set_def            ~> Inf_set_def
  5588 
  5589     Sup_def                 ~> Sup_Inf
  5590     Sup_bool_eq             ~> Sup_bool_def
  5591     Sup_fun_eq              ~> Sup_fun_def
  5592     Sup_set_eq              ~> Sup_set_def
  5593 
  5594     listsp_meetI            ~> listsp_infI
  5595     listsp_meet_eq          ~> listsp_inf_eq
  5596 
  5597     meet_min                ~> inf_min
  5598     join_max                ~> sup_max
  5599 
  5600 * Added syntactic class "size"; overloaded constant "size" now has
  5601 type "'a::size ==> bool"
  5602 
  5603 * Internal reorganisation of `size' of datatypes: size theorems
  5604 "foo.size" are no longer subsumed by "foo.simps" (but are still
  5605 simplification rules by default!); theorems "prod.size" now named
  5606 "*.size".
  5607 
  5608 * Class "div" now inherits from class "times" rather than "type".
  5609 INCOMPATIBILITY.
  5610 
  5611 * HOL/Finite_Set: "name-space" locales Lattice, Distrib_lattice,
  5612 Linorder etc.  have disappeared; operations defined in terms of
  5613 fold_set now are named Inf_fin, Sup_fin.  INCOMPATIBILITY.
  5614 
  5615 * HOL/Nat: neq0_conv no longer declared as iff.  INCOMPATIBILITY.
  5616 
  5617 * HOL-Word: New extensive library and type for generic, fixed size
  5618 machine words, with arithmetic, bit-wise, shifting and rotating
  5619 operations, reflection into int, nat, and bool lists, automation for
  5620 linear arithmetic (by automatic reflection into nat or int), including
  5621 lemmas on overflow and monotonicity.  Instantiated to all appropriate
  5622 arithmetic type classes, supporting automatic simplification of
  5623 numerals on all operations.
  5624 
  5625 * Library/Boolean_Algebra: locales for abstract boolean algebras.
  5626 
  5627 * Library/Numeral_Type: numbers as types, e.g. TYPE(32).
  5628 
  5629 * Code generator library theories:
  5630   - Code_Integer represents HOL integers by big integer literals in target
  5631     languages.
  5632   - Code_Char represents HOL characters by character literals in target
  5633     languages.
  5634   - Code_Char_chr like Code_Char, but also offers treatment of character
  5635     codes; includes Code_Integer.
  5636   - Executable_Set allows to generate code for finite sets using lists.
  5637   - Executable_Rat implements rational numbers as triples (sign, enumerator,
  5638     denominator).
  5639   - Executable_Real implements a subset of real numbers, namly those
  5640     representable by rational numbers.
  5641   - Efficient_Nat implements natural numbers by integers, which in general will
  5642     result in higher efficency; pattern matching with 0/Suc is eliminated;
  5643     includes Code_Integer.
  5644   - Code_Index provides an additional datatype index which is mapped to
  5645     target-language built-in integers.
  5646   - Code_Message provides an additional datatype message_string which is isomorphic to
  5647     strings; messages are mapped to target-language strings.
  5648 
  5649 * New package for inductive predicates
  5650 
  5651   An n-ary predicate p with m parameters z_1, ..., z_m can now be defined via
  5652 
  5653     inductive
  5654       p :: "U_1 => ... => U_m => T_1 => ... => T_n => bool"
  5655       for z_1 :: U_1 and ... and z_n :: U_m
  5656     where
  5657       rule_1: "... ==> p z_1 ... z_m t_1_1 ... t_1_n"
  5658     | ...
  5659 
  5660   with full support for type-inference, rather than
  5661 
  5662     consts s :: "U_1 => ... => U_m => (T_1 * ... * T_n) set"
  5663 
  5664     abbreviation p :: "U_1 => ... => U_m => T_1 => ... => T_n => bool"
  5665     where "p z_1 ... z_m x_1 ... x_n == (x_1, ..., x_n) : s z_1 ... z_m"
  5666 
  5667     inductive "s z_1 ... z_m"
  5668     intros
  5669       rule_1: "... ==> (t_1_1, ..., t_1_n) : s z_1 ... z_m"
  5670       ...
  5671 
  5672   For backward compatibility, there is a wrapper allowing inductive
  5673   sets to be defined with the new package via
  5674 
  5675     inductive_set
  5676       s :: "U_1 => ... => U_m => (T_1 * ... * T_n) set"
  5677       for z_1 :: U_1 and ... and z_n :: U_m
  5678     where
  5679       rule_1: "... ==> (t_1_1, ..., t_1_n) : s z_1 ... z_m"
  5680     | ...
  5681 
  5682   or
  5683 
  5684     inductive_set
  5685       s :: "U_1 => ... => U_m => (T_1 * ... * T_n) set"
  5686       and p :: "U_1 => ... => U_m => T_1 => ... => T_n => bool"
  5687       for z_1 :: U_1 and ... and z_n :: U_m
  5688     where
  5689       "p z_1 ... z_m x_1 ... x_n == (x_1, ..., x_n) : s z_1 ... z_m"
  5690     | rule_1: "... ==> p z_1 ... z_m t_1_1 ... t_1_n"
  5691     | ...
  5692 
  5693   if the additional syntax "p ..." is required.
  5694 
  5695   Numerous examples can be found in the subdirectories src/HOL/Auth,
  5696   src/HOL/Bali, src/HOL/Induct, and src/HOL/MicroJava.
  5697 
  5698   INCOMPATIBILITIES:
  5699 
  5700   - Since declaration and definition of inductive sets or predicates
  5701     is no longer separated, abbreviations involving the newly
  5702     introduced sets or predicates must be specified together with the
  5703     introduction rules after the 'where' keyword (see above), rather
  5704     than before the actual inductive definition.
  5705 
  5706   - The variables in induction and elimination rules are now
  5707     quantified in the order of their occurrence in the introduction
  5708     rules, rather than in alphabetical order. Since this may break
  5709     some proofs, these proofs either have to be repaired, e.g. by
  5710     reordering the variables a_i_1 ... a_i_{k_i} in Isar 'case'
  5711     statements of the form
  5712 
  5713       case (rule_i a_i_1 ... a_i_{k_i})
  5714 
  5715     or the old order of quantification has to be restored by explicitly adding
  5716     meta-level quantifiers in the introduction rules, i.e.
  5717 
  5718       | rule_i: "!!a_i_1 ... a_i_{k_i}. ... ==> p z_1 ... z_m t_i_1 ... t_i_n"
  5719 
  5720   - The format of the elimination rules is now
  5721 
  5722       p z_1 ... z_m x_1 ... x_n ==>
  5723         (!!a_1_1 ... a_1_{k_1}. x_1 = t_1_1 ==> ... ==> x_n = t_1_n ==> ... ==> P)
  5724         ==> ... ==> P
  5725 
  5726     for predicates and
  5727 
  5728       (x_1, ..., x_n) : s z_1 ... z_m ==>
  5729         (!!a_1_1 ... a_1_{k_1}. x_1 = t_1_1 ==> ... ==> x_n = t_1_n ==> ... ==> P)
  5730         ==> ... ==> P
  5731 
  5732     for sets rather than
  5733 
  5734       x : s z_1 ... z_m ==>
  5735         (!!a_1_1 ... a_1_{k_1}. x = (t_1_1, ..., t_1_n) ==> ... ==> P)
  5736         ==> ... ==> P
  5737 
  5738     This may require terms in goals to be expanded to n-tuples
  5739     (e.g. using case_tac or simplification with the split_paired_all
  5740     rule) before the above elimination rule is applicable.
  5741 
  5742   - The elimination or case analysis rules for (mutually) inductive
  5743     sets or predicates are now called "p_1.cases" ... "p_k.cases". The
  5744     list of rules "p_1_..._p_k.elims" is no longer available.
  5745 
  5746 * New package "function"/"fun" for general recursive functions,
  5747 supporting mutual and nested recursion, definitions in local contexts,
  5748 more general pattern matching and partiality. See HOL/ex/Fundefs.thy
  5749 for small examples, and the separate tutorial on the function
  5750 package. The old recdef "package" is still available as before, but
  5751 users are encouraged to use the new package.
  5752 
  5753 * Method "lexicographic_order" automatically synthesizes termination
  5754 relations as lexicographic combinations of size measures.
  5755 
  5756 * Case-expressions allow arbitrary constructor-patterns (including
  5757 "_") and take their order into account, like in functional
  5758 programming.  Internally, this is translated into nested
  5759 case-expressions; missing cases are added and mapped to the predefined
  5760 constant "undefined". In complicated cases printing may no longer show
  5761 the original input but the internal form. Lambda-abstractions allow
  5762 the same form of pattern matching: "% pat1 => e1 | ..." is an
  5763 abbreviation for "%x. case x of pat1 => e1 | ..." where x is a new
  5764 variable.
  5765 
  5766 * IntDef: The constant "int :: nat => int" has been removed; now "int"
  5767 is an abbreviation for "of_nat :: nat => int". The simplification
  5768 rules for "of_nat" have been changed to work like "int" did
  5769 previously.  Potential INCOMPATIBILITY:
  5770   - "of_nat (Suc m)" simplifies to "1 + of_nat m" instead of "of_nat m + 1"
  5771   - of_nat_diff and of_nat_mult are no longer default simp rules
  5772 
  5773 * Method "algebra" solves polynomial equations over (semi)rings using
  5774 Groebner bases. The (semi)ring structure is defined by locales and the
  5775 tool setup depends on that generic context. Installing the method for
  5776 a specific type involves instantiating the locale and possibly adding
  5777 declarations for computation on the coefficients.  The method is
  5778 already instantiated for natural numbers and for the axiomatic class
  5779 of idoms with numerals.  See also the paper by Chaieb and Wenzel at
  5780 CALCULEMUS 2007 for the general principles underlying this
  5781 architecture of context-aware proof-tools.
  5782 
  5783 * Method "ferrack" implements quantifier elimination over
  5784 special-purpose dense linear orders using locales (analogous to
  5785 "algebra"). The method is already installed for class
  5786 {ordered_field,recpower,number_ring} which subsumes real, hyperreal,
  5787 rat, etc.
  5788 
  5789 * Former constant "List.op @" now named "List.append".  Use ML
  5790 antiquotations @{const_name List.append} or @{term " ... @ ... "} to
  5791 circumvent possible incompatibilities when working on ML level.
  5792 
  5793 * primrec: missing cases mapped to "undefined" instead of "arbitrary".
  5794 
  5795 * New function listsum :: 'a list => 'a for arbitrary monoids.
  5796 Special syntax: "SUM x <- xs. f x" (and latex variants)
  5797 
  5798 * New syntax for Haskell-like list comprehension (input only), eg.
  5799 [(x,y). x <- xs, y <- ys, x ~= y], see also src/HOL/List.thy.
  5800 
  5801 * The special syntax for function "filter" has changed from [x :
  5802 xs. P] to [x <- xs. P] to avoid an ambiguity caused by list
  5803 comprehension syntax, and for uniformity.  INCOMPATIBILITY.
  5804 
  5805 * [a..b] is now defined for arbitrary linear orders.  It used to be
  5806 defined on nat only, as an abbreviation for [a..<Suc b]
  5807 INCOMPATIBILITY.
  5808 
  5809 * Renamed lemma "set_take_whileD"  to "set_takeWhileD".
  5810 
  5811 * New functions "sorted" and "sort" in src/HOL/List.thy.
  5812 
  5813 * New lemma collection field_simps (an extension of ring_simps) for
  5814 manipulating (in)equations involving division. Multiplies with all
  5815 denominators that can be proved to be non-zero (in equations) or
  5816 positive/negative (in inequations).
  5817 
  5818 * Lemma collections ring_eq_simps, group_eq_simps and ring_distrib
  5819 have been improved and renamed to ring_simps, group_simps and
  5820 ring_distribs.  Removed lemmas field_xyz in theory Ring_and_Field
  5821 because they were subsumed by lemmas xyz.  INCOMPATIBILITY.
  5822 
  5823 * Theory Library/Commutative_Ring: switched from recdef to function
  5824 package; constants add, mul, pow now curried.  Infix syntax for
  5825 algebraic operations.
  5826 
  5827 * Dropped redundant lemma def_imp_eq in favor of meta_eq_to_obj_eq.
  5828 INCOMPATIBILITY.
  5829 
  5830 * Dropped redundant lemma if_def2 in favor of if_bool_eq_conj.
  5831 INCOMPATIBILITY.
  5832 
  5833 * HOL/records: generalised field-update to take a function on the
  5834 field rather than the new value: r(|A := x|) is translated to A_update
  5835 (K x) r The K-combinator that is internally used is called K_record.
  5836 INCOMPATIBILITY: Usage of the plain update functions has to be
  5837 adapted.
  5838 
  5839 * Class "semiring_0" now contains annihilation axioms x * 0 = 0 and 0
  5840 * x = 0, which are required for a semiring.  Richer structures do not
  5841 inherit from semiring_0 anymore, because this property is a theorem
  5842 there, not an axiom.  INCOMPATIBILITY: In instances of semiring_0,
  5843 there is more to prove, but this is mostly trivial.
  5844 
  5845 * Class "recpower" is generalized to arbitrary monoids, not just
  5846 commutative semirings.  INCOMPATIBILITY: may need to incorporate
  5847 commutativity or semiring properties additionally.
  5848 
  5849 * Constant "List.list_all2" in List.thy now uses authentic syntax.
  5850 INCOMPATIBILITY: translations containing list_all2 may go wrong,
  5851 better use 'abbreviation'.
  5852 
  5853 * Renamed constant "List.op mem" to "List.member".  INCOMPATIBILITY.
  5854 
  5855 * Numeral syntax: type 'bin' which was a mere type copy of 'int' has
  5856 been abandoned in favour of plain 'int'.  INCOMPATIBILITY --
  5857 significant changes for setting up numeral syntax for types:
  5858   - New constants Numeral.pred and Numeral.succ instead
  5859       of former Numeral.bin_pred and Numeral.bin_succ.
  5860   - Use integer operations instead of bin_add, bin_mult and so on.
  5861   - Numeral simplification theorems named Numeral.numeral_simps instead of Bin_simps.
  5862   - ML structure Bin_Simprocs now named Int_Numeral_Base_Simprocs.
  5863 
  5864 See src/HOL/Integ/IntArith.thy for an example setup.
  5865 
  5866 * Command 'normal_form' computes the normal form of a term that may
  5867 contain free variables.  For example ``normal_form "rev [a, b, c]"''
  5868 produces ``[b, c, a]'' (without proof).  This command is suitable for
  5869 heavy-duty computations because the functions are compiled to ML
  5870 first.  Correspondingly, a method "normalization" is provided.  See
  5871 further src/HOL/ex/NormalForm.thy and src/Tools/nbe.ML.
  5872 
  5873 * Alternative iff syntax "A <-> B" for equality on bool (with priority
  5874 25 like -->); output depends on the "iff" print_mode, the default is
  5875 "A = B" (with priority 50).
  5876 
  5877 * Relations less (<) and less_eq (<=) are also available on type bool.
  5878 Modified syntax to disallow nesting without explicit parentheses,
  5879 e.g. "(x < y) < z" or "x < (y < z)", but NOT "x < y < z".  Potential
  5880 INCOMPATIBILITY.
  5881 
  5882 * "LEAST x:A. P" expands to "LEAST x. x:A & P" (input only).
  5883 
  5884 * Relation composition operator "op O" now has precedence 75 and binds
  5885 stronger than union and intersection. INCOMPATIBILITY.
  5886 
  5887 * The old set interval syntax "{m..n(}" (and relatives) has been
  5888 removed.  Use "{m..<n}" (and relatives) instead.
  5889 
  5890 * In the context of the assumption "~(s = t)" the Simplifier rewrites
  5891 "t = s" to False (by simproc "neq").  INCOMPATIBILITY, consider using
  5892 ``declare [[simproc del: neq]]''.
  5893 
  5894 * Simplifier: "m dvd n" where m and n are numbers is evaluated to
  5895 True/False.
  5896 
  5897 * Theorem Cons_eq_map_conv no longer declared as "simp".
  5898 
  5899 * Theorem setsum_mult renamed to setsum_right_distrib.
  5900 
  5901 * Prefer ex1I over ex_ex1I in single-step reasoning, e.g. by the
  5902 ``rule'' method.
  5903 
  5904 * Reimplemented methods "sat" and "satx", with several improvements:
  5905 goals no longer need to be stated as "<prems> ==> False", equivalences
  5906 (i.e. "=" on type bool) are handled, variable names of the form
  5907 "lit_<n>" are no longer reserved, significant speedup.
  5908 
  5909 * Methods "sat" and "satx" can now replay MiniSat proof traces.
  5910 zChaff is still supported as well.
  5911 
  5912 * 'inductive' and 'datatype': provide projections of mutual rules,
  5913 bundled as foo_bar.inducts;
  5914 
  5915 * Library: moved theories Parity, GCD, Binomial, Infinite_Set to
  5916 Library.
  5917 
  5918 * Library: moved theory Accessible_Part to main HOL.
  5919 
  5920 * Library: added theory Coinductive_List of potentially infinite lists
  5921 as greatest fixed-point.
  5922 
  5923 * Library: added theory AssocList which implements (finite) maps as
  5924 association lists.
  5925 
  5926 * Method "evaluation" solves goals (i.e. a boolean expression)
  5927 efficiently by compiling it to ML.  The goal is "proved" (via an
  5928 oracle) if it evaluates to True.
  5929 
  5930 * Linear arithmetic now splits certain operators (e.g. min, max, abs)
  5931 also when invoked by the simplifier.  This results in the Simplifier
  5932 being more powerful on arithmetic goals.  INCOMPATIBILITY.
  5933 Configuration option fast_arith_split_limit=0 recovers the old
  5934 behavior.
  5935 
  5936 * Support for hex (0x20) and binary (0b1001) numerals.
  5937 
  5938 * New method: reify eqs (t), where eqs are equations for an
  5939 interpretation I :: 'a list => 'b => 'c and t::'c is an optional
  5940 parameter, computes a term s::'b and a list xs::'a list and proves the
  5941 theorem I xs s = t. This is also known as reification or quoting. The
  5942 resulting theorem is applied to the subgoal to substitute t with I xs
  5943 s.  If t is omitted, the subgoal itself is reified.
  5944 
  5945 * New method: reflection corr_thm eqs (t). The parameters eqs and (t)
  5946 are as explained above. corr_thm is a theorem for I vs (f t) = I vs t,
  5947 where f is supposed to be a computable function (in the sense of code
  5948 generattion). The method uses reify to compute s and xs as above then
  5949 applies corr_thm and uses normalization by evaluation to "prove" f s =
  5950 r and finally gets the theorem t = r, which is again applied to the
  5951 subgoal. An Example is available in src/HOL/ex/ReflectionEx.thy.
  5952 
  5953 * Reflection: Automatic reification now handels binding, an example is
  5954 available in src/HOL/ex/ReflectionEx.thy
  5955 
  5956 * HOL-Statespace: ``State Spaces: The Locale Way'' introduces a
  5957 command 'statespace' that is similar to 'record', but introduces an
  5958 abstract specification based on the locale infrastructure instead of
  5959 HOL types.  This leads to extra flexibility in composing state spaces,
  5960 in particular multiple inheritance and renaming of components.
  5961 
  5962 
  5963 *** HOL-Complex ***
  5964 
  5965 * Hyperreal: Functions root and sqrt are now defined on negative real
  5966 inputs so that root n (- x) = - root n x and sqrt (- x) = - sqrt x.
  5967 Nonnegativity side conditions have been removed from many lemmas, so
  5968 that more subgoals may now be solved by simplification; potential
  5969 INCOMPATIBILITY.
  5970 
  5971 * Real: new type classes formalize real normed vector spaces and
  5972 algebras, using new overloaded constants scaleR :: real => 'a => 'a
  5973 and norm :: 'a => real.
  5974 
  5975 * Real: constant of_real :: real => 'a::real_algebra_1 injects from
  5976 reals into other types. The overloaded constant Reals :: 'a set is now
  5977 defined as range of_real; potential INCOMPATIBILITY.
  5978 
  5979 * Real: proper support for ML code generation, including 'quickcheck'.
  5980 Reals are implemented as arbitrary precision rationals.
  5981 
  5982 * Hyperreal: Several constants that previously worked only for the
  5983 reals have been generalized, so they now work over arbitrary vector
  5984 spaces. Type annotations may need to be added in some cases; potential
  5985 INCOMPATIBILITY.
  5986 
  5987   Infinitesimal  :: ('a::real_normed_vector) star set
  5988   HFinite        :: ('a::real_normed_vector) star set
  5989   HInfinite      :: ('a::real_normed_vector) star set
  5990   approx         :: ('a::real_normed_vector) star => 'a star => bool
  5991   monad          :: ('a::real_normed_vector) star => 'a star set
  5992   galaxy         :: ('a::real_normed_vector) star => 'a star set
  5993   (NS)LIMSEQ     :: [nat => 'a::real_normed_vector, 'a] => bool
  5994   (NS)convergent :: (nat => 'a::real_normed_vector) => bool
  5995   (NS)Bseq       :: (nat => 'a::real_normed_vector) => bool
  5996   (NS)Cauchy     :: (nat => 'a::real_normed_vector) => bool
  5997   (NS)LIM        :: ['a::real_normed_vector => 'b::real_normed_vector, 'a, 'b] => bool
  5998   is(NS)Cont     :: ['a::real_normed_vector => 'b::real_normed_vector, 'a] => bool
  5999   deriv          :: ['a::real_normed_field => 'a, 'a, 'a] => bool
  6000   sgn            :: 'a::real_normed_vector => 'a
  6001   exp            :: 'a::{recpower,real_normed_field,banach} => 'a
  6002 
  6003 * Complex: Some complex-specific constants are now abbreviations for
  6004 overloaded ones: complex_of_real = of_real, cmod = norm, hcmod =
  6005 hnorm.  Other constants have been entirely removed in favor of the
  6006 polymorphic versions (INCOMPATIBILITY):
  6007 
  6008   approx        <-- capprox
  6009   HFinite       <-- CFinite
  6010   HInfinite     <-- CInfinite
  6011   Infinitesimal <-- CInfinitesimal
  6012   monad         <-- cmonad
  6013   galaxy        <-- cgalaxy
  6014   (NS)LIM       <-- (NS)CLIM, (NS)CRLIM
  6015   is(NS)Cont    <-- is(NS)Contc, is(NS)contCR
  6016   (ns)deriv     <-- (ns)cderiv
  6017 
  6018 
  6019 *** HOL-Algebra ***
  6020 
  6021 * Formalisation of ideals and the quotient construction over rings.
  6022 
  6023 * Order and lattice theory no longer based on records.
  6024 INCOMPATIBILITY.
  6025 
  6026 * Renamed lemmas least_carrier -> least_closed and greatest_carrier ->
  6027 greatest_closed.  INCOMPATIBILITY.
  6028 
  6029 * Method algebra is now set up via an attribute.  For examples see
  6030 Ring.thy.  INCOMPATIBILITY: the method is now weaker on combinations
  6031 of algebraic structures.
  6032 
  6033 * Renamed theory CRing to Ring.
  6034 
  6035 
  6036 *** HOL-Nominal ***
  6037 
  6038 * Substantial, yet incomplete support for nominal datatypes (binding
  6039 structures) based on HOL-Nominal logic.  See src/HOL/Nominal and
  6040 src/HOL/Nominal/Examples.  Prospective users should consult
  6041 http://isabelle.in.tum.de/nominal/
  6042 
  6043 
  6044 *** ML ***
  6045 
  6046 * ML basics: just one true type int, which coincides with IntInf.int
  6047 (even on SML/NJ).
  6048 
  6049 * ML within Isar: antiquotations allow to embed statically-checked
  6050 formal entities in the source, referring to the context available at
  6051 compile-time.  For example:
  6052 
  6053 ML {* @{sort "{zero,one}"} *}
  6054 ML {* @{typ "'a => 'b"} *}
  6055 ML {* @{term "%x. x"} *}
  6056 ML {* @{prop "x == y"} *}
  6057 ML {* @{ctyp "'a => 'b"} *}
  6058 ML {* @{cterm "%x. x"} *}
  6059 ML {* @{cprop "x == y"} *}
  6060 ML {* @{thm asm_rl} *}
  6061 ML {* @{thms asm_rl} *}
  6062 ML {* @{type_name c} *}
  6063 ML {* @{type_syntax c} *}
  6064 ML {* @{const_name c} *}
  6065 ML {* @{const_syntax c} *}
  6066 ML {* @{context} *}
  6067 ML {* @{theory} *}
  6068 ML {* @{theory Pure} *}
  6069 ML {* @{theory_ref} *}
  6070 ML {* @{theory_ref Pure} *}
  6071 ML {* @{simpset} *}
  6072 ML {* @{claset} *}
  6073 ML {* @{clasimpset} *}
  6074 
  6075 The same works for sources being ``used'' within an Isar context.
  6076 
  6077 * ML in Isar: improved error reporting; extra verbosity with
  6078 ML_Context.trace enabled.
  6079 
  6080 * Pure/General/table.ML: the join operations now works via exceptions
  6081 DUP/SAME instead of type option. This is simpler in simple cases, and
  6082 admits slightly more efficient complex applications.
  6083 
  6084 * Pure: 'advanced' translation functions (parse_translation etc.) now
  6085 use Context.generic instead of just theory.
  6086 
  6087 * Pure: datatype Context.generic joins theory/Proof.context and
  6088 provides some facilities for code that works in either kind of
  6089 context, notably GenericDataFun for uniform theory and proof data.
  6090 
  6091 * Pure: simplified internal attribute type, which is now always
  6092 Context.generic * thm -> Context.generic * thm. Global (theory) vs.
  6093 local (Proof.context) attributes have been discontinued, while
  6094 minimizing code duplication. Thm.rule_attribute and
  6095 Thm.declaration_attribute build canonical attributes; see also structure
  6096 Context for further operations on Context.generic, notably
  6097 GenericDataFun. INCOMPATIBILITY, need to adapt attribute type
  6098 declarations and definitions.
  6099 
  6100 * Context data interfaces (Theory/Proof/GenericDataFun): removed
  6101 name/print, uninitialized data defaults to ad-hoc copy of empty value,
  6102 init only required for impure data. INCOMPATIBILITY: empty really need
  6103 to be empty (no dependencies on theory content!)
  6104 
  6105 * Pure/kernel: consts certification ignores sort constraints given in
  6106 signature declarations. (This information is not relevant to the
  6107 logic, but only for type inference.) SIGNIFICANT INTERNAL CHANGE,
  6108 potential INCOMPATIBILITY.
  6109 
  6110 * Pure: axiomatic type classes are now purely definitional, with
  6111 explicit proofs of class axioms and super class relations performed
  6112 internally. See Pure/axclass.ML for the main internal interfaces --
  6113 notably AxClass.define_class supercedes AxClass.add_axclass, and
  6114 AxClass.axiomatize_class/classrel/arity supersede
  6115 Sign.add_classes/classrel/arities.
  6116 
  6117 * Pure/Isar: Args/Attrib parsers operate on Context.generic --
  6118 global/local versions on theory vs. Proof.context have been
  6119 discontinued; Attrib.syntax and Method.syntax have been adapted
  6120 accordingly.  INCOMPATIBILITY, need to adapt parser expressions for
  6121 attributes, methods, etc.
  6122 
  6123 * Pure: several functions of signature "... -> theory -> theory * ..."
  6124 have been reoriented to "... -> theory -> ... * theory" in order to
  6125 allow natural usage in combination with the ||>, ||>>, |-> and
  6126 fold_map combinators.
  6127 
  6128 * Pure: official theorem names (closed derivations) and additional
  6129 comments (tags) are now strictly separate.  Name hints -- which are
  6130 maintained as tags -- may be attached any time without affecting the
  6131 derivation.
  6132 
  6133 * Pure: primitive rule lift_rule now takes goal cterm instead of an
  6134 actual goal state (thm).  Use Thm.lift_rule (Thm.cprem_of st i) to
  6135 achieve the old behaviour.
  6136 
  6137 * Pure: the "Goal" constant is now called "prop", supporting a
  6138 slightly more general idea of ``protecting'' meta-level rule
  6139 statements.
  6140 
  6141 * Pure: Logic.(un)varify only works in a global context, which is now
  6142 enforced instead of silently assumed.  INCOMPATIBILITY, may use
  6143 Logic.legacy_(un)varify as temporary workaround.
  6144 
  6145 * Pure: structure Name provides scalable operations for generating
  6146 internal variable names, notably Name.variants etc.  This replaces
  6147 some popular functions from term.ML:
  6148 
  6149   Term.variant		->  Name.variant
  6150   Term.variantlist	->  Name.variant_list
  6151   Term.invent_names	->  Name.invent_list
  6152 
  6153 Note that low-level renaming rarely occurs in new code -- operations
  6154 from structure Variable are used instead (see below).
  6155 
  6156 * Pure: structure Variable provides fundamental operations for proper
  6157 treatment of fixed/schematic variables in a context.  For example,
  6158 Variable.import introduces fixes for schematics of given facts and
  6159 Variable.export reverses the effect (up to renaming) -- this replaces
  6160 various freeze_thaw operations.
  6161 
  6162 * Pure: structure Goal provides simple interfaces for
  6163 init/conclude/finish and tactical prove operations (replacing former
  6164 Tactic.prove).  Goal.prove is the canonical way to prove results
  6165 within a given context; Goal.prove_global is a degraded version for
  6166 theory level goals, including a global Drule.standard.  Note that
  6167 OldGoals.prove_goalw_cterm has long been obsolete, since it is
  6168 ill-behaved in a local proof context (e.g. with local fixes/assumes or
  6169 in a locale context).
  6170 
  6171 * Pure/Syntax: generic interfaces for parsing (Syntax.parse_term etc.)
  6172 and type checking (Syntax.check_term etc.), with common combinations
  6173 (Syntax.read_term etc.). These supersede former Sign.read_term etc.
  6174 which are considered legacy and await removal.
  6175 
  6176 * Pure/Syntax: generic interfaces for type unchecking
  6177 (Syntax.uncheck_terms etc.) and unparsing (Syntax.unparse_term etc.),
  6178 with common combinations (Syntax.pretty_term, Syntax.string_of_term
  6179 etc.).  Former Sign.pretty_term, Sign.string_of_term etc. are still
  6180 available for convenience, but refer to the very same operations using
  6181 a mere theory instead of a full context.
  6182 
  6183 * Isar: simplified treatment of user-level errors, using exception
  6184 ERROR of string uniformly.  Function error now merely raises ERROR,
  6185 without any side effect on output channels.  The Isar toplevel takes
  6186 care of proper display of ERROR exceptions.  ML code may use plain
  6187 handle/can/try; cat_error may be used to concatenate errors like this:
  6188 
  6189   ... handle ERROR msg => cat_error msg "..."
  6190 
  6191 Toplevel ML code (run directly or through the Isar toplevel) may be
  6192 embedded into the Isar toplevel with exception display/debug like
  6193 this:
  6194 
  6195   Isar.toplevel (fn () => ...)
  6196 
  6197 INCOMPATIBILITY, removed special transform_error facilities, removed
  6198 obsolete variants of user-level exceptions (ERROR_MESSAGE,
  6199 Context.PROOF, ProofContext.CONTEXT, Proof.STATE, ProofHistory.FAIL)
  6200 -- use plain ERROR instead.
  6201 
  6202 * Isar: theory setup now has type (theory -> theory), instead of a
  6203 list.  INCOMPATIBILITY, may use #> to compose setup functions.
  6204 
  6205 * Isar: ML toplevel pretty printer for type Proof.context, subject to
  6206 ProofContext.debug/verbose flags.
  6207 
  6208 * Isar: Toplevel.theory_to_proof admits transactions that modify the
  6209 theory before entering a proof state.  Transactions now always see a
  6210 quasi-functional intermediate checkpoint, both in interactive and
  6211 batch mode.
  6212 
  6213 * Isar: simplified interfaces for outer syntax.  Renamed
  6214 OuterSyntax.add_keywords to OuterSyntax.keywords.  Removed
  6215 OuterSyntax.add_parsers -- this functionality is now included in
  6216 OuterSyntax.command etc.  INCOMPATIBILITY.
  6217 
  6218 * Simplifier: the simpset of a running simplification process now
  6219 contains a proof context (cf. Simplifier.the_context), which is the
  6220 very context that the initial simpset has been retrieved from (by
  6221 simpset_of/local_simpset_of).  Consequently, all plug-in components
  6222 (solver, looper etc.) may depend on arbitrary proof data.
  6223 
  6224 * Simplifier.inherit_context inherits the proof context (plus the
  6225 local bounds) of the current simplification process; any simproc
  6226 etc. that calls the Simplifier recursively should do this!  Removed
  6227 former Simplifier.inherit_bounds, which is already included here --
  6228 INCOMPATIBILITY.  Tools based on low-level rewriting may even have to
  6229 specify an explicit context using Simplifier.context/theory_context.
  6230 
  6231 * Simplifier/Classical Reasoner: more abstract interfaces
  6232 change_simpset/claset for modifying the simpset/claset reference of a
  6233 theory; raw versions simpset/claset_ref etc. have been discontinued --
  6234 INCOMPATIBILITY.
  6235 
  6236 * Provers: more generic wrt. syntax of object-logics, avoid hardwired
  6237 "Trueprop" etc.
  6238 
  6239